Effects of Frontal Lobe Lesions on Intentionalityand Random Physical Phenomena

MORRIS FREEDMAN

Behavioural Neurology Program and Rotman Research Institute Baycrest CentreFor Geriatric Care Department of Medicine (Division of Neurology)

Mt Sinai Hospital University Health Network and University of TorontoToronto Canada

e-mail morrisfreedmanutorontoca

STANLEY JEFFERS

York University Toronto Canada

KAREN SAEGER

Behavioural Neurology Program Baycrest Centre For Geriatric Care Toronto Canada

MALCOLM BINNS

Rotman Research Institute Baycrest Centre For Geriatric Care Toronto Canada

SANDRA BLACK

Rotman Research Institute Baycrest Centre For Geriatric CareDepartment of Medicine (Division of Neurology)

Sunnybrook and Womenrsquos College Health Sciences Centre and University of TorontoToronto Canada

AbstractmdashAlthough data from the PEAR program at Princeton Universityappear to support a role for intentionality in determining physical phenomenathe use of theoretically based controls raises concerns about validity of thefindings We re-examined claims from the PEAR lab using experimentallyderived control data in a study of patients with frontal lobe brain damage andnormal subjects The rationale for includingfrontal patients follows a suggestionthat reduced self-awareness may facilitate effects of intentionality on physicalphenomena Frontal patients may have reduced self-awareness a state noteasily achieved by normal subjects and may provide a good model for studyingthe role of consciousness on physical events within a conceptual frameworkthat maximizes the likelihood of detecting possible effects We found a signif-icant effect of intentionality on random physical phenomena in a patient withleft frontal damage that was directed contralateral to his lesion Moreover theeffect was replicated

Keywords consciousnessmdashself-awarenessmdashintentionalitymdashfrontal lobe dam-agemdashrandom event generator

Although several studies claim to support a role for intentionality in determiningphysical phenomena (Schmidt 1969 Schmidt amp Pantas 1972 Jahn amp Dunne

Journal of Scienti c Exploration Vol 17 No 4 pp 651ndash668 2003 0892-331003

651

1987a Radin amp Nelson 1989 Jahn et al 1997) concerns about research designand a lack of theoretical models (Alcock 1987 Jeffers in press) as well asnegative studies (Hall et al 1977 Jeffers amp Sloan 1992 Ibison amp Jeffers1998) have been important sources of criticism of the literature in this areaA major methodological problem in research design relates to the issue ofinadequate experimental controls (Jeffers in press) For example Robert Jahnand his colleagues from the Princeton Engineering Anomalies Research (PEAR)programhave reported statisticallysignificanteffects whereby subjectsor lsquolsquoopera-torsrsquorsquo have successfully influenced the statistical distribution of outcomes froma Random Event Generator (REG) (Jahn et al 1987a Jahn et al 1997)A concern relates to the nature of the machine calibration data used forcomparison to subject generated data (Jeffers in press) Rather than collectingcontrol data in close temporal proximity to the period when a subject is trying toinfluence the REG output the PEAR protocol relies on the assumption that theREG output is always random in the absence of operator influence This approachdoes not control for potentially unrecognised factors that may affect the REGoutput and therefore casts doubt on the interpretation of the experimentalfindings In addition there are theoretical issues raising questions about thefindings from Jahnrsquos group such as the independenceof the reported effects fromtime and distance that are difficult to reconcile (Jahn et al 1997 Jeffers inpress) Nevertheless Jahn and his colleagues have amassed a great wealth of datato support their conclusions that an individualrsquos conscious intention can alter thestatistical distribution of random physical phenomena Because their findingswould have immense significance if validated we re-examined their claimsusing a methodology with well-designed control conditions

Some highly interesting but speculative ideas relating anomalous physicalactivity to consciousness have been advanced by Jahn and Dunne (Jahn et al1987a Jahn amp Dunne 1986) They proposed a metaphor for consciousnessbased upon quantum mechanical concepts that relates consciousness toanomalous physical phenomena Based on data from the PEAR lab theysuggest that consciousness has the potential to influence random physical eventsand that this effect is maximal when consciousness is exhibiting lsquolsquowavepropertiesrsquorsquo rather than lsquolsquoparticlersquorsquo properties Although it is unclear howconsciousness can be characterised in physical terms the analogy has interestingimplications when taken a step further Jahn and Dunne propose that the waveproperties of consciousness correlate best with a state in which individuals areable to divert their attention away from their self-awareness in relation to eventsaround them This analogy suggests that states of reduced self-awareness mayfacilitate the effects of consciousness on physical phenomena Self-awareness isa highly complex neurological function comprised of several hierarchical levelsranging from visceral knowledge to more abstract concepts of self-image There isa well-established literature suggesting that this function is mediated by the frontallobes and that frontal lesions are associated with reduced self-awareness (Stuss ampBenson 1986 Stuss 1991 Carver amp Scheier 1991) a state that is not easily

652 M Freedman et al

achievable by normal individualsStudying patientswith frontal lobe lesions mayprovide a good model for testing the hypotheses from the PEAR lab within thecontext of a conceptual framework that would maximise the likelihood ofdetecting effects if these in fact exist We report our findings in patients withfrontal lobe lesions and in normal subjects as well as replication data in onesubject with left frontal brain damage

Methods

Subjects

Subjects with frontal lobe brain damage (n ˆ 6) and normal subjects withoutbrain lesions (n ˆ 6) were studied The frontal group consisted of subjects withlesions in the following brain regions bilateral frontal lobes (n ˆ 4) left frontallobe (n ˆ 1) and right frontal lobe (n ˆ 1) Table 1 shows the age and gender ofthe subjects and the etiology of the brain lesions in the patients All subjects withfrontal brain damage except S1 had a CT or MRI scan of the head document-ing the site of the lesion The CT and MRI scans were interpreted blind to thehypotheses by observers with experience in interpreting neuroradiological scansThe patient without a scan had a bilateral frontal leucotomy many years earlierThe normal subjects were comprised of four research staff at one of the studysites (two of whom are authors) and two relatives of the research staff

Experimental Procedure

Subjects were asked to sit in front of a computer monitor showing an imageof a horizontal arrow or bar The midline of the screen was indicated by a lineThe subjects were instructed to concentrate on moving the image to the right(intention right) or left (intention left) of the midline or not to concentrate on

TABLE 1Subject Profiles

Subject Age Gender Etiology

S1 Bilateral frontal 70 F Frontal leucotomyS2 Bilateral frontal 60 F Subarachnoid hemorrhageS3 Bilateral frontal 58 M Subarachnoid hemorrhageS4 Bilateral frontal 61 M Frontal leucotomyS5 Left frontal 45 M Tension pneumocephalusS6 Right frontal 70 F InfarctS7 Normal 25 F NAa

S8 Normal 43 M NAS9 Normal 61 M NA

S10 Normal 69 M NAS11 Normal 51 M NAS12 Normal 26 F NA

a NA ˆ not applicable

Effects of Frontal Lobe Lesions on Intentionality 653

moving the image (baseline intention) For the right and left intentions theimage was an arrow for the baseline intention it was a bar (Figure 1 andashc) Thepurpose of selecting an arrow or bar was to provide an ongoing cue to helpsubjects maintain their attention on the specified intent

The order of intentions (right left or baseline) followed a predeterminedsequence as shown below (ie R L B L B ) Subjects who completed more thanthe following 12 intentions repeated the same cycle until they finished the study

R L BL B RR B LL R B

All six patients and two of the six normal subjects were tested with anexaminer in the room The four normal subjects who were research staff (S7 S8S11 S12) were tested without an examiner present The latter is in keeping withthe protocol in the PEAR program where laboratory staff maintain a minimalpresence during the experiments (Jahn et al 1987a) Each of the six patientsand the two normal subjects tested by an examiner were initially seated facingthe examiner and were given the following instructions lsquolsquoThere are some peoplewho believe that if we concentrate on something hard enough we can affect howthings happen Now we donrsquot know if this is true but we have undertaken to testthis out We would like to see if there is a possibility that people can influencesomething just by concentrating on itrsquorsquo Subjects were then instructed to face thecomputer screen which displayed an arrow pointing in the right or in the leftdirection with the tip at the midline The experimenter then continued with theinstructions lsquolsquoNow on this screen there is an arrow What I would like you to dois concentrate on making the arrow move in the direction that it is pointingI want to see how your concentration can affect the position of the arrow Thearrow will sometimes be on the right and sometimes on the left of the screen butI want you to keep the arrow on the leftright side as much as possible Do youhave any questions Remember I want you to try to keep the arrow on the leftright side as much as possible Irsquod like you to begin nowrsquorsquo

Fig 1 Illustration of computer screen showing initial position of arrow or bar for each intention(a) Intention Right (b) Intention Left and (c) Intention Baseline

654 M Freedman et al

The examiner then started the computer program which was designed to movethe arrow according to the output of a Random Event Generator (REG) The REGproduces a random series of 0rsquos and 1rsquos based upon a sampling of an electronicnoise pattern at pre-set regular intervals The sign of the signal at the time ofsampling (ie positive or negative) is compared with a regularly alternatingsequence of positive and negative pulses If there is a match (ie negative-negative or positive-positive) a lsquolsquo1rsquorsquo is generated If there is no match then a lsquolsquo0rsquorsquois generated (Jahn et al 1987a) Sampling occurred at a rate of 200 per secondThe data were summed as 200-sample bits with expected numerical value of 100due to a 50 chance of a positive-positive or negative-negativematch occurringat each sampling Each sample of 200 bits of data comprised a trial

The position of the arrow relative to the midline of the screen movedrightward or leftward according to the cumulative mean of the trials with themidline representing a cumulative mean of 100 An arrow with the tip to theright of midline represented a mean of greater than 100 and an arrow with the tipto the left corresponded to a mean of less than 100 The REG used for the ex-periment was a portable model of the larger device that has been used in thePEAR program at Princeton University (Jahn et al 1987a Nelson et al 1984Nelson et al 1989 Jahn et al 1987b Jahn et al 1997) The portable REG wasobtained from the PEAR program

Each intention (right left or baseline) consisted of 10 blocks of 100 trialsand lasted approximately 15 minutes At the end of each block of 100 trialsthe examiner confirmed that the subject understood the task answered anyquestions and initiated the next block of 100 trials After each block of 100trials the position of the arrow tip or bar was reset to the midline

After each intention for right left or baseline a control run of 1000 trials wascarried out without anyone in the room This completed one full session of 1000trials for the relevant intention and 1000 control trials Participation was spreadacross several sessions The total number of trials varied across subjects due todifferences in the time that they could devote to the study

Results

Primary analyses were carried out to test whether there were significanteffects of intention in the frontal subjects (bilateral frontal left frontal rightfrontal and frontal subjects pooled) and in the normal group Secondary analyseswere carried out to determine whether there were effects for individual subjects

Primary Analyses

Figure 2 (andashc) shows the mean output of the REG for each intention (rightleft baseline) and the mean control output for the patients with frontal lesionsand the normal subjects Table 2 shows the corresponding number of intentionand control trials

Effects of Frontal Lobe Lesions on Intentionality 655

T-tests were carried out using SAS System for Windows (The SAS Systemfor Windows Release 612 1996) to determine whether there were signifi-cant differences between intention and control conditions for the right left andbaseline intentions Separate analyses were carried out for patients with lesionsin the frontal lobes bilaterally left frontal lobe right frontal lobe all frontalpatients pooled and for the normal subjects This resulted in a total of 15analyses Using a Bonferroni correction for 15 multiple comparisons a p-value 00515 or 0003 would be required for statistical significance As shown in

Fig 2 Mean REG output for patients with brain lesions in bilateral frontal (BF) left frontal (LF)and right frontal (RF) regions frontal patients pooled (FR) and normal subjects (NORM)(a) Intention Right (b) Intention Left and (c) Intention Baseline

656 M Freedman et al

Table 3 there were no significant differences between the intention and controlconditions for the right left or baseline intentions in the bilateral frontal sub-jects right frontal subject or normal subjects In addition there were no signifi-cant differences between the intention and control conditions for the pooledgroup of frontal subjects

For the left frontal patient there was a significant difference between theintention and control conditions for the right intention (p ˆ 00015) The effectwas in the direction of intention and was significant even after Bonferroni cor-rection for multiple comparisons There were no significant differences betweenthe intention and control conditions for the left or baseline intentions in the leftfrontal patient

T-tests were carried out to determine whether there were significantdifferences between the intention and control conditions for the right left orbaseline intentions for the entire sample of frontal and normal subjects pooledThere were no significant differences for any intention (right t[304898] ˆ00480 p ˆ 096 left t[302998] ˆ 06709 p ˆ 050 baseline t[299998] ˆiexcl17053 p ˆ 009)

Secondary Analyses

Secondary analyses were carried out using t-tests to determine whether therewere significant differences between intention and control conditions for eachsubject Table 4 shows the results of the analysis for each subject Significantdifferences at the p ˆ 005 level were found between the intention and controlconditions in three instances intention right (p ˆ 00015) (S5 left frontalpatient) baseline (p ˆ 003) (S8 normal subject) intention right (p ˆ 0004) (S9

Fig 2 (Continued)

Effects of Frontal Lobe Lesions on Intentionality 657

normal subject) The effect was in the intended direction for S5 and S9 Usinga Bonferroni correction for 36 multiple comparisons a p-value 00536 or00014 would be required for statistical significance None of these p-valuesexceeded this threshold

As stated in the methodology section four normal subjects were research staff(S7 S8 S11 S12) and were tested without an examiner in the room In additionsubjects S9 and S10 were relatives of the research staff but were tested with anexaminer present These six subjects were divided into two groups for separateanalysis One group was comprised of the normal subjects who were researchstaff (S7 S8 S11 S12) and the other group was comprised of subjects S9 andS10 T-tests were carried out to determine whether there were significantdifferences between the intention and control conditions for the right left andbaseline intentions for these two groups This resulted in a total of six analysesUsing a Bonferroni correction for six multiple comparisons a p-value of 0056 or 0008 would be required for statistical significance There were no sig-nificant differences between the intention and control conditions for the groupcomprised of subjects S7 S8 S11 and S12 (right p ˆ 069 left p ˆ 061baseline p ˆ 024) For the group comprised of subjects S9 and S10 there wasa difference between the intention and control conditions for the right intention(p ˆ 00164) that was in the direction of intention However this was notsignificant after correction for multiple comparisons There were no significantdifferences between the intention and control conditions for the left (p ˆ 014) orbaseline (p ˆ 048) intentions for the group comprised of S9 and S10

In addition to statistical control for multiple comparisons using a Bonferronicorrection we performed an additional study to experimentally control for thenumerous statistical tests lsquolsquoPseudodatarsquorsquo were created by repeating the entire ex-perimental procedure except that there was no subject or experimenter in theroom during the intention condition Two sets of pseudodata were generated thefirst at the same site where the original data were collected (pseudodata 1) andthe second in a different building (pseudodata 2) The data were labeled to cor-respond to the intentions of the subjects as in the real study

As shown in Tables 5a and 5b the number of statistically significant resultsranged from one (pseudodata 1) to four (pseudodata 2) Moreover one of the

TABLE 2Total Number of Trials (Includes Intention Plus Control Trials)

Subjects Right Left Baseline

Bilateral frontals 200900a 201000 200000Left frontal 6000 8000 6000Right frontal 2000 2000 2000Frontals pooled 208900a 211000 208000Normals 96000 92000 92000

a There were 100 fewer intention trials than control trials for the right intention in the bilateralfrontals

658 M Freedman et al

significance levels was greater in the pseudodata than in the real data Howevernone of the effects were in the direction of intention and in three instances thesignificant effects occurred during the baseline intention

Replication Study

The primary analyses showed a statistically significant effect in the direc-tion of intention to the right for the patient with a left frontal lesion This effectwas significant even after Bonferroni correction for multiple comparisons Toexamine the possible effects in this patient further we carried out a second studyto determine whether the findings could be replicated The planned hypothesiswas that he would show a statistically significant effect of intention to the rightbut not to the left or in the baseline intention (ie that the findings would be thesame as before) The left frontal subject was retested using the methodologydescribed for the original study

During testing a printer error (data was always printed after each block of 100trials) occurred at the end of the fifth block of 100 trials during a baseline sessionand the program exited while the bar was still on the screen Although thecomplete data for these 100 trials were still captured a replacement set of 1000baseline and control trials for the entire intention was collected on a separateday However the possible impact of the printer error is considered negligibledue to the small number of trials involved the fact that none of the data werelost and because it occurred during the baseline condition in which the subjectwas asked not to concentrate on moving the arrow On another occasiona baseline session should have preceded a right intention session but the order was

TABLE 3Analysis of Frontal and Normal Subjects (Intention vs Control Conditions)

Subjects Intention t df p-value

Bilateral frontals right 11920 200898 023left 01213 200998 090baseline iexcl13874 199998 017

Left frontal right iexcl31691 5998 00015R

left 00524 7998 096baseline iexcl03656 5998 071

Right frontal right 00375 1998 097left iexcl08754 1998 038baseline iexcl16084 1998 011

Frontals pooled right 06377 208898 052left 00441 210998 096baseline iexcl15794 207998 011

Normals right iexcl08538 95998 039left 11474 91998 025baseline iexcl07036 91998 048

Note R ˆ Direction of effect was to the right

Effects of Frontal Lobe Lesions on Intentionality 659

reversed This resulted in two right intention and two baseline sessions being runconsecutively The right intention sessions were on the same day whereas thebaseline sessions were on different days Another occurrence consisted of thesubject seeing double arrows during a right intention session Finally during a left

TABLE 4Single Subject Analysis

Mean REG Output

Subject Intention Intention Control t p-value Trials (n)

S1 right 9998 10001 04982 062 40000left 9997 10003 08491 040 40000baseline 10003 10004 02000 084 40000

S2 right 9999 10007 08302 041 22900a

left 9988 9991 03145 075 24000baseline 10003 10001 iexcl01708 086 24000

S3 right 9989 10002 17414 008 34000left 10011 10000 iexcl13408 018 33000baseline 10013 10004 iexcl10462 030 32000

S4 right 10002 10002 iexcl00341 097 104000left 9999 10000 02488 080 104000baseline 10003 9997 iexcl13896 016 104000

S5 right 10019 9962 iexcl31691 00015R 6000left 9999 10000 00524 096 8000baseline 10001 9994 iexcl03656 071 6000

S6 right 10007 10008 00375 097 2000left 9991 9964 iexcl08754 038 2000baseline 10009 9959 iexcl16084 011 2000

S7 right 9984 9991 06708 050 16000left 10002 9998 iexcl03481 073 16000baseline 9994 9994 iexcl00022 100 16000

S8 right 10000 10006 07071 048 24000left 10003 10002 iexcl01489 088 24000baseline 10009 9989 iexcl21274 00334R 24000

S9 right 10021 9991 iexcl28790 00040R 18000left 10000 10012 10737 028 16000baseline 9994 9999 04375 066 16000b

S10 right 10001 10004 01845 085 6000left 9988 10014 11721 024 4000baseline 9974 9990 07088 048 4000

S11 right 10002 10002 iexcl00477 096 16000left 9999 10005 06132 054 16000baseline 10005 10004 iexcl01077 091 16000

S12 right 10007 10000 iexcl06514 051 16000left 9995 10007 10216 031 16000baseline 10000 10003 02445 081 16000

Note n ˆ number of intention trials Dagger control trials R ˆ direction of effect was to the righta There were 100 fewer intention trials than control trialsb 500 baseline control trials were collected incorrectly and replaced by trials collected at a different

time Although this introduced a potential bias the difference between the intention and controlconditions was too far from significant for a small number of 500 trials to have had an impact

660 M Freedman et al

intention session the building intercom came on at the end of the fourth block of100 trials The subject stated that this affected his concentration only slightly Inall cases described above the complete set of data was used for analysesincluding the replacement trials that followed the printer error (ie no data werediscarded)

TABLE 5aPseudodata Analysis 1

Mean REG Output

Subject Intention Intention Control t p-value Trials (n)

PS1 right 9997 10005 11636 024 40000left 10011 10003 iexcl10648 029 40000baseline 10007 9995 iexcl17586 008 40000

PS2 right 9995 10010 15873 011 22900left 9998 10004 06576 051 24000baseline 9994 10006 14214 016 24000

PS3 right 9996 10002 08405 040 34000left 9991 10002 13635 017 32000a

baseline 10007 9999 iexcl09442 035 32000PS4 right 10003 9999 iexcl10667 029 104000

left 10006 9999 iexcl17887 007 104000baseline 10001 9999 iexcl02984 077 104000

PS5 right 9982 9993 06322 053 6000left 10020 10000 iexcl12473 021 8000baseline 9988 9993 02343 081 6000

PS6 right 9999 10031 10198 031 2000left 9979 10006 08638 039 2000baseline 9963 10031 21474 00319L 2000

PS7 right 10002 10005 02463 081 16000left 9999 10013 12288 022 16000baseline 10005 9999 iexcl05196 060 16000

PS8 right 10006 9996 iexcl11546 025 24000left 10005 10007 02370 081 24000baseline 10000 9997 iexcl02564 080 24000

PS9 right 9995 10011 15199 013 18000left 9996 10002 05973 055 16000baseline 10005 9995 iexcl08764 038 16000

PS10 right 10007 9993 iexcl07168 047 6000left 9997 10016 08457 040 4000baseline 9981 9987 02792 078 4000

PS11 right 10005 10014 08604 039 16000left 10001 10006 04769 063 16000baseline 10009 9997 iexcl10813 028 16000

PS12 right 9993 10010 15528 012 16000left 10001 10012 09288 035 16000baseline 9998 10013 13483 018 16000

Note n ˆ number of intention trials Dagger control trials L ˆ direction of effect was to the lefta Pseudodata contained 500 fewer intention and control trials as compared to real data

Effects of Frontal Lobe Lesions on Intentionality 661

Results of Replication Study

Figure 3 shows the mean REG output for each intention (right left baseline)and the control output for the left frontal subject There were 19000 intentionand control trials for the right intention and 20000 trials for each of the inten-tion and control conditions for the left and baseline intentions

TABLE 5bPseudodata Analysis 2

Mean REG Output

Subject Intention Intention Control t p-value Trials (n)

PS1 right 9995 10001 08327 041 40000left 9996 9995 iexcl02329 082 40000baseline 10003 10000 iexcl04994 062 40000

PS2 right 10001 9995 iexcl06746 050 22900left 10000 9991 iexcl09721 033 24000baseline 9996 9999 03060 076 24000

PS3 right 10011 10008 iexcl04347 066 34000left 9998 10001 03522 072 32000a

baseline 10005 10000 iexcl06143 054 32000PS4 right 9994 10005 26336 00084L 104000

left 10001 9995 iexcl13071 019 104000baseline 9998 9997 iexcl02664 079 104000

PS5 right 9986 10005 10582 029 6000left 10003 10005 01187 091 8000baseline 10002 9995 iexcl03727 071 6000

PS6 right 9968 9993 07972 043 2000left 9990 9976 iexcl04402 066 2000baseline 10008 9960 iexcl15583 012 2000

PS7 right 9990 9995 04285 067 16000left 10018 10002 iexcl14678 014 16000baseline 9983 9996 11654 024 16000

PS8 right 10008 10005 iexcl02559 080 24000left 10016 9984 iexcl35151 00004R 24000baseline 10002 10002 iexcl00648 095 24000

PS9 right 10001 9998 iexcl02441 081 18000left 10017 9998 iexcl16610 010 16000baseline 9993 9983 iexcl08404 040 16000

PS10 right 10009 10014 02525 080 6000left 10018 10000 iexcl07789 044 4000baseline 9987 10022 15658 012 4000

PS11 right 10006 10010 03433 073 16000left 9992 10000 07129 048 16000baseline 10007 9986 iexcl19258 00541R 16000

PS12 right 9997 9986 iexcl10275 030 16000left 10002 9999 iexcl02967 077 16000baseline 9978 10001 20961 00361L 16000

Note n ˆ number of intention trials Dagger control trials R ˆ direction of effect was to the right L ˆdirection of effect was to the left

a Pseudodata contained 500 fewer intention and control trials as compared to real data

662 M Freedman et al

T-tests were carried out using SAS System for Windows Release 8 todetermine whether there were significant differences between the intention andcontrol conditions for the right left and baseline intentions There wasa significant difference between the intention and control conditions for the rightintention (t[37998]ˆiexcl253 pˆ 00115) but not for the other two intentions (leftt[39998] ˆ 049 p ˆ 06269 and baseline t[39998] ˆ iexcl007 p ˆ 09418) Thesignificant effect for the right intention was in the direction of intention as in theoriginal study

Since the investigation was run in blocks of 1000 trials we examined the datausing blocks as the unit of measurement and the averaged output across trials asthe measure in each block An F-ratio test with condition (intention vs control)as the source of variance of interest and the interaction between condition andblock as the error term showed an effect of condition (p ˆ 00578) This suggeststhat the left frontal subject would show comparable findings on another series of19 pairs of blocks of 1000 trials with 94 confidence

Whereas the mean REG output for the right intention was above the expectedvalue of 100 the output for the control condition was below 100 (Figure 3) Toexamine the possibility that the significant difference between the intention andcontrol condition was due solely to low output during the control condition asopposed to high output during the intention condition we tested whether theoutput for the intention condition was significantly different from a constantvalue of 100 This value is the theoretical mean output of the REG assumingtruly random and unbiased output The control values for the left and baselineconditions are approximately equal to this theoretical mean The REG output forthe right intention was significantly different than 100 and in the direction ofintention (t ˆ 201 p ˆ 0045 two-tailed test) For the control condition themean REG output was less than (although not significantly different from) 100(t ˆ iexcl156 p ˆ 012)

Fig 3 Mean REG output for the left frontal patient in replication study

Effects of Frontal Lobe Lesions on Intentionality 663

Figure 4 (andashc) shows the mean output of the REG broken down byconsecutive 1000 trial sessions for each intention (right left and baseline) andfor the control output For the right intention Figure 4 shows that on most pairsof blocks of 1000 trials the mean REG output follows a fairly consistent patternin which the means are either higher compared to the control condition or aboutthe same In contrast the data for each of the left and baseline conditions showa pattern with less separation between conditions

Fig 4 Mean REG output for each 1000 trial session in replication study (a) Intention Right (b)Intention Left and (c) Intention Baseline

664 M Freedman et al

Discussion

We examined the claims from the PEAR program that an individualrsquosintentions can influence the statistical distribution of random physical phenom-ena Our main focus was to test these claims using well-designed control condi-tions in a population that might maximize the likelihood of detecting any effectsthat may exist Our choice of studying patients with frontal lobe lesions wasbased on the concept that the potential to influence random physical events maybe optimized when attention is diverted from self-awareness (ie such as whenself-awareness is reduced) This state can occur following frontal brain damage(Stuss et al 1986 Stuss 1991 Carver et al 1991) Our findings showed asignificant effect for the right intention in a subject with left frontal braindamage Moreover this effect was in the direction of intention In contrast therewere no significant effects in the other groups (ie bilateral frontal right frontalpooled frontal or normal subjects)

Although the result in the left frontal patient on the right intention wasstatistically significant even after correction for multiple comparisons thisfinding was interpreted with great caution First it was based on a relativelysmall number of intention trials (n ˆ 3000) and was derived from only onesubject Second we found a p-value for an individual lsquolsquosubjectrsquorsquo in the pseu-dodata that would also have met criteria for significance using a Bonferronicorrection and that was even less than the p-value of 00015 obtained fromthe left frontal subject Although the effect of this lsquolsquopseudopatientrsquorsquo was not inthe direction of intention the fact that it was significant raised caution forinterpreting the effect in the left frontal patient However the replication of thefindings in the left frontal patient in a second well-designed study for each of the

Fig 4 (Continued)

Effects of Frontal Lobe Lesions on Intentionality 665

three intentions right left and baseline suggests that the effect in this subjectmay be more than chance occurrence

Additional support for a reliable effect in the left frontal subject comes fromfurther analysis of his data suggesting about a 94 chance of his showinga similar finding if another series of 19 pairs of blocks of 1000 trials were runagain Furthermore the profile of REG output data for the right intentionshowed a fairly consistent separation whereas this was not the case for the left orbaseline intentions

A comment is warranted about the REG output for the left frontal patientbeing lower for the control condition on the right intention or more to the leftas compared to the control conditions for the left or baseline intentions Onemight argue that the significant effect on the right intention was an artifact ofcomparison to control data that was in the left direction and that this widened thedifference between the intention and control data However this argument is nottenable because the effect on the right intention was significant even when theREG output was compared to a theoretical mean of 100mdasha value which is ap-proximately equal to the control means for the left intention (9999985) and thebaseline intention (1000011)

The patientrsquos cognitive deficits and brain lesion have been describedelsewhere (Marras et al 1998) He suffered from a tension pneumocephaluswhich resulted in cognitive deficits and epileptic seizures The cognitive deficitsinclude decreased mental flexibility on the Trail-Making Test poor attentionreduced fluency and impaired spatial planning and visuospatial problem solvingMRI showed an extensive left frontal lesion but the right frontal lobe was intactPsychometric testing and SPECT suggested the addition of right frontal dysfunc-tion The SPECT findings provide a measure of function as opposed to structureand were subtle

As indicated above frontal lesions have been associated with reduced self-awareness (Stuss et al 1986 Stuss 1991 Carver et al 1991) a state that isdifficult for normal individuals to achieve The rationale for studying patientswith damage to the frontal lobes was that decreased self-awareness might facili-tate the effects of intentionality on random physical phenomena Brain regionsthat mediate neurological processes underlying self-awareness include thefrontal lobes bilaterally particularly on the right (Stuss et al 2001a Stuss ampAlexander 2000a Stuss amp Alexander 2000b Stuss et al 2001b) Whereas thepatientrsquos right-sided brain dysfunction may have been insufficient to producea deficit in self-awareness the extensive lesion on the left may have resulted inreduced self-awareness when attention was directed towards the right Howeverit remains unclear why positive results should be found only following damageto the left frontal region and not after bilateral or right frontal lesions Onespeculation is that the effect on random physical events may require reducedself-awareness combined with relatively intact attentional mechanisms Theassociation of frontal lobe lesions especially on the right with impaired at-tention (Stuss amp Levine 2002) may explain the negative findings in the setting

666 M Freedman et al

of bilateral or right frontal damage Whether deficits related to left frontalabnormalities can explain the observed effects on the REG output warrantsfurther study Moreover the question whether normal processes associated withintact frontal lobe function together with preserved self-awareness may serve toinhibit effects of intentionality on random physical phenomena needs to beaddressed

The strength of our conclusions rests largely on a well-designed methodol-ogy and replication of our findings Although our results did not replicate thefindings reported by Jahn and his colleagues in normal subjects (Jahn et al1987a 1997) they support their claims that intentionality can alter the output ofa random event generator Furthermore our findings suggest that patients withfrontal lobe lesions may serve as a good model for future studies of the effects ofconsciousness on random physical events

Acknowledgments

We gratefully acknowledge Dr Donald Stuss for his helpful advice andsupport throughout the study and for reviewing the manuscript and for hishelpful suggestions Dr Louis Siminovitch for helpful advice and supportPatricia Ebert and Dr Donald Stuss for assisting with analysis of lesion local-ization for the CT and MRI scans Janice Pogue for statistical advice LeeFerreira for testing the subjects John Sloan for assistance with data collec-tion Tania Xerri for assistance with data analyses and Tania Xerri MaedeanBrown Wanna Mar and Melissa Ziraldo for preparation of materials for pub-lication We also gratefully acknowledge Professor Robert Jahn BrendaDunne Dr Roger Nelson Dr York Dobyns and John Bradish at the PEARlab for their helpful advice and for providing the portable REG and softwarefor the study This study was supported by the Fetzer Institute Dr Freedmanwas supported by the Saul A Silverman Family Foundation Toronto Canadaas a Canada-International Scientific exchange program (CISEPO) project

References

Alcock J E (1987) ParapsychologyScience of the Anomalous or Search for the Soul Behavioraland Brain Sciences 10 553ndash643

Carver C S amp Scheier M F (1991) Self-Regulation and the Self In J Strauss amp G R Goethals(Eds) The Self Interdisciplinary Approaches (pp 168ndash207) New York Springer-Verlag

Hall J Kim C McElroy B amp Shimony A (1977) Wave-packet reduction as a medium ofcommunication Foundations of Physics 7 759ndash767

Ibison M amp Jeffers S (1998) A Double-Slit Diffraction Experiment to Investigate Claims ofConsciousness-Related Anomalies Journal of Scienti c Exploration 12 543ndash550

Jahn R G amp Dunne B J (1986) On the Quantum Mechanics of Consciousness with Applicationto Anomalous Phenomena Foundations of Physics 16 721ndash772

Jahn R G amp Dunne B J (1987a) Margins of Reality The Role of Consciousness in the PhysicalWorld San Diego Harcourt Brace Jovanovich

Jahn R G Dunne B J amp Nelson R D (1987b) Engineering Anomalies Research Journal ofScienti c Exploration 1 21ndash50

Effects of Frontal Lobe Lesions on Intentionality 667

1987a Radin amp Nelson 1989 Jahn et al 1997) concerns about research designand a lack of theoretical models (Alcock 1987 Jeffers in press) as well asnegative studies (Hall et al 1977 Jeffers amp Sloan 1992 Ibison amp Jeffers1998) have been important sources of criticism of the literature in this areaA major methodological problem in research design relates to the issue ofinadequate experimental controls (Jeffers in press) For example Robert Jahnand his colleagues from the Princeton Engineering Anomalies Research (PEAR)programhave reported statisticallysignificanteffects whereby subjectsor lsquolsquoopera-torsrsquorsquo have successfully influenced the statistical distribution of outcomes froma Random Event Generator (REG) (Jahn et al 1987a Jahn et al 1997)A concern relates to the nature of the machine calibration data used forcomparison to subject generated data (Jeffers in press) Rather than collectingcontrol data in close temporal proximity to the period when a subject is trying toinfluence the REG output the PEAR protocol relies on the assumption that theREG output is always random in the absence of operator influence This approachdoes not control for potentially unrecognised factors that may affect the REGoutput and therefore casts doubt on the interpretation of the experimentalfindings In addition there are theoretical issues raising questions about thefindings from Jahnrsquos group such as the independenceof the reported effects fromtime and distance that are difficult to reconcile (Jahn et al 1997 Jeffers inpress) Nevertheless Jahn and his colleagues have amassed a great wealth of datato support their conclusions that an individualrsquos conscious intention can alter thestatistical distribution of random physical phenomena Because their findingswould have immense significance if validated we re-examined their claimsusing a methodology with well-designed control conditions

Some highly interesting but speculative ideas relating anomalous physicalactivity to consciousness have been advanced by Jahn and Dunne (Jahn et al1987a Jahn amp Dunne 1986) They proposed a metaphor for consciousnessbased upon quantum mechanical concepts that relates consciousness toanomalous physical phenomena Based on data from the PEAR lab theysuggest that consciousness has the potential to influence random physical eventsand that this effect is maximal when consciousness is exhibiting lsquolsquowavepropertiesrsquorsquo rather than lsquolsquoparticlersquorsquo properties Although it is unclear howconsciousness can be characterised in physical terms the analogy has interestingimplications when taken a step further Jahn and Dunne propose that the waveproperties of consciousness correlate best with a state in which individuals areable to divert their attention away from their self-awareness in relation to eventsaround them This analogy suggests that states of reduced self-awareness mayfacilitate the effects of consciousness on physical phenomena Self-awareness isa highly complex neurological function comprised of several hierarchical levelsranging from visceral knowledge to more abstract concepts of self-image There isa well-established literature suggesting that this function is mediated by the frontallobes and that frontal lesions are associated with reduced self-awareness (Stuss ampBenson 1986 Stuss 1991 Carver amp Scheier 1991) a state that is not easily

652 M Freedman et al

achievable by normal individualsStudying patientswith frontal lobe lesions mayprovide a good model for testing the hypotheses from the PEAR lab within thecontext of a conceptual framework that would maximise the likelihood ofdetecting effects if these in fact exist We report our findings in patients withfrontal lobe lesions and in normal subjects as well as replication data in onesubject with left frontal brain damage

Methods

Subjects

Subjects with frontal lobe brain damage (n ˆ 6) and normal subjects withoutbrain lesions (n ˆ 6) were studied The frontal group consisted of subjects withlesions in the following brain regions bilateral frontal lobes (n ˆ 4) left frontallobe (n ˆ 1) and right frontal lobe (n ˆ 1) Table 1 shows the age and gender ofthe subjects and the etiology of the brain lesions in the patients All subjects withfrontal brain damage except S1 had a CT or MRI scan of the head document-ing the site of the lesion The CT and MRI scans were interpreted blind to thehypotheses by observers with experience in interpreting neuroradiological scansThe patient without a scan had a bilateral frontal leucotomy many years earlierThe normal subjects were comprised of four research staff at one of the studysites (two of whom are authors) and two relatives of the research staff

Experimental Procedure

Subjects were asked to sit in front of a computer monitor showing an imageof a horizontal arrow or bar The midline of the screen was indicated by a lineThe subjects were instructed to concentrate on moving the image to the right(intention right) or left (intention left) of the midline or not to concentrate on

TABLE 1Subject Profiles

Subject Age Gender Etiology

S1 Bilateral frontal 70 F Frontal leucotomyS2 Bilateral frontal 60 F Subarachnoid hemorrhageS3 Bilateral frontal 58 M Subarachnoid hemorrhageS4 Bilateral frontal 61 M Frontal leucotomyS5 Left frontal 45 M Tension pneumocephalusS6 Right frontal 70 F InfarctS7 Normal 25 F NAa

S8 Normal 43 M NAS9 Normal 61 M NA

S10 Normal 69 M NAS11 Normal 51 M NAS12 Normal 26 F NA

a NA ˆ not applicable

Effects of Frontal Lobe Lesions on Intentionality 653

moving the image (baseline intention) For the right and left intentions theimage was an arrow for the baseline intention it was a bar (Figure 1 andashc) Thepurpose of selecting an arrow or bar was to provide an ongoing cue to helpsubjects maintain their attention on the specified intent

The order of intentions (right left or baseline) followed a predeterminedsequence as shown below (ie R L B L B ) Subjects who completed more thanthe following 12 intentions repeated the same cycle until they finished the study

R L BL B RR B LL R B

All six patients and two of the six normal subjects were tested with anexaminer in the room The four normal subjects who were research staff (S7 S8S11 S12) were tested without an examiner present The latter is in keeping withthe protocol in the PEAR program where laboratory staff maintain a minimalpresence during the experiments (Jahn et al 1987a) Each of the six patientsand the two normal subjects tested by an examiner were initially seated facingthe examiner and were given the following instructions lsquolsquoThere are some peoplewho believe that if we concentrate on something hard enough we can affect howthings happen Now we donrsquot know if this is true but we have undertaken to testthis out We would like to see if there is a possibility that people can influencesomething just by concentrating on itrsquorsquo Subjects were then instructed to face thecomputer screen which displayed an arrow pointing in the right or in the leftdirection with the tip at the midline The experimenter then continued with theinstructions lsquolsquoNow on this screen there is an arrow What I would like you to dois concentrate on making the arrow move in the direction that it is pointingI want to see how your concentration can affect the position of the arrow Thearrow will sometimes be on the right and sometimes on the left of the screen butI want you to keep the arrow on the leftright side as much as possible Do youhave any questions Remember I want you to try to keep the arrow on the leftright side as much as possible Irsquod like you to begin nowrsquorsquo

Fig 1 Illustration of computer screen showing initial position of arrow or bar for each intention(a) Intention Right (b) Intention Left and (c) Intention Baseline

654 M Freedman et al

The examiner then started the computer program which was designed to movethe arrow according to the output of a Random Event Generator (REG) The REGproduces a random series of 0rsquos and 1rsquos based upon a sampling of an electronicnoise pattern at pre-set regular intervals The sign of the signal at the time ofsampling (ie positive or negative) is compared with a regularly alternatingsequence of positive and negative pulses If there is a match (ie negative-negative or positive-positive) a lsquolsquo1rsquorsquo is generated If there is no match then a lsquolsquo0rsquorsquois generated (Jahn et al 1987a) Sampling occurred at a rate of 200 per secondThe data were summed as 200-sample bits with expected numerical value of 100due to a 50 chance of a positive-positive or negative-negativematch occurringat each sampling Each sample of 200 bits of data comprised a trial

The position of the arrow relative to the midline of the screen movedrightward or leftward according to the cumulative mean of the trials with themidline representing a cumulative mean of 100 An arrow with the tip to theright of midline represented a mean of greater than 100 and an arrow with the tipto the left corresponded to a mean of less than 100 The REG used for the ex-periment was a portable model of the larger device that has been used in thePEAR program at Princeton University (Jahn et al 1987a Nelson et al 1984Nelson et al 1989 Jahn et al 1987b Jahn et al 1997) The portable REG wasobtained from the PEAR program

Each intention (right left or baseline) consisted of 10 blocks of 100 trialsand lasted approximately 15 minutes At the end of each block of 100 trialsthe examiner confirmed that the subject understood the task answered anyquestions and initiated the next block of 100 trials After each block of 100trials the position of the arrow tip or bar was reset to the midline

After each intention for right left or baseline a control run of 1000 trials wascarried out without anyone in the room This completed one full session of 1000trials for the relevant intention and 1000 control trials Participation was spreadacross several sessions The total number of trials varied across subjects due todifferences in the time that they could devote to the study

Results

Primary analyses were carried out to test whether there were significanteffects of intention in the frontal subjects (bilateral frontal left frontal rightfrontal and frontal subjects pooled) and in the normal group Secondary analyseswere carried out to determine whether there were effects for individual subjects

Primary Analyses

Figure 2 (andashc) shows the mean output of the REG for each intention (rightleft baseline) and the mean control output for the patients with frontal lesionsand the normal subjects Table 2 shows the corresponding number of intentionand control trials

Effects of Frontal Lobe Lesions on Intentionality 655

T-tests were carried out using SAS System for Windows (The SAS Systemfor Windows Release 612 1996) to determine whether there were signifi-cant differences between intention and control conditions for the right left andbaseline intentions Separate analyses were carried out for patients with lesionsin the frontal lobes bilaterally left frontal lobe right frontal lobe all frontalpatients pooled and for the normal subjects This resulted in a total of 15analyses Using a Bonferroni correction for 15 multiple comparisons a p-value 00515 or 0003 would be required for statistical significance As shown in

Fig 2 Mean REG output for patients with brain lesions in bilateral frontal (BF) left frontal (LF)and right frontal (RF) regions frontal patients pooled (FR) and normal subjects (NORM)(a) Intention Right (b) Intention Left and (c) Intention Baseline

656 M Freedman et al

Table 3 there were no significant differences between the intention and controlconditions for the right left or baseline intentions in the bilateral frontal sub-jects right frontal subject or normal subjects In addition there were no signifi-cant differences between the intention and control conditions for the pooledgroup of frontal subjects

For the left frontal patient there was a significant difference between theintention and control conditions for the right intention (p ˆ 00015) The effectwas in the direction of intention and was significant even after Bonferroni cor-rection for multiple comparisons There were no significant differences betweenthe intention and control conditions for the left or baseline intentions in the leftfrontal patient

T-tests were carried out to determine whether there were significantdifferences between the intention and control conditions for the right left orbaseline intentions for the entire sample of frontal and normal subjects pooledThere were no significant differences for any intention (right t[304898] ˆ00480 p ˆ 096 left t[302998] ˆ 06709 p ˆ 050 baseline t[299998] ˆiexcl17053 p ˆ 009)

Secondary Analyses

Secondary analyses were carried out using t-tests to determine whether therewere significant differences between intention and control conditions for eachsubject Table 4 shows the results of the analysis for each subject Significantdifferences at the p ˆ 005 level were found between the intention and controlconditions in three instances intention right (p ˆ 00015) (S5 left frontalpatient) baseline (p ˆ 003) (S8 normal subject) intention right (p ˆ 0004) (S9

Fig 2 (Continued)

Effects of Frontal Lobe Lesions on Intentionality 657

normal subject) The effect was in the intended direction for S5 and S9 Usinga Bonferroni correction for 36 multiple comparisons a p-value 00536 or00014 would be required for statistical significance None of these p-valuesexceeded this threshold

As stated in the methodology section four normal subjects were research staff(S7 S8 S11 S12) and were tested without an examiner in the room In additionsubjects S9 and S10 were relatives of the research staff but were tested with anexaminer present These six subjects were divided into two groups for separateanalysis One group was comprised of the normal subjects who were researchstaff (S7 S8 S11 S12) and the other group was comprised of subjects S9 andS10 T-tests were carried out to determine whether there were significantdifferences between the intention and control conditions for the right left andbaseline intentions for these two groups This resulted in a total of six analysesUsing a Bonferroni correction for six multiple comparisons a p-value of 0056 or 0008 would be required for statistical significance There were no sig-nificant differences between the intention and control conditions for the groupcomprised of subjects S7 S8 S11 and S12 (right p ˆ 069 left p ˆ 061baseline p ˆ 024) For the group comprised of subjects S9 and S10 there wasa difference between the intention and control conditions for the right intention(p ˆ 00164) that was in the direction of intention However this was notsignificant after correction for multiple comparisons There were no significantdifferences between the intention and control conditions for the left (p ˆ 014) orbaseline (p ˆ 048) intentions for the group comprised of S9 and S10

In addition to statistical control for multiple comparisons using a Bonferronicorrection we performed an additional study to experimentally control for thenumerous statistical tests lsquolsquoPseudodatarsquorsquo were created by repeating the entire ex-perimental procedure except that there was no subject or experimenter in theroom during the intention condition Two sets of pseudodata were generated thefirst at the same site where the original data were collected (pseudodata 1) andthe second in a different building (pseudodata 2) The data were labeled to cor-respond to the intentions of the subjects as in the real study

As shown in Tables 5a and 5b the number of statistically significant resultsranged from one (pseudodata 1) to four (pseudodata 2) Moreover one of the

TABLE 2Total Number of Trials (Includes Intention Plus Control Trials)

Subjects Right Left Baseline

Bilateral frontals 200900a 201000 200000Left frontal 6000 8000 6000Right frontal 2000 2000 2000Frontals pooled 208900a 211000 208000Normals 96000 92000 92000

a There were 100 fewer intention trials than control trials for the right intention in the bilateralfrontals

658 M Freedman et al

significance levels was greater in the pseudodata than in the real data Howevernone of the effects were in the direction of intention and in three instances thesignificant effects occurred during the baseline intention

Replication Study

The primary analyses showed a statistically significant effect in the direc-tion of intention to the right for the patient with a left frontal lesion This effectwas significant even after Bonferroni correction for multiple comparisons Toexamine the possible effects in this patient further we carried out a second studyto determine whether the findings could be replicated The planned hypothesiswas that he would show a statistically significant effect of intention to the rightbut not to the left or in the baseline intention (ie that the findings would be thesame as before) The left frontal subject was retested using the methodologydescribed for the original study

During testing a printer error (data was always printed after each block of 100trials) occurred at the end of the fifth block of 100 trials during a baseline sessionand the program exited while the bar was still on the screen Although thecomplete data for these 100 trials were still captured a replacement set of 1000baseline and control trials for the entire intention was collected on a separateday However the possible impact of the printer error is considered negligibledue to the small number of trials involved the fact that none of the data werelost and because it occurred during the baseline condition in which the subjectwas asked not to concentrate on moving the arrow On another occasiona baseline session should have preceded a right intention session but the order was

TABLE 3Analysis of Frontal and Normal Subjects (Intention vs Control Conditions)

Subjects Intention t df p-value

Bilateral frontals right 11920 200898 023left 01213 200998 090baseline iexcl13874 199998 017

Left frontal right iexcl31691 5998 00015R

left 00524 7998 096baseline iexcl03656 5998 071

Right frontal right 00375 1998 097left iexcl08754 1998 038baseline iexcl16084 1998 011

Frontals pooled right 06377 208898 052left 00441 210998 096baseline iexcl15794 207998 011

Normals right iexcl08538 95998 039left 11474 91998 025baseline iexcl07036 91998 048

Note R ˆ Direction of effect was to the right

Effects of Frontal Lobe Lesions on Intentionality 659

reversed This resulted in two right intention and two baseline sessions being runconsecutively The right intention sessions were on the same day whereas thebaseline sessions were on different days Another occurrence consisted of thesubject seeing double arrows during a right intention session Finally during a left

TABLE 4Single Subject Analysis

Mean REG Output

Subject Intention Intention Control t p-value Trials (n)

S1 right 9998 10001 04982 062 40000left 9997 10003 08491 040 40000baseline 10003 10004 02000 084 40000

S2 right 9999 10007 08302 041 22900a

left 9988 9991 03145 075 24000baseline 10003 10001 iexcl01708 086 24000

S3 right 9989 10002 17414 008 34000left 10011 10000 iexcl13408 018 33000baseline 10013 10004 iexcl10462 030 32000

S4 right 10002 10002 iexcl00341 097 104000left 9999 10000 02488 080 104000baseline 10003 9997 iexcl13896 016 104000

S5 right 10019 9962 iexcl31691 00015R 6000left 9999 10000 00524 096 8000baseline 10001 9994 iexcl03656 071 6000

S6 right 10007 10008 00375 097 2000left 9991 9964 iexcl08754 038 2000baseline 10009 9959 iexcl16084 011 2000

S7 right 9984 9991 06708 050 16000left 10002 9998 iexcl03481 073 16000baseline 9994 9994 iexcl00022 100 16000

S8 right 10000 10006 07071 048 24000left 10003 10002 iexcl01489 088 24000baseline 10009 9989 iexcl21274 00334R 24000

S9 right 10021 9991 iexcl28790 00040R 18000left 10000 10012 10737 028 16000baseline 9994 9999 04375 066 16000b

S10 right 10001 10004 01845 085 6000left 9988 10014 11721 024 4000baseline 9974 9990 07088 048 4000

S11 right 10002 10002 iexcl00477 096 16000left 9999 10005 06132 054 16000baseline 10005 10004 iexcl01077 091 16000

S12 right 10007 10000 iexcl06514 051 16000left 9995 10007 10216 031 16000baseline 10000 10003 02445 081 16000

Note n ˆ number of intention trials Dagger control trials R ˆ direction of effect was to the righta There were 100 fewer intention trials than control trialsb 500 baseline control trials were collected incorrectly and replaced by trials collected at a different

time Although this introduced a potential bias the difference between the intention and controlconditions was too far from significant for a small number of 500 trials to have had an impact

660 M Freedman et al

intention session the building intercom came on at the end of the fourth block of100 trials The subject stated that this affected his concentration only slightly Inall cases described above the complete set of data was used for analysesincluding the replacement trials that followed the printer error (ie no data werediscarded)

TABLE 5aPseudodata Analysis 1

Mean REG Output

Subject Intention Intention Control t p-value Trials (n)

PS1 right 9997 10005 11636 024 40000left 10011 10003 iexcl10648 029 40000baseline 10007 9995 iexcl17586 008 40000

PS2 right 9995 10010 15873 011 22900left 9998 10004 06576 051 24000baseline 9994 10006 14214 016 24000

PS3 right 9996 10002 08405 040 34000left 9991 10002 13635 017 32000a

baseline 10007 9999 iexcl09442 035 32000PS4 right 10003 9999 iexcl10667 029 104000

left 10006 9999 iexcl17887 007 104000baseline 10001 9999 iexcl02984 077 104000

PS5 right 9982 9993 06322 053 6000left 10020 10000 iexcl12473 021 8000baseline 9988 9993 02343 081 6000

PS6 right 9999 10031 10198 031 2000left 9979 10006 08638 039 2000baseline 9963 10031 21474 00319L 2000

PS7 right 10002 10005 02463 081 16000left 9999 10013 12288 022 16000baseline 10005 9999 iexcl05196 060 16000

PS8 right 10006 9996 iexcl11546 025 24000left 10005 10007 02370 081 24000baseline 10000 9997 iexcl02564 080 24000

PS9 right 9995 10011 15199 013 18000left 9996 10002 05973 055 16000baseline 10005 9995 iexcl08764 038 16000

PS10 right 10007 9993 iexcl07168 047 6000left 9997 10016 08457 040 4000baseline 9981 9987 02792 078 4000

PS11 right 10005 10014 08604 039 16000left 10001 10006 04769 063 16000baseline 10009 9997 iexcl10813 028 16000

PS12 right 9993 10010 15528 012 16000left 10001 10012 09288 035 16000baseline 9998 10013 13483 018 16000

Note n ˆ number of intention trials Dagger control trials L ˆ direction of effect was to the lefta Pseudodata contained 500 fewer intention and control trials as compared to real data

Effects of Frontal Lobe Lesions on Intentionality 661

Results of Replication Study

Figure 3 shows the mean REG output for each intention (right left baseline)and the control output for the left frontal subject There were 19000 intentionand control trials for the right intention and 20000 trials for each of the inten-tion and control conditions for the left and baseline intentions

TABLE 5bPseudodata Analysis 2

Mean REG Output

Subject Intention Intention Control t p-value Trials (n)

PS1 right 9995 10001 08327 041 40000left 9996 9995 iexcl02329 082 40000baseline 10003 10000 iexcl04994 062 40000

PS2 right 10001 9995 iexcl06746 050 22900left 10000 9991 iexcl09721 033 24000baseline 9996 9999 03060 076 24000

PS3 right 10011 10008 iexcl04347 066 34000left 9998 10001 03522 072 32000a

baseline 10005 10000 iexcl06143 054 32000PS4 right 9994 10005 26336 00084L 104000

left 10001 9995 iexcl13071 019 104000baseline 9998 9997 iexcl02664 079 104000

PS5 right 9986 10005 10582 029 6000left 10003 10005 01187 091 8000baseline 10002 9995 iexcl03727 071 6000

PS6 right 9968 9993 07972 043 2000left 9990 9976 iexcl04402 066 2000baseline 10008 9960 iexcl15583 012 2000

PS7 right 9990 9995 04285 067 16000left 10018 10002 iexcl14678 014 16000baseline 9983 9996 11654 024 16000

PS8 right 10008 10005 iexcl02559 080 24000left 10016 9984 iexcl35151 00004R 24000baseline 10002 10002 iexcl00648 095 24000

PS9 right 10001 9998 iexcl02441 081 18000left 10017 9998 iexcl16610 010 16000baseline 9993 9983 iexcl08404 040 16000

PS10 right 10009 10014 02525 080 6000left 10018 10000 iexcl07789 044 4000baseline 9987 10022 15658 012 4000

PS11 right 10006 10010 03433 073 16000left 9992 10000 07129 048 16000baseline 10007 9986 iexcl19258 00541R 16000

PS12 right 9997 9986 iexcl10275 030 16000left 10002 9999 iexcl02967 077 16000baseline 9978 10001 20961 00361L 16000

Note n ˆ number of intention trials Dagger control trials R ˆ direction of effect was to the right L ˆdirection of effect was to the left

a Pseudodata contained 500 fewer intention and control trials as compared to real data

662 M Freedman et al

T-tests were carried out using SAS System for Windows Release 8 todetermine whether there were significant differences between the intention andcontrol conditions for the right left and baseline intentions There wasa significant difference between the intention and control conditions for the rightintention (t[37998]ˆiexcl253 pˆ 00115) but not for the other two intentions (leftt[39998] ˆ 049 p ˆ 06269 and baseline t[39998] ˆ iexcl007 p ˆ 09418) Thesignificant effect for the right intention was in the direction of intention as in theoriginal study

Since the investigation was run in blocks of 1000 trials we examined the datausing blocks as the unit of measurement and the averaged output across trials asthe measure in each block An F-ratio test with condition (intention vs control)as the source of variance of interest and the interaction between condition andblock as the error term showed an effect of condition (p ˆ 00578) This suggeststhat the left frontal subject would show comparable findings on another series of19 pairs of blocks of 1000 trials with 94 confidence

Whereas the mean REG output for the right intention was above the expectedvalue of 100 the output for the control condition was below 100 (Figure 3) Toexamine the possibility that the significant difference between the intention andcontrol condition was due solely to low output during the control condition asopposed to high output during the intention condition we tested whether theoutput for the intention condition was significantly different from a constantvalue of 100 This value is the theoretical mean output of the REG assumingtruly random and unbiased output The control values for the left and baselineconditions are approximately equal to this theoretical mean The REG output forthe right intention was significantly different than 100 and in the direction ofintention (t ˆ 201 p ˆ 0045 two-tailed test) For the control condition themean REG output was less than (although not significantly different from) 100(t ˆ iexcl156 p ˆ 012)

Fig 3 Mean REG output for the left frontal patient in replication study

Effects of Frontal Lobe Lesions on Intentionality 663

Figure 4 (andashc) shows the mean output of the REG broken down byconsecutive 1000 trial sessions for each intention (right left and baseline) andfor the control output For the right intention Figure 4 shows that on most pairsof blocks of 1000 trials the mean REG output follows a fairly consistent patternin which the means are either higher compared to the control condition or aboutthe same In contrast the data for each of the left and baseline conditions showa pattern with less separation between conditions

Fig 4 Mean REG output for each 1000 trial session in replication study (a) Intention Right (b)Intention Left and (c) Intention Baseline

664 M Freedman et al

Discussion

We examined the claims from the PEAR program that an individualrsquosintentions can influence the statistical distribution of random physical phenom-ena Our main focus was to test these claims using well-designed control condi-tions in a population that might maximize the likelihood of detecting any effectsthat may exist Our choice of studying patients with frontal lobe lesions wasbased on the concept that the potential to influence random physical events maybe optimized when attention is diverted from self-awareness (ie such as whenself-awareness is reduced) This state can occur following frontal brain damage(Stuss et al 1986 Stuss 1991 Carver et al 1991) Our findings showed asignificant effect for the right intention in a subject with left frontal braindamage Moreover this effect was in the direction of intention In contrast therewere no significant effects in the other groups (ie bilateral frontal right frontalpooled frontal or normal subjects)

Although the result in the left frontal patient on the right intention wasstatistically significant even after correction for multiple comparisons thisfinding was interpreted with great caution First it was based on a relativelysmall number of intention trials (n ˆ 3000) and was derived from only onesubject Second we found a p-value for an individual lsquolsquosubjectrsquorsquo in the pseu-dodata that would also have met criteria for significance using a Bonferronicorrection and that was even less than the p-value of 00015 obtained fromthe left frontal subject Although the effect of this lsquolsquopseudopatientrsquorsquo was not inthe direction of intention the fact that it was significant raised caution forinterpreting the effect in the left frontal patient However the replication of thefindings in the left frontal patient in a second well-designed study for each of the

Fig 4 (Continued)

Effects of Frontal Lobe Lesions on Intentionality 665

three intentions right left and baseline suggests that the effect in this subjectmay be more than chance occurrence

Additional support for a reliable effect in the left frontal subject comes fromfurther analysis of his data suggesting about a 94 chance of his showinga similar finding if another series of 19 pairs of blocks of 1000 trials were runagain Furthermore the profile of REG output data for the right intentionshowed a fairly consistent separation whereas this was not the case for the left orbaseline intentions

A comment is warranted about the REG output for the left frontal patientbeing lower for the control condition on the right intention or more to the leftas compared to the control conditions for the left or baseline intentions Onemight argue that the significant effect on the right intention was an artifact ofcomparison to control data that was in the left direction and that this widened thedifference between the intention and control data However this argument is nottenable because the effect on the right intention was significant even when theREG output was compared to a theoretical mean of 100mdasha value which is ap-proximately equal to the control means for the left intention (9999985) and thebaseline intention (1000011)

The patientrsquos cognitive deficits and brain lesion have been describedelsewhere (Marras et al 1998) He suffered from a tension pneumocephaluswhich resulted in cognitive deficits and epileptic seizures The cognitive deficitsinclude decreased mental flexibility on the Trail-Making Test poor attentionreduced fluency and impaired spatial planning and visuospatial problem solvingMRI showed an extensive left frontal lesion but the right frontal lobe was intactPsychometric testing and SPECT suggested the addition of right frontal dysfunc-tion The SPECT findings provide a measure of function as opposed to structureand were subtle

As indicated above frontal lesions have been associated with reduced self-awareness (Stuss et al 1986 Stuss 1991 Carver et al 1991) a state that isdifficult for normal individuals to achieve The rationale for studying patientswith damage to the frontal lobes was that decreased self-awareness might facili-tate the effects of intentionality on random physical phenomena Brain regionsthat mediate neurological processes underlying self-awareness include thefrontal lobes bilaterally particularly on the right (Stuss et al 2001a Stuss ampAlexander 2000a Stuss amp Alexander 2000b Stuss et al 2001b) Whereas thepatientrsquos right-sided brain dysfunction may have been insufficient to producea deficit in self-awareness the extensive lesion on the left may have resulted inreduced self-awareness when attention was directed towards the right Howeverit remains unclear why positive results should be found only following damageto the left frontal region and not after bilateral or right frontal lesions Onespeculation is that the effect on random physical events may require reducedself-awareness combined with relatively intact attentional mechanisms Theassociation of frontal lobe lesions especially on the right with impaired at-tention (Stuss amp Levine 2002) may explain the negative findings in the setting

666 M Freedman et al

of bilateral or right frontal damage Whether deficits related to left frontalabnormalities can explain the observed effects on the REG output warrantsfurther study Moreover the question whether normal processes associated withintact frontal lobe function together with preserved self-awareness may serve toinhibit effects of intentionality on random physical phenomena needs to beaddressed

The strength of our conclusions rests largely on a well-designed methodol-ogy and replication of our findings Although our results did not replicate thefindings reported by Jahn and his colleagues in normal subjects (Jahn et al1987a 1997) they support their claims that intentionality can alter the output ofa random event generator Furthermore our findings suggest that patients withfrontal lobe lesions may serve as a good model for future studies of the effects ofconsciousness on random physical events

Acknowledgments

We gratefully acknowledge Dr Donald Stuss for his helpful advice andsupport throughout the study and for reviewing the manuscript and for hishelpful suggestions Dr Louis Siminovitch for helpful advice and supportPatricia Ebert and Dr Donald Stuss for assisting with analysis of lesion local-ization for the CT and MRI scans Janice Pogue for statistical advice LeeFerreira for testing the subjects John Sloan for assistance with data collec-tion Tania Xerri for assistance with data analyses and Tania Xerri MaedeanBrown Wanna Mar and Melissa Ziraldo for preparation of materials for pub-lication We also gratefully acknowledge Professor Robert Jahn BrendaDunne Dr Roger Nelson Dr York Dobyns and John Bradish at the PEARlab for their helpful advice and for providing the portable REG and softwarefor the study This study was supported by the Fetzer Institute Dr Freedmanwas supported by the Saul A Silverman Family Foundation Toronto Canadaas a Canada-International Scientific exchange program (CISEPO) project

References

Alcock J E (1987) ParapsychologyScience of the Anomalous or Search for the Soul Behavioraland Brain Sciences 10 553ndash643

Carver C S amp Scheier M F (1991) Self-Regulation and the Self In J Strauss amp G R Goethals(Eds) The Self Interdisciplinary Approaches (pp 168ndash207) New York Springer-Verlag

Hall J Kim C McElroy B amp Shimony A (1977) Wave-packet reduction as a medium ofcommunication Foundations of Physics 7 759ndash767

Ibison M amp Jeffers S (1998) A Double-Slit Diffraction Experiment to Investigate Claims ofConsciousness-Related Anomalies Journal of Scienti c Exploration 12 543ndash550

Jahn R G amp Dunne B J (1986) On the Quantum Mechanics of Consciousness with Applicationto Anomalous Phenomena Foundations of Physics 16 721ndash772

Jahn R G amp Dunne B J (1987a) Margins of Reality The Role of Consciousness in the PhysicalWorld San Diego Harcourt Brace Jovanovich

Jahn R G Dunne B J amp Nelson R D (1987b) Engineering Anomalies Research Journal ofScienti c Exploration 1 21ndash50

Effects of Frontal Lobe Lesions on Intentionality 667

achievable by normal individualsStudying patientswith frontal lobe lesions mayprovide a good model for testing the hypotheses from the PEAR lab within thecontext of a conceptual framework that would maximise the likelihood ofdetecting effects if these in fact exist We report our findings in patients withfrontal lobe lesions and in normal subjects as well as replication data in onesubject with left frontal brain damage

Methods

Subjects

Subjects with frontal lobe brain damage (n ˆ 6) and normal subjects withoutbrain lesions (n ˆ 6) were studied The frontal group consisted of subjects withlesions in the following brain regions bilateral frontal lobes (n ˆ 4) left frontallobe (n ˆ 1) and right frontal lobe (n ˆ 1) Table 1 shows the age and gender ofthe subjects and the etiology of the brain lesions in the patients All subjects withfrontal brain damage except S1 had a CT or MRI scan of the head document-ing the site of the lesion The CT and MRI scans were interpreted blind to thehypotheses by observers with experience in interpreting neuroradiological scansThe patient without a scan had a bilateral frontal leucotomy many years earlierThe normal subjects were comprised of four research staff at one of the studysites (two of whom are authors) and two relatives of the research staff

Experimental Procedure

Subjects were asked to sit in front of a computer monitor showing an imageof a horizontal arrow or bar The midline of the screen was indicated by a lineThe subjects were instructed to concentrate on moving the image to the right(intention right) or left (intention left) of the midline or not to concentrate on

TABLE 1Subject Profiles

Subject Age Gender Etiology

S1 Bilateral frontal 70 F Frontal leucotomyS2 Bilateral frontal 60 F Subarachnoid hemorrhageS3 Bilateral frontal 58 M Subarachnoid hemorrhageS4 Bilateral frontal 61 M Frontal leucotomyS5 Left frontal 45 M Tension pneumocephalusS6 Right frontal 70 F InfarctS7 Normal 25 F NAa

S8 Normal 43 M NAS9 Normal 61 M NA

S10 Normal 69 M NAS11 Normal 51 M NAS12 Normal 26 F NA

a NA ˆ not applicable

Effects of Frontal Lobe Lesions on Intentionality 653

moving the image (baseline intention) For the right and left intentions theimage was an arrow for the baseline intention it was a bar (Figure 1 andashc) Thepurpose of selecting an arrow or bar was to provide an ongoing cue to helpsubjects maintain their attention on the specified intent

The order of intentions (right left or baseline) followed a predeterminedsequence as shown below (ie R L B L B ) Subjects who completed more thanthe following 12 intentions repeated the same cycle until they finished the study

R L BL B RR B LL R B

All six patients and two of the six normal subjects were tested with anexaminer in the room The four normal subjects who were research staff (S7 S8S11 S12) were tested without an examiner present The latter is in keeping withthe protocol in the PEAR program where laboratory staff maintain a minimalpresence during the experiments (Jahn et al 1987a) Each of the six patientsand the two normal subjects tested by an examiner were initially seated facingthe examiner and were given the following instructions lsquolsquoThere are some peoplewho believe that if we concentrate on something hard enough we can affect howthings happen Now we donrsquot know if this is true but we have undertaken to testthis out We would like to see if there is a possibility that people can influencesomething just by concentrating on itrsquorsquo Subjects were then instructed to face thecomputer screen which displayed an arrow pointing in the right or in the leftdirection with the tip at the midline The experimenter then continued with theinstructions lsquolsquoNow on this screen there is an arrow What I would like you to dois concentrate on making the arrow move in the direction that it is pointingI want to see how your concentration can affect the position of the arrow Thearrow will sometimes be on the right and sometimes on the left of the screen butI want you to keep the arrow on the leftright side as much as possible Do youhave any questions Remember I want you to try to keep the arrow on the leftright side as much as possible Irsquod like you to begin nowrsquorsquo

Fig 1 Illustration of computer screen showing initial position of arrow or bar for each intention(a) Intention Right (b) Intention Left and (c) Intention Baseline

654 M Freedman et al

The examiner then started the computer program which was designed to movethe arrow according to the output of a Random Event Generator (REG) The REGproduces a random series of 0rsquos and 1rsquos based upon a sampling of an electronicnoise pattern at pre-set regular intervals The sign of the signal at the time ofsampling (ie positive or negative) is compared with a regularly alternatingsequence of positive and negative pulses If there is a match (ie negative-negative or positive-positive) a lsquolsquo1rsquorsquo is generated If there is no match then a lsquolsquo0rsquorsquois generated (Jahn et al 1987a) Sampling occurred at a rate of 200 per secondThe data were summed as 200-sample bits with expected numerical value of 100due to a 50 chance of a positive-positive or negative-negativematch occurringat each sampling Each sample of 200 bits of data comprised a trial

The position of the arrow relative to the midline of the screen movedrightward or leftward according to the cumulative mean of the trials with themidline representing a cumulative mean of 100 An arrow with the tip to theright of midline represented a mean of greater than 100 and an arrow with the tipto the left corresponded to a mean of less than 100 The REG used for the ex-periment was a portable model of the larger device that has been used in thePEAR program at Princeton University (Jahn et al 1987a Nelson et al 1984Nelson et al 1989 Jahn et al 1987b Jahn et al 1997) The portable REG wasobtained from the PEAR program

Each intention (right left or baseline) consisted of 10 blocks of 100 trialsand lasted approximately 15 minutes At the end of each block of 100 trialsthe examiner confirmed that the subject understood the task answered anyquestions and initiated the next block of 100 trials After each block of 100trials the position of the arrow tip or bar was reset to the midline

After each intention for right left or baseline a control run of 1000 trials wascarried out without anyone in the room This completed one full session of 1000trials for the relevant intention and 1000 control trials Participation was spreadacross several sessions The total number of trials varied across subjects due todifferences in the time that they could devote to the study

Results

Primary analyses were carried out to test whether there were significanteffects of intention in the frontal subjects (bilateral frontal left frontal rightfrontal and frontal subjects pooled) and in the normal group Secondary analyseswere carried out to determine whether there were effects for individual subjects

Primary Analyses

Figure 2 (andashc) shows the mean output of the REG for each intention (rightleft baseline) and the mean control output for the patients with frontal lesionsand the normal subjects Table 2 shows the corresponding number of intentionand control trials

Effects of Frontal Lobe Lesions on Intentionality 655

T-tests were carried out using SAS System for Windows (The SAS Systemfor Windows Release 612 1996) to determine whether there were signifi-cant differences between intention and control conditions for the right left andbaseline intentions Separate analyses were carried out for patients with lesionsin the frontal lobes bilaterally left frontal lobe right frontal lobe all frontalpatients pooled and for the normal subjects This resulted in a total of 15analyses Using a Bonferroni correction for 15 multiple comparisons a p-value 00515 or 0003 would be required for statistical significance As shown in

Fig 2 Mean REG output for patients with brain lesions in bilateral frontal (BF) left frontal (LF)and right frontal (RF) regions frontal patients pooled (FR) and normal subjects (NORM)(a) Intention Right (b) Intention Left and (c) Intention Baseline

656 M Freedman et al

Table 3 there were no significant differences between the intention and controlconditions for the right left or baseline intentions in the bilateral frontal sub-jects right frontal subject or normal subjects In addition there were no signifi-cant differences between the intention and control conditions for the pooledgroup of frontal subjects

For the left frontal patient there was a significant difference between theintention and control conditions for the right intention (p ˆ 00015) The effectwas in the direction of intention and was significant even after Bonferroni cor-rection for multiple comparisons There were no significant differences betweenthe intention and control conditions for the left or baseline intentions in the leftfrontal patient

T-tests were carried out to determine whether there were significantdifferences between the intention and control conditions for the right left orbaseline intentions for the entire sample of frontal and normal subjects pooledThere were no significant differences for any intention (right t[304898] ˆ00480 p ˆ 096 left t[302998] ˆ 06709 p ˆ 050 baseline t[299998] ˆiexcl17053 p ˆ 009)

Secondary Analyses

Secondary analyses were carried out using t-tests to determine whether therewere significant differences between intention and control conditions for eachsubject Table 4 shows the results of the analysis for each subject Significantdifferences at the p ˆ 005 level were found between the intention and controlconditions in three instances intention right (p ˆ 00015) (S5 left frontalpatient) baseline (p ˆ 003) (S8 normal subject) intention right (p ˆ 0004) (S9

Fig 2 (Continued)

Effects of Frontal Lobe Lesions on Intentionality 657

normal subject) The effect was in the intended direction for S5 and S9 Usinga Bonferroni correction for 36 multiple comparisons a p-value 00536 or00014 would be required for statistical significance None of these p-valuesexceeded this threshold

As stated in the methodology section four normal subjects were research staff(S7 S8 S11 S12) and were tested without an examiner in the room In additionsubjects S9 and S10 were relatives of the research staff but were tested with anexaminer present These six subjects were divided into two groups for separateanalysis One group was comprised of the normal subjects who were researchstaff (S7 S8 S11 S12) and the other group was comprised of subjects S9 andS10 T-tests were carried out to determine whether there were significantdifferences between the intention and control conditions for the right left andbaseline intentions for these two groups This resulted in a total of six analysesUsing a Bonferroni correction for six multiple comparisons a p-value of 0056 or 0008 would be required for statistical significance There were no sig-nificant differences between the intention and control conditions for the groupcomprised of subjects S7 S8 S11 and S12 (right p ˆ 069 left p ˆ 061baseline p ˆ 024) For the group comprised of subjects S9 and S10 there wasa difference between the intention and control conditions for the right intention(p ˆ 00164) that was in the direction of intention However this was notsignificant after correction for multiple comparisons There were no significantdifferences between the intention and control conditions for the left (p ˆ 014) orbaseline (p ˆ 048) intentions for the group comprised of S9 and S10

In addition to statistical control for multiple comparisons using a Bonferronicorrection we performed an additional study to experimentally control for thenumerous statistical tests lsquolsquoPseudodatarsquorsquo were created by repeating the entire ex-perimental procedure except that there was no subject or experimenter in theroom during the intention condition Two sets of pseudodata were generated thefirst at the same site where the original data were collected (pseudodata 1) andthe second in a different building (pseudodata 2) The data were labeled to cor-respond to the intentions of the subjects as in the real study

As shown in Tables 5a and 5b the number of statistically significant resultsranged from one (pseudodata 1) to four (pseudodata 2) Moreover one of the

TABLE 2Total Number of Trials (Includes Intention Plus Control Trials)

Subjects Right Left Baseline

Bilateral frontals 200900a 201000 200000Left frontal 6000 8000 6000Right frontal 2000 2000 2000Frontals pooled 208900a 211000 208000Normals 96000 92000 92000

a There were 100 fewer intention trials than control trials for the right intention in the bilateralfrontals

658 M Freedman et al

significance levels was greater in the pseudodata than in the real data Howevernone of the effects were in the direction of intention and in three instances thesignificant effects occurred during the baseline intention

Replication Study

The primary analyses showed a statistically significant effect in the direc-tion of intention to the right for the patient with a left frontal lesion This effectwas significant even after Bonferroni correction for multiple comparisons Toexamine the possible effects in this patient further we carried out a second studyto determine whether the findings could be replicated The planned hypothesiswas that he would show a statistically significant effect of intention to the rightbut not to the left or in the baseline intention (ie that the findings would be thesame as before) The left frontal subject was retested using the methodologydescribed for the original study

During testing a printer error (data was always printed after each block of 100trials) occurred at the end of the fifth block of 100 trials during a baseline sessionand the program exited while the bar was still on the screen Although thecomplete data for these 100 trials were still captured a replacement set of 1000baseline and control trials for the entire intention was collected on a separateday However the possible impact of the printer error is considered negligibledue to the small number of trials involved the fact that none of the data werelost and because it occurred during the baseline condition in which the subjectwas asked not to concentrate on moving the arrow On another occasiona baseline session should have preceded a right intention session but the order was

TABLE 3Analysis of Frontal and Normal Subjects (Intention vs Control Conditions)

Subjects Intention t df p-value

Bilateral frontals right 11920 200898 023left 01213 200998 090baseline iexcl13874 199998 017

Left frontal right iexcl31691 5998 00015R

left 00524 7998 096baseline iexcl03656 5998 071

Right frontal right 00375 1998 097left iexcl08754 1998 038baseline iexcl16084 1998 011

Frontals pooled right 06377 208898 052left 00441 210998 096baseline iexcl15794 207998 011

Normals right iexcl08538 95998 039left 11474 91998 025baseline iexcl07036 91998 048

Note R ˆ Direction of effect was to the right

Effects of Frontal Lobe Lesions on Intentionality 659

reversed This resulted in two right intention and two baseline sessions being runconsecutively The right intention sessions were on the same day whereas thebaseline sessions were on different days Another occurrence consisted of thesubject seeing double arrows during a right intention session Finally during a left

TABLE 4Single Subject Analysis

Mean REG Output

Subject Intention Intention Control t p-value Trials (n)

S1 right 9998 10001 04982 062 40000left 9997 10003 08491 040 40000baseline 10003 10004 02000 084 40000

S2 right 9999 10007 08302 041 22900a

left 9988 9991 03145 075 24000baseline 10003 10001 iexcl01708 086 24000

S3 right 9989 10002 17414 008 34000left 10011 10000 iexcl13408 018 33000baseline 10013 10004 iexcl10462 030 32000

S4 right 10002 10002 iexcl00341 097 104000left 9999 10000 02488 080 104000baseline 10003 9997 iexcl13896 016 104000

S5 right 10019 9962 iexcl31691 00015R 6000left 9999 10000 00524 096 8000baseline 10001 9994 iexcl03656 071 6000

S6 right 10007 10008 00375 097 2000left 9991 9964 iexcl08754 038 2000baseline 10009 9959 iexcl16084 011 2000

S7 right 9984 9991 06708 050 16000left 10002 9998 iexcl03481 073 16000baseline 9994 9994 iexcl00022 100 16000

S8 right 10000 10006 07071 048 24000left 10003 10002 iexcl01489 088 24000baseline 10009 9989 iexcl21274 00334R 24000

S9 right 10021 9991 iexcl28790 00040R 18000left 10000 10012 10737 028 16000baseline 9994 9999 04375 066 16000b

S10 right 10001 10004 01845 085 6000left 9988 10014 11721 024 4000baseline 9974 9990 07088 048 4000

S11 right 10002 10002 iexcl00477 096 16000left 9999 10005 06132 054 16000baseline 10005 10004 iexcl01077 091 16000

S12 right 10007 10000 iexcl06514 051 16000left 9995 10007 10216 031 16000baseline 10000 10003 02445 081 16000

Note n ˆ number of intention trials Dagger control trials R ˆ direction of effect was to the righta There were 100 fewer intention trials than control trialsb 500 baseline control trials were collected incorrectly and replaced by trials collected at a different

time Although this introduced a potential bias the difference between the intention and controlconditions was too far from significant for a small number of 500 trials to have had an impact

660 M Freedman et al

intention session the building intercom came on at the end of the fourth block of100 trials The subject stated that this affected his concentration only slightly Inall cases described above the complete set of data was used for analysesincluding the replacement trials that followed the printer error (ie no data werediscarded)

TABLE 5aPseudodata Analysis 1

Mean REG Output

Subject Intention Intention Control t p-value Trials (n)

PS1 right 9997 10005 11636 024 40000left 10011 10003 iexcl10648 029 40000baseline 10007 9995 iexcl17586 008 40000

PS2 right 9995 10010 15873 011 22900left 9998 10004 06576 051 24000baseline 9994 10006 14214 016 24000

PS3 right 9996 10002 08405 040 34000left 9991 10002 13635 017 32000a

baseline 10007 9999 iexcl09442 035 32000PS4 right 10003 9999 iexcl10667 029 104000

left 10006 9999 iexcl17887 007 104000baseline 10001 9999 iexcl02984 077 104000

PS5 right 9982 9993 06322 053 6000left 10020 10000 iexcl12473 021 8000baseline 9988 9993 02343 081 6000

PS6 right 9999 10031 10198 031 2000left 9979 10006 08638 039 2000baseline 9963 10031 21474 00319L 2000

PS7 right 10002 10005 02463 081 16000left 9999 10013 12288 022 16000baseline 10005 9999 iexcl05196 060 16000

PS8 right 10006 9996 iexcl11546 025 24000left 10005 10007 02370 081 24000baseline 10000 9997 iexcl02564 080 24000

PS9 right 9995 10011 15199 013 18000left 9996 10002 05973 055 16000baseline 10005 9995 iexcl08764 038 16000

PS10 right 10007 9993 iexcl07168 047 6000left 9997 10016 08457 040 4000baseline 9981 9987 02792 078 4000

PS11 right 10005 10014 08604 039 16000left 10001 10006 04769 063 16000baseline 10009 9997 iexcl10813 028 16000

PS12 right 9993 10010 15528 012 16000left 10001 10012 09288 035 16000baseline 9998 10013 13483 018 16000

Note n ˆ number of intention trials Dagger control trials L ˆ direction of effect was to the lefta Pseudodata contained 500 fewer intention and control trials as compared to real data

Effects of Frontal Lobe Lesions on Intentionality 661

Results of Replication Study

Figure 3 shows the mean REG output for each intention (right left baseline)and the control output for the left frontal subject There were 19000 intentionand control trials for the right intention and 20000 trials for each of the inten-tion and control conditions for the left and baseline intentions

TABLE 5bPseudodata Analysis 2

Mean REG Output

Subject Intention Intention Control t p-value Trials (n)

PS1 right 9995 10001 08327 041 40000left 9996 9995 iexcl02329 082 40000baseline 10003 10000 iexcl04994 062 40000

PS2 right 10001 9995 iexcl06746 050 22900left 10000 9991 iexcl09721 033 24000baseline 9996 9999 03060 076 24000

PS3 right 10011 10008 iexcl04347 066 34000left 9998 10001 03522 072 32000a

baseline 10005 10000 iexcl06143 054 32000PS4 right 9994 10005 26336 00084L 104000

left 10001 9995 iexcl13071 019 104000baseline 9998 9997 iexcl02664 079 104000

PS5 right 9986 10005 10582 029 6000left 10003 10005 01187 091 8000baseline 10002 9995 iexcl03727 071 6000

PS6 right 9968 9993 07972 043 2000left 9990 9976 iexcl04402 066 2000baseline 10008 9960 iexcl15583 012 2000

PS7 right 9990 9995 04285 067 16000left 10018 10002 iexcl14678 014 16000baseline 9983 9996 11654 024 16000

PS8 right 10008 10005 iexcl02559 080 24000left 10016 9984 iexcl35151 00004R 24000baseline 10002 10002 iexcl00648 095 24000

PS9 right 10001 9998 iexcl02441 081 18000left 10017 9998 iexcl16610 010 16000baseline 9993 9983 iexcl08404 040 16000

PS10 right 10009 10014 02525 080 6000left 10018 10000 iexcl07789 044 4000baseline 9987 10022 15658 012 4000

PS11 right 10006 10010 03433 073 16000left 9992 10000 07129 048 16000baseline 10007 9986 iexcl19258 00541R 16000

PS12 right 9997 9986 iexcl10275 030 16000left 10002 9999 iexcl02967 077 16000baseline 9978 10001 20961 00361L 16000

Note n ˆ number of intention trials Dagger control trials R ˆ direction of effect was to the right L ˆdirection of effect was to the left

a Pseudodata contained 500 fewer intention and control trials as compared to real data

662 M Freedman et al

T-tests were carried out using SAS System for Windows Release 8 todetermine whether there were significant differences between the intention andcontrol conditions for the right left and baseline intentions There wasa significant difference between the intention and control conditions for the rightintention (t[37998]ˆiexcl253 pˆ 00115) but not for the other two intentions (leftt[39998] ˆ 049 p ˆ 06269 and baseline t[39998] ˆ iexcl007 p ˆ 09418) Thesignificant effect for the right intention was in the direction of intention as in theoriginal study

Since the investigation was run in blocks of 1000 trials we examined the datausing blocks as the unit of measurement and the averaged output across trials asthe measure in each block An F-ratio test with condition (intention vs control)as the source of variance of interest and the interaction between condition andblock as the error term showed an effect of condition (p ˆ 00578) This suggeststhat the left frontal subject would show comparable findings on another series of19 pairs of blocks of 1000 trials with 94 confidence

Whereas the mean REG output for the right intention was above the expectedvalue of 100 the output for the control condition was below 100 (Figure 3) Toexamine the possibility that the significant difference between the intention andcontrol condition was due solely to low output during the control condition asopposed to high output during the intention condition we tested whether theoutput for the intention condition was significantly different from a constantvalue of 100 This value is the theoretical mean output of the REG assumingtruly random and unbiased output The control values for the left and baselineconditions are approximately equal to this theoretical mean The REG output forthe right intention was significantly different than 100 and in the direction ofintention (t ˆ 201 p ˆ 0045 two-tailed test) For the control condition themean REG output was less than (although not significantly different from) 100(t ˆ iexcl156 p ˆ 012)

Fig 3 Mean REG output for the left frontal patient in replication study

Effects of Frontal Lobe Lesions on Intentionality 663

Figure 4 (andashc) shows the mean output of the REG broken down byconsecutive 1000 trial sessions for each intention (right left and baseline) andfor the control output For the right intention Figure 4 shows that on most pairsof blocks of 1000 trials the mean REG output follows a fairly consistent patternin which the means are either higher compared to the control condition or aboutthe same In contrast the data for each of the left and baseline conditions showa pattern with less separation between conditions

Fig 4 Mean REG output for each 1000 trial session in replication study (a) Intention Right (b)Intention Left and (c) Intention Baseline

664 M Freedman et al

Discussion

We examined the claims from the PEAR program that an individualrsquosintentions can influence the statistical distribution of random physical phenom-ena Our main focus was to test these claims using well-designed control condi-tions in a population that might maximize the likelihood of detecting any effectsthat may exist Our choice of studying patients with frontal lobe lesions wasbased on the concept that the potential to influence random physical events maybe optimized when attention is diverted from self-awareness (ie such as whenself-awareness is reduced) This state can occur following frontal brain damage(Stuss et al 1986 Stuss 1991 Carver et al 1991) Our findings showed asignificant effect for the right intention in a subject with left frontal braindamage Moreover this effect was in the direction of intention In contrast therewere no significant effects in the other groups (ie bilateral frontal right frontalpooled frontal or normal subjects)

Although the result in the left frontal patient on the right intention wasstatistically significant even after correction for multiple comparisons thisfinding was interpreted with great caution First it was based on a relativelysmall number of intention trials (n ˆ 3000) and was derived from only onesubject Second we found a p-value for an individual lsquolsquosubjectrsquorsquo in the pseu-dodata that would also have met criteria for significance using a Bonferronicorrection and that was even less than the p-value of 00015 obtained fromthe left frontal subject Although the effect of this lsquolsquopseudopatientrsquorsquo was not inthe direction of intention the fact that it was significant raised caution forinterpreting the effect in the left frontal patient However the replication of thefindings in the left frontal patient in a second well-designed study for each of the

Fig 4 (Continued)

Effects of Frontal Lobe Lesions on Intentionality 665

three intentions right left and baseline suggests that the effect in this subjectmay be more than chance occurrence

Additional support for a reliable effect in the left frontal subject comes fromfurther analysis of his data suggesting about a 94 chance of his showinga similar finding if another series of 19 pairs of blocks of 1000 trials were runagain Furthermore the profile of REG output data for the right intentionshowed a fairly consistent separation whereas this was not the case for the left orbaseline intentions

A comment is warranted about the REG output for the left frontal patientbeing lower for the control condition on the right intention or more to the leftas compared to the control conditions for the left or baseline intentions Onemight argue that the significant effect on the right intention was an artifact ofcomparison to control data that was in the left direction and that this widened thedifference between the intention and control data However this argument is nottenable because the effect on the right intention was significant even when theREG output was compared to a theoretical mean of 100mdasha value which is ap-proximately equal to the control means for the left intention (9999985) and thebaseline intention (1000011)

The patientrsquos cognitive deficits and brain lesion have been describedelsewhere (Marras et al 1998) He suffered from a tension pneumocephaluswhich resulted in cognitive deficits and epileptic seizures The cognitive deficitsinclude decreased mental flexibility on the Trail-Making Test poor attentionreduced fluency and impaired spatial planning and visuospatial problem solvingMRI showed an extensive left frontal lesion but the right frontal lobe was intactPsychometric testing and SPECT suggested the addition of right frontal dysfunc-tion The SPECT findings provide a measure of function as opposed to structureand were subtle

As indicated above frontal lesions have been associated with reduced self-awareness (Stuss et al 1986 Stuss 1991 Carver et al 1991) a state that isdifficult for normal individuals to achieve The rationale for studying patientswith damage to the frontal lobes was that decreased self-awareness might facili-tate the effects of intentionality on random physical phenomena Brain regionsthat mediate neurological processes underlying self-awareness include thefrontal lobes bilaterally particularly on the right (Stuss et al 2001a Stuss ampAlexander 2000a Stuss amp Alexander 2000b Stuss et al 2001b) Whereas thepatientrsquos right-sided brain dysfunction may have been insufficient to producea deficit in self-awareness the extensive lesion on the left may have resulted inreduced self-awareness when attention was directed towards the right Howeverit remains unclear why positive results should be found only following damageto the left frontal region and not after bilateral or right frontal lesions Onespeculation is that the effect on random physical events may require reducedself-awareness combined with relatively intact attentional mechanisms Theassociation of frontal lobe lesions especially on the right with impaired at-tention (Stuss amp Levine 2002) may explain the negative findings in the setting

666 M Freedman et al

of bilateral or right frontal damage Whether deficits related to left frontalabnormalities can explain the observed effects on the REG output warrantsfurther study Moreover the question whether normal processes associated withintact frontal lobe function together with preserved self-awareness may serve toinhibit effects of intentionality on random physical phenomena needs to beaddressed

The strength of our conclusions rests largely on a well-designed methodol-ogy and replication of our findings Although our results did not replicate thefindings reported by Jahn and his colleagues in normal subjects (Jahn et al1987a 1997) they support their claims that intentionality can alter the output ofa random event generator Furthermore our findings suggest that patients withfrontal lobe lesions may serve as a good model for future studies of the effects ofconsciousness on random physical events

Acknowledgments

We gratefully acknowledge Dr Donald Stuss for his helpful advice andsupport throughout the study and for reviewing the manuscript and for hishelpful suggestions Dr Louis Siminovitch for helpful advice and supportPatricia Ebert and Dr Donald Stuss for assisting with analysis of lesion local-ization for the CT and MRI scans Janice Pogue for statistical advice LeeFerreira for testing the subjects John Sloan for assistance with data collec-tion Tania Xerri for assistance with data analyses and Tania Xerri MaedeanBrown Wanna Mar and Melissa Ziraldo for preparation of materials for pub-lication We also gratefully acknowledge Professor Robert Jahn BrendaDunne Dr Roger Nelson Dr York Dobyns and John Bradish at the PEARlab for their helpful advice and for providing the portable REG and softwarefor the study This study was supported by the Fetzer Institute Dr Freedmanwas supported by the Saul A Silverman Family Foundation Toronto Canadaas a Canada-International Scientific exchange program (CISEPO) project

References

Alcock J E (1987) ParapsychologyScience of the Anomalous or Search for the Soul Behavioraland Brain Sciences 10 553ndash643

Carver C S amp Scheier M F (1991) Self-Regulation and the Self In J Strauss amp G R Goethals(Eds) The Self Interdisciplinary Approaches (pp 168ndash207) New York Springer-Verlag

Hall J Kim C McElroy B amp Shimony A (1977) Wave-packet reduction as a medium ofcommunication Foundations of Physics 7 759ndash767

Ibison M amp Jeffers S (1998) A Double-Slit Diffraction Experiment to Investigate Claims ofConsciousness-Related Anomalies Journal of Scienti c Exploration 12 543ndash550

Jahn R G amp Dunne B J (1986) On the Quantum Mechanics of Consciousness with Applicationto Anomalous Phenomena Foundations of Physics 16 721ndash772

Jahn R G amp Dunne B J (1987a) Margins of Reality The Role of Consciousness in the PhysicalWorld San Diego Harcourt Brace Jovanovich

Jahn R G Dunne B J amp Nelson R D (1987b) Engineering Anomalies Research Journal ofScienti c Exploration 1 21ndash50

Effects of Frontal Lobe Lesions on Intentionality 667

moving the image (baseline intention) For the right and left intentions theimage was an arrow for the baseline intention it was a bar (Figure 1 andashc) Thepurpose of selecting an arrow or bar was to provide an ongoing cue to helpsubjects maintain their attention on the specified intent

The order of intentions (right left or baseline) followed a predeterminedsequence as shown below (ie R L B L B ) Subjects who completed more thanthe following 12 intentions repeated the same cycle until they finished the study

R L BL B RR B LL R B

All six patients and two of the six normal subjects were tested with anexaminer in the room The four normal subjects who were research staff (S7 S8S11 S12) were tested without an examiner present The latter is in keeping withthe protocol in the PEAR program where laboratory staff maintain a minimalpresence during the experiments (Jahn et al 1987a) Each of the six patientsand the two normal subjects tested by an examiner were initially seated facingthe examiner and were given the following instructions lsquolsquoThere are some peoplewho believe that if we concentrate on something hard enough we can affect howthings happen Now we donrsquot know if this is true but we have undertaken to testthis out We would like to see if there is a possibility that people can influencesomething just by concentrating on itrsquorsquo Subjects were then instructed to face thecomputer screen which displayed an arrow pointing in the right or in the leftdirection with the tip at the midline The experimenter then continued with theinstructions lsquolsquoNow on this screen there is an arrow What I would like you to dois concentrate on making the arrow move in the direction that it is pointingI want to see how your concentration can affect the position of the arrow Thearrow will sometimes be on the right and sometimes on the left of the screen butI want you to keep the arrow on the leftright side as much as possible Do youhave any questions Remember I want you to try to keep the arrow on the leftright side as much as possible Irsquod like you to begin nowrsquorsquo

Fig 1 Illustration of computer screen showing initial position of arrow or bar for each intention(a) Intention Right (b) Intention Left and (c) Intention Baseline

654 M Freedman et al

The examiner then started the computer program which was designed to movethe arrow according to the output of a Random Event Generator (REG) The REGproduces a random series of 0rsquos and 1rsquos based upon a sampling of an electronicnoise pattern at pre-set regular intervals The sign of the signal at the time ofsampling (ie positive or negative) is compared with a regularly alternatingsequence of positive and negative pulses If there is a match (ie negative-negative or positive-positive) a lsquolsquo1rsquorsquo is generated If there is no match then a lsquolsquo0rsquorsquois generated (Jahn et al 1987a) Sampling occurred at a rate of 200 per secondThe data were summed as 200-sample bits with expected numerical value of 100due to a 50 chance of a positive-positive or negative-negativematch occurringat each sampling Each sample of 200 bits of data comprised a trial

The position of the arrow relative to the midline of the screen movedrightward or leftward according to the cumulative mean of the trials with themidline representing a cumulative mean of 100 An arrow with the tip to theright of midline represented a mean of greater than 100 and an arrow with the tipto the left corresponded to a mean of less than 100 The REG used for the ex-periment was a portable model of the larger device that has been used in thePEAR program at Princeton University (Jahn et al 1987a Nelson et al 1984Nelson et al 1989 Jahn et al 1987b Jahn et al 1997) The portable REG wasobtained from the PEAR program

Each intention (right left or baseline) consisted of 10 blocks of 100 trialsand lasted approximately 15 minutes At the end of each block of 100 trialsthe examiner confirmed that the subject understood the task answered anyquestions and initiated the next block of 100 trials After each block of 100trials the position of the arrow tip or bar was reset to the midline

After each intention for right left or baseline a control run of 1000 trials wascarried out without anyone in the room This completed one full session of 1000trials for the relevant intention and 1000 control trials Participation was spreadacross several sessions The total number of trials varied across subjects due todifferences in the time that they could devote to the study

Results

Primary analyses were carried out to test whether there were significanteffects of intention in the frontal subjects (bilateral frontal left frontal rightfrontal and frontal subjects pooled) and in the normal group Secondary analyseswere carried out to determine whether there were effects for individual subjects

Primary Analyses

Figure 2 (andashc) shows the mean output of the REG for each intention (rightleft baseline) and the mean control output for the patients with frontal lesionsand the normal subjects Table 2 shows the corresponding number of intentionand control trials

Effects of Frontal Lobe Lesions on Intentionality 655

T-tests were carried out using SAS System for Windows (The SAS Systemfor Windows Release 612 1996) to determine whether there were signifi-cant differences between intention and control conditions for the right left andbaseline intentions Separate analyses were carried out for patients with lesionsin the frontal lobes bilaterally left frontal lobe right frontal lobe all frontalpatients pooled and for the normal subjects This resulted in a total of 15analyses Using a Bonferroni correction for 15 multiple comparisons a p-value 00515 or 0003 would be required for statistical significance As shown in

Fig 2 Mean REG output for patients with brain lesions in bilateral frontal (BF) left frontal (LF)and right frontal (RF) regions frontal patients pooled (FR) and normal subjects (NORM)(a) Intention Right (b) Intention Left and (c) Intention Baseline

656 M Freedman et al

Table 3 there were no significant differences between the intention and controlconditions for the right left or baseline intentions in the bilateral frontal sub-jects right frontal subject or normal subjects In addition there were no signifi-cant differences between the intention and control conditions for the pooledgroup of frontal subjects

For the left frontal patient there was a significant difference between theintention and control conditions for the right intention (p ˆ 00015) The effectwas in the direction of intention and was significant even after Bonferroni cor-rection for multiple comparisons There were no significant differences betweenthe intention and control conditions for the left or baseline intentions in the leftfrontal patient

T-tests were carried out to determine whether there were significantdifferences between the intention and control conditions for the right left orbaseline intentions for the entire sample of frontal and normal subjects pooledThere were no significant differences for any intention (right t[304898] ˆ00480 p ˆ 096 left t[302998] ˆ 06709 p ˆ 050 baseline t[299998] ˆiexcl17053 p ˆ 009)

Secondary Analyses

Secondary analyses were carried out using t-tests to determine whether therewere significant differences between intention and control conditions for eachsubject Table 4 shows the results of the analysis for each subject Significantdifferences at the p ˆ 005 level were found between the intention and controlconditions in three instances intention right (p ˆ 00015) (S5 left frontalpatient) baseline (p ˆ 003) (S8 normal subject) intention right (p ˆ 0004) (S9

Fig 2 (Continued)

Effects of Frontal Lobe Lesions on Intentionality 657

normal subject) The effect was in the intended direction for S5 and S9 Usinga Bonferroni correction for 36 multiple comparisons a p-value 00536 or00014 would be required for statistical significance None of these p-valuesexceeded this threshold

As stated in the methodology section four normal subjects were research staff(S7 S8 S11 S12) and were tested without an examiner in the room In additionsubjects S9 and S10 were relatives of the research staff but were tested with anexaminer present These six subjects were divided into two groups for separateanalysis One group was comprised of the normal subjects who were researchstaff (S7 S8 S11 S12) and the other group was comprised of subjects S9 andS10 T-tests were carried out to determine whether there were significantdifferences between the intention and control conditions for the right left andbaseline intentions for these two groups This resulted in a total of six analysesUsing a Bonferroni correction for six multiple comparisons a p-value of 0056 or 0008 would be required for statistical significance There were no sig-nificant differences between the intention and control conditions for the groupcomprised of subjects S7 S8 S11 and S12 (right p ˆ 069 left p ˆ 061baseline p ˆ 024) For the group comprised of subjects S9 and S10 there wasa difference between the intention and control conditions for the right intention(p ˆ 00164) that was in the direction of intention However this was notsignificant after correction for multiple comparisons There were no significantdifferences between the intention and control conditions for the left (p ˆ 014) orbaseline (p ˆ 048) intentions for the group comprised of S9 and S10

In addition to statistical control for multiple comparisons using a Bonferronicorrection we performed an additional study to experimentally control for thenumerous statistical tests lsquolsquoPseudodatarsquorsquo were created by repeating the entire ex-perimental procedure except that there was no subject or experimenter in theroom during the intention condition Two sets of pseudodata were generated thefirst at the same site where the original data were collected (pseudodata 1) andthe second in a different building (pseudodata 2) The data were labeled to cor-respond to the intentions of the subjects as in the real study

As shown in Tables 5a and 5b the number of statistically significant resultsranged from one (pseudodata 1) to four (pseudodata 2) Moreover one of the

TABLE 2Total Number of Trials (Includes Intention Plus Control Trials)

Subjects Right Left Baseline

Bilateral frontals 200900a 201000 200000Left frontal 6000 8000 6000Right frontal 2000 2000 2000Frontals pooled 208900a 211000 208000Normals 96000 92000 92000

a There were 100 fewer intention trials than control trials for the right intention in the bilateralfrontals

658 M Freedman et al

significance levels was greater in the pseudodata than in the real data Howevernone of the effects were in the direction of intention and in three instances thesignificant effects occurred during the baseline intention

Replication Study

The primary analyses showed a statistically significant effect in the direc-tion of intention to the right for the patient with a left frontal lesion This effectwas significant even after Bonferroni correction for multiple comparisons Toexamine the possible effects in this patient further we carried out a second studyto determine whether the findings could be replicated The planned hypothesiswas that he would show a statistically significant effect of intention to the rightbut not to the left or in the baseline intention (ie that the findings would be thesame as before) The left frontal subject was retested using the methodologydescribed for the original study

During testing a printer error (data was always printed after each block of 100trials) occurred at the end of the fifth block of 100 trials during a baseline sessionand the program exited while the bar was still on the screen Although thecomplete data for these 100 trials were still captured a replacement set of 1000baseline and control trials for the entire intention was collected on a separateday However the possible impact of the printer error is considered negligibledue to the small number of trials involved the fact that none of the data werelost and because it occurred during the baseline condition in which the subjectwas asked not to concentrate on moving the arrow On another occasiona baseline session should have preceded a right intention session but the order was

TABLE 3Analysis of Frontal and Normal Subjects (Intention vs Control Conditions)

Subjects Intention t df p-value

Bilateral frontals right 11920 200898 023left 01213 200998 090baseline iexcl13874 199998 017

Left frontal right iexcl31691 5998 00015R

left 00524 7998 096baseline iexcl03656 5998 071

Right frontal right 00375 1998 097left iexcl08754 1998 038baseline iexcl16084 1998 011

Frontals pooled right 06377 208898 052left 00441 210998 096baseline iexcl15794 207998 011

Normals right iexcl08538 95998 039left 11474 91998 025baseline iexcl07036 91998 048

Note R ˆ Direction of effect was to the right

Effects of Frontal Lobe Lesions on Intentionality 659

reversed This resulted in two right intention and two baseline sessions being runconsecutively The right intention sessions were on the same day whereas thebaseline sessions were on different days Another occurrence consisted of thesubject seeing double arrows during a right intention session Finally during a left

TABLE 4Single Subject Analysis

Mean REG Output

Subject Intention Intention Control t p-value Trials (n)

S1 right 9998 10001 04982 062 40000left 9997 10003 08491 040 40000baseline 10003 10004 02000 084 40000

S2 right 9999 10007 08302 041 22900a

left 9988 9991 03145 075 24000baseline 10003 10001 iexcl01708 086 24000

S3 right 9989 10002 17414 008 34000left 10011 10000 iexcl13408 018 33000baseline 10013 10004 iexcl10462 030 32000

S4 right 10002 10002 iexcl00341 097 104000left 9999 10000 02488 080 104000baseline 10003 9997 iexcl13896 016 104000

S5 right 10019 9962 iexcl31691 00015R 6000left 9999 10000 00524 096 8000baseline 10001 9994 iexcl03656 071 6000

S6 right 10007 10008 00375 097 2000left 9991 9964 iexcl08754 038 2000baseline 10009 9959 iexcl16084 011 2000

S7 right 9984 9991 06708 050 16000left 10002 9998 iexcl03481 073 16000baseline 9994 9994 iexcl00022 100 16000

S8 right 10000 10006 07071 048 24000left 10003 10002 iexcl01489 088 24000baseline 10009 9989 iexcl21274 00334R 24000

S9 right 10021 9991 iexcl28790 00040R 18000left 10000 10012 10737 028 16000baseline 9994 9999 04375 066 16000b

S10 right 10001 10004 01845 085 6000left 9988 10014 11721 024 4000baseline 9974 9990 07088 048 4000

S11 right 10002 10002 iexcl00477 096 16000left 9999 10005 06132 054 16000baseline 10005 10004 iexcl01077 091 16000

S12 right 10007 10000 iexcl06514 051 16000left 9995 10007 10216 031 16000baseline 10000 10003 02445 081 16000

Note n ˆ number of intention trials Dagger control trials R ˆ direction of effect was to the righta There were 100 fewer intention trials than control trialsb 500 baseline control trials were collected incorrectly and replaced by trials collected at a different

time Although this introduced a potential bias the difference between the intention and controlconditions was too far from significant for a small number of 500 trials to have had an impact

660 M Freedman et al

intention session the building intercom came on at the end of the fourth block of100 trials The subject stated that this affected his concentration only slightly Inall cases described above the complete set of data was used for analysesincluding the replacement trials that followed the printer error (ie no data werediscarded)

TABLE 5aPseudodata Analysis 1

Mean REG Output

Subject Intention Intention Control t p-value Trials (n)

PS1 right 9997 10005 11636 024 40000left 10011 10003 iexcl10648 029 40000baseline 10007 9995 iexcl17586 008 40000

PS2 right 9995 10010 15873 011 22900left 9998 10004 06576 051 24000baseline 9994 10006 14214 016 24000

PS3 right 9996 10002 08405 040 34000left 9991 10002 13635 017 32000a

baseline 10007 9999 iexcl09442 035 32000PS4 right 10003 9999 iexcl10667 029 104000

left 10006 9999 iexcl17887 007 104000baseline 10001 9999 iexcl02984 077 104000

PS5 right 9982 9993 06322 053 6000left 10020 10000 iexcl12473 021 8000baseline 9988 9993 02343 081 6000

PS6 right 9999 10031 10198 031 2000left 9979 10006 08638 039 2000baseline 9963 10031 21474 00319L 2000

PS7 right 10002 10005 02463 081 16000left 9999 10013 12288 022 16000baseline 10005 9999 iexcl05196 060 16000

PS8 right 10006 9996 iexcl11546 025 24000left 10005 10007 02370 081 24000baseline 10000 9997 iexcl02564 080 24000

PS9 right 9995 10011 15199 013 18000left 9996 10002 05973 055 16000baseline 10005 9995 iexcl08764 038 16000

PS10 right 10007 9993 iexcl07168 047 6000left 9997 10016 08457 040 4000baseline 9981 9987 02792 078 4000

PS11 right 10005 10014 08604 039 16000left 10001 10006 04769 063 16000baseline 10009 9997 iexcl10813 028 16000

PS12 right 9993 10010 15528 012 16000left 10001 10012 09288 035 16000baseline 9998 10013 13483 018 16000

Note n ˆ number of intention trials Dagger control trials L ˆ direction of effect was to the lefta Pseudodata contained 500 fewer intention and control trials as compared to real data

Effects of Frontal Lobe Lesions on Intentionality 661

Results of Replication Study

Figure 3 shows the mean REG output for each intention (right left baseline)and the control output for the left frontal subject There were 19000 intentionand control trials for the right intention and 20000 trials for each of the inten-tion and control conditions for the left and baseline intentions

TABLE 5bPseudodata Analysis 2

Mean REG Output

Subject Intention Intention Control t p-value Trials (n)

PS1 right 9995 10001 08327 041 40000left 9996 9995 iexcl02329 082 40000baseline 10003 10000 iexcl04994 062 40000

PS2 right 10001 9995 iexcl06746 050 22900left 10000 9991 iexcl09721 033 24000baseline 9996 9999 03060 076 24000

PS3 right 10011 10008 iexcl04347 066 34000left 9998 10001 03522 072 32000a

baseline 10005 10000 iexcl06143 054 32000PS4 right 9994 10005 26336 00084L 104000

left 10001 9995 iexcl13071 019 104000baseline 9998 9997 iexcl02664 079 104000

PS5 right 9986 10005 10582 029 6000left 10003 10005 01187 091 8000baseline 10002 9995 iexcl03727 071 6000

PS6 right 9968 9993 07972 043 2000left 9990 9976 iexcl04402 066 2000baseline 10008 9960 iexcl15583 012 2000

PS7 right 9990 9995 04285 067 16000left 10018 10002 iexcl14678 014 16000baseline 9983 9996 11654 024 16000

PS8 right 10008 10005 iexcl02559 080 24000left 10016 9984 iexcl35151 00004R 24000baseline 10002 10002 iexcl00648 095 24000

PS9 right 10001 9998 iexcl02441 081 18000left 10017 9998 iexcl16610 010 16000baseline 9993 9983 iexcl08404 040 16000

PS10 right 10009 10014 02525 080 6000left 10018 10000 iexcl07789 044 4000baseline 9987 10022 15658 012 4000

PS11 right 10006 10010 03433 073 16000left 9992 10000 07129 048 16000baseline 10007 9986 iexcl19258 00541R 16000

PS12 right 9997 9986 iexcl10275 030 16000left 10002 9999 iexcl02967 077 16000baseline 9978 10001 20961 00361L 16000

Note n ˆ number of intention trials Dagger control trials R ˆ direction of effect was to the right L ˆdirection of effect was to the left

a Pseudodata contained 500 fewer intention and control trials as compared to real data

662 M Freedman et al

T-tests were carried out using SAS System for Windows Release 8 todetermine whether there were significant differences between the intention andcontrol conditions for the right left and baseline intentions There wasa significant difference between the intention and control conditions for the rightintention (t[37998]ˆiexcl253 pˆ 00115) but not for the other two intentions (leftt[39998] ˆ 049 p ˆ 06269 and baseline t[39998] ˆ iexcl007 p ˆ 09418) Thesignificant effect for the right intention was in the direction of intention as in theoriginal study

Since the investigation was run in blocks of 1000 trials we examined the datausing blocks as the unit of measurement and the averaged output across trials asthe measure in each block An F-ratio test with condition (intention vs control)as the source of variance of interest and the interaction between condition andblock as the error term showed an effect of condition (p ˆ 00578) This suggeststhat the left frontal subject would show comparable findings on another series of19 pairs of blocks of 1000 trials with 94 confidence

Whereas the mean REG output for the right intention was above the expectedvalue of 100 the output for the control condition was below 100 (Figure 3) Toexamine the possibility that the significant difference between the intention andcontrol condition was due solely to low output during the control condition asopposed to high output during the intention condition we tested whether theoutput for the intention condition was significantly different from a constantvalue of 100 This value is the theoretical mean output of the REG assumingtruly random and unbiased output The control values for the left and baselineconditions are approximately equal to this theoretical mean The REG output forthe right intention was significantly different than 100 and in the direction ofintention (t ˆ 201 p ˆ 0045 two-tailed test) For the control condition themean REG output was less than (although not significantly different from) 100(t ˆ iexcl156 p ˆ 012)

Fig 3 Mean REG output for the left frontal patient in replication study

Effects of Frontal Lobe Lesions on Intentionality 663

Figure 4 (andashc) shows the mean output of the REG broken down byconsecutive 1000 trial sessions for each intention (right left and baseline) andfor the control output For the right intention Figure 4 shows that on most pairsof blocks of 1000 trials the mean REG output follows a fairly consistent patternin which the means are either higher compared to the control condition or aboutthe same In contrast the data for each of the left and baseline conditions showa pattern with less separation between conditions

Fig 4 Mean REG output for each 1000 trial session in replication study (a) Intention Right (b)Intention Left and (c) Intention Baseline

664 M Freedman et al

Discussion

We examined the claims from the PEAR program that an individualrsquosintentions can influence the statistical distribution of random physical phenom-ena Our main focus was to test these claims using well-designed control condi-tions in a population that might maximize the likelihood of detecting any effectsthat may exist Our choice of studying patients with frontal lobe lesions wasbased on the concept that the potential to influence random physical events maybe optimized when attention is diverted from self-awareness (ie such as whenself-awareness is reduced) This state can occur following frontal brain damage(Stuss et al 1986 Stuss 1991 Carver et al 1991) Our findings showed asignificant effect for the right intention in a subject with left frontal braindamage Moreover this effect was in the direction of intention In contrast therewere no significant effects in the other groups (ie bilateral frontal right frontalpooled frontal or normal subjects)

Although the result in the left frontal patient on the right intention wasstatistically significant even after correction for multiple comparisons thisfinding was interpreted with great caution First it was based on a relativelysmall number of intention trials (n ˆ 3000) and was derived from only onesubject Second we found a p-value for an individual lsquolsquosubjectrsquorsquo in the pseu-dodata that would also have met criteria for significance using a Bonferronicorrection and that was even less than the p-value of 00015 obtained fromthe left frontal subject Although the effect of this lsquolsquopseudopatientrsquorsquo was not inthe direction of intention the fact that it was significant raised caution forinterpreting the effect in the left frontal patient However the replication of thefindings in the left frontal patient in a second well-designed study for each of the

Fig 4 (Continued)

Effects of Frontal Lobe Lesions on Intentionality 665

three intentions right left and baseline suggests that the effect in this subjectmay be more than chance occurrence

Additional support for a reliable effect in the left frontal subject comes fromfurther analysis of his data suggesting about a 94 chance of his showinga similar finding if another series of 19 pairs of blocks of 1000 trials were runagain Furthermore the profile of REG output data for the right intentionshowed a fairly consistent separation whereas this was not the case for the left orbaseline intentions

A comment is warranted about the REG output for the left frontal patientbeing lower for the control condition on the right intention or more to the leftas compared to the control conditions for the left or baseline intentions Onemight argue that the significant effect on the right intention was an artifact ofcomparison to control data that was in the left direction and that this widened thedifference between the intention and control data However this argument is nottenable because the effect on the right intention was significant even when theREG output was compared to a theoretical mean of 100mdasha value which is ap-proximately equal to the control means for the left intention (9999985) and thebaseline intention (1000011)

The patientrsquos cognitive deficits and brain lesion have been describedelsewhere (Marras et al 1998) He suffered from a tension pneumocephaluswhich resulted in cognitive deficits and epileptic seizures The cognitive deficitsinclude decreased mental flexibility on the Trail-Making Test poor attentionreduced fluency and impaired spatial planning and visuospatial problem solvingMRI showed an extensive left frontal lesion but the right frontal lobe was intactPsychometric testing and SPECT suggested the addition of right frontal dysfunc-tion The SPECT findings provide a measure of function as opposed to structureand were subtle

As indicated above frontal lesions have been associated with reduced self-awareness (Stuss et al 1986 Stuss 1991 Carver et al 1991) a state that isdifficult for normal individuals to achieve The rationale for studying patientswith damage to the frontal lobes was that decreased self-awareness might facili-tate the effects of intentionality on random physical phenomena Brain regionsthat mediate neurological processes underlying self-awareness include thefrontal lobes bilaterally particularly on the right (Stuss et al 2001a Stuss ampAlexander 2000a Stuss amp Alexander 2000b Stuss et al 2001b) Whereas thepatientrsquos right-sided brain dysfunction may have been insufficient to producea deficit in self-awareness the extensive lesion on the left may have resulted inreduced self-awareness when attention was directed towards the right Howeverit remains unclear why positive results should be found only following damageto the left frontal region and not after bilateral or right frontal lesions Onespeculation is that the effect on random physical events may require reducedself-awareness combined with relatively intact attentional mechanisms Theassociation of frontal lobe lesions especially on the right with impaired at-tention (Stuss amp Levine 2002) may explain the negative findings in the setting

666 M Freedman et al

of bilateral or right frontal damage Whether deficits related to left frontalabnormalities can explain the observed effects on the REG output warrantsfurther study Moreover the question whether normal processes associated withintact frontal lobe function together with preserved self-awareness may serve toinhibit effects of intentionality on random physical phenomena needs to beaddressed

The strength of our conclusions rests largely on a well-designed methodol-ogy and replication of our findings Although our results did not replicate thefindings reported by Jahn and his colleagues in normal subjects (Jahn et al1987a 1997) they support their claims that intentionality can alter the output ofa random event generator Furthermore our findings suggest that patients withfrontal lobe lesions may serve as a good model for future studies of the effects ofconsciousness on random physical events

Acknowledgments

We gratefully acknowledge Dr Donald Stuss for his helpful advice andsupport throughout the study and for reviewing the manuscript and for hishelpful suggestions Dr Louis Siminovitch for helpful advice and supportPatricia Ebert and Dr Donald Stuss for assisting with analysis of lesion local-ization for the CT and MRI scans Janice Pogue for statistical advice LeeFerreira for testing the subjects John Sloan for assistance with data collec-tion Tania Xerri for assistance with data analyses and Tania Xerri MaedeanBrown Wanna Mar and Melissa Ziraldo for preparation of materials for pub-lication We also gratefully acknowledge Professor Robert Jahn BrendaDunne Dr Roger Nelson Dr York Dobyns and John Bradish at the PEARlab for their helpful advice and for providing the portable REG and softwarefor the study This study was supported by the Fetzer Institute Dr Freedmanwas supported by the Saul A Silverman Family Foundation Toronto Canadaas a Canada-International Scientific exchange program (CISEPO) project

References

Alcock J E (1987) ParapsychologyScience of the Anomalous or Search for the Soul Behavioraland Brain Sciences 10 553ndash643

Carver C S amp Scheier M F (1991) Self-Regulation and the Self In J Strauss amp G R Goethals(Eds) The Self Interdisciplinary Approaches (pp 168ndash207) New York Springer-Verlag

Hall J Kim C McElroy B amp Shimony A (1977) Wave-packet reduction as a medium ofcommunication Foundations of Physics 7 759ndash767

Ibison M amp Jeffers S (1998) A Double-Slit Diffraction Experiment to Investigate Claims ofConsciousness-Related Anomalies Journal of Scienti c Exploration 12 543ndash550

Jahn R G amp Dunne B J (1986) On the Quantum Mechanics of Consciousness with Applicationto Anomalous Phenomena Foundations of Physics 16 721ndash772

Jahn R G amp Dunne B J (1987a) Margins of Reality The Role of Consciousness in the PhysicalWorld San Diego Harcourt Brace Jovanovich

Jahn R G Dunne B J amp Nelson R D (1987b) Engineering Anomalies Research Journal ofScienti c Exploration 1 21ndash50

Effects of Frontal Lobe Lesions on Intentionality 667

The examiner then started the computer program which was designed to movethe arrow according to the output of a Random Event Generator (REG) The REGproduces a random series of 0rsquos and 1rsquos based upon a sampling of an electronicnoise pattern at pre-set regular intervals The sign of the signal at the time ofsampling (ie positive or negative) is compared with a regularly alternatingsequence of positive and negative pulses If there is a match (ie negative-negative or positive-positive) a lsquolsquo1rsquorsquo is generated If there is no match then a lsquolsquo0rsquorsquois generated (Jahn et al 1987a) Sampling occurred at a rate of 200 per secondThe data were summed as 200-sample bits with expected numerical value of 100due to a 50 chance of a positive-positive or negative-negativematch occurringat each sampling Each sample of 200 bits of data comprised a trial

The position of the arrow relative to the midline of the screen movedrightward or leftward according to the cumulative mean of the trials with themidline representing a cumulative mean of 100 An arrow with the tip to theright of midline represented a mean of greater than 100 and an arrow with the tipto the left corresponded to a mean of less than 100 The REG used for the ex-periment was a portable model of the larger device that has been used in thePEAR program at Princeton University (Jahn et al 1987a Nelson et al 1984Nelson et al 1989 Jahn et al 1987b Jahn et al 1997) The portable REG wasobtained from the PEAR program

Each intention (right left or baseline) consisted of 10 blocks of 100 trialsand lasted approximately 15 minutes At the end of each block of 100 trialsthe examiner confirmed that the subject understood the task answered anyquestions and initiated the next block of 100 trials After each block of 100trials the position of the arrow tip or bar was reset to the midline

After each intention for right left or baseline a control run of 1000 trials wascarried out without anyone in the room This completed one full session of 1000trials for the relevant intention and 1000 control trials Participation was spreadacross several sessions The total number of trials varied across subjects due todifferences in the time that they could devote to the study

Results

Primary analyses were carried out to test whether there were significanteffects of intention in the frontal subjects (bilateral frontal left frontal rightfrontal and frontal subjects pooled) and in the normal group Secondary analyseswere carried out to determine whether there were effects for individual subjects

Primary Analyses

Figure 2 (andashc) shows the mean output of the REG for each intention (rightleft baseline) and the mean control output for the patients with frontal lesionsand the normal subjects Table 2 shows the corresponding number of intentionand control trials

Effects of Frontal Lobe Lesions on Intentionality 655

T-tests were carried out using SAS System for Windows (The SAS Systemfor Windows Release 612 1996) to determine whether there were signifi-cant differences between intention and control conditions for the right left andbaseline intentions Separate analyses were carried out for patients with lesionsin the frontal lobes bilaterally left frontal lobe right frontal lobe all frontalpatients pooled and for the normal subjects This resulted in a total of 15analyses Using a Bonferroni correction for 15 multiple comparisons a p-value 00515 or 0003 would be required for statistical significance As shown in

Fig 2 Mean REG output for patients with brain lesions in bilateral frontal (BF) left frontal (LF)and right frontal (RF) regions frontal patients pooled (FR) and normal subjects (NORM)(a) Intention Right (b) Intention Left and (c) Intention Baseline

656 M Freedman et al

Table 3 there were no significant differences between the intention and controlconditions for the right left or baseline intentions in the bilateral frontal sub-jects right frontal subject or normal subjects In addition there were no signifi-cant differences between the intention and control conditions for the pooledgroup of frontal subjects

For the left frontal patient there was a significant difference between theintention and control conditions for the right intention (p ˆ 00015) The effectwas in the direction of intention and was significant even after Bonferroni cor-rection for multiple comparisons There were no significant differences betweenthe intention and control conditions for the left or baseline intentions in the leftfrontal patient

T-tests were carried out to determine whether there were significantdifferences between the intention and control conditions for the right left orbaseline intentions for the entire sample of frontal and normal subjects pooledThere were no significant differences for any intention (right t[304898] ˆ00480 p ˆ 096 left t[302998] ˆ 06709 p ˆ 050 baseline t[299998] ˆiexcl17053 p ˆ 009)

Secondary Analyses

Secondary analyses were carried out using t-tests to determine whether therewere significant differences between intention and control conditions for eachsubject Table 4 shows the results of the analysis for each subject Significantdifferences at the p ˆ 005 level were found between the intention and controlconditions in three instances intention right (p ˆ 00015) (S5 left frontalpatient) baseline (p ˆ 003) (S8 normal subject) intention right (p ˆ 0004) (S9

Fig 2 (Continued)

Effects of Frontal Lobe Lesions on Intentionality 657

normal subject) The effect was in the intended direction for S5 and S9 Usinga Bonferroni correction for 36 multiple comparisons a p-value 00536 or00014 would be required for statistical significance None of these p-valuesexceeded this threshold

As stated in the methodology section four normal subjects were research staff(S7 S8 S11 S12) and were tested without an examiner in the room In additionsubjects S9 and S10 were relatives of the research staff but were tested with anexaminer present These six subjects were divided into two groups for separateanalysis One group was comprised of the normal subjects who were researchstaff (S7 S8 S11 S12) and the other group was comprised of subjects S9 andS10 T-tests were carried out to determine whether there were significantdifferences between the intention and control conditions for the right left andbaseline intentions for these two groups This resulted in a total of six analysesUsing a Bonferroni correction for six multiple comparisons a p-value of 0056 or 0008 would be required for statistical significance There were no sig-nificant differences between the intention and control conditions for the groupcomprised of subjects S7 S8 S11 and S12 (right p ˆ 069 left p ˆ 061baseline p ˆ 024) For the group comprised of subjects S9 and S10 there wasa difference between the intention and control conditions for the right intention(p ˆ 00164) that was in the direction of intention However this was notsignificant after correction for multiple comparisons There were no significantdifferences between the intention and control conditions for the left (p ˆ 014) orbaseline (p ˆ 048) intentions for the group comprised of S9 and S10

In addition to statistical control for multiple comparisons using a Bonferronicorrection we performed an additional study to experimentally control for thenumerous statistical tests lsquolsquoPseudodatarsquorsquo were created by repeating the entire ex-perimental procedure except that there was no subject or experimenter in theroom during the intention condition Two sets of pseudodata were generated thefirst at the same site where the original data were collected (pseudodata 1) andthe second in a different building (pseudodata 2) The data were labeled to cor-respond to the intentions of the subjects as in the real study

As shown in Tables 5a and 5b the number of statistically significant resultsranged from one (pseudodata 1) to four (pseudodata 2) Moreover one of the

TABLE 2Total Number of Trials (Includes Intention Plus Control Trials)

Subjects Right Left Baseline

Bilateral frontals 200900a 201000 200000Left frontal 6000 8000 6000Right frontal 2000 2000 2000Frontals pooled 208900a 211000 208000Normals 96000 92000 92000

a There were 100 fewer intention trials than control trials for the right intention in the bilateralfrontals

658 M Freedman et al

significance levels was greater in the pseudodata than in the real data Howevernone of the effects were in the direction of intention and in three instances thesignificant effects occurred during the baseline intention

Replication Study

The primary analyses showed a statistically significant effect in the direc-tion of intention to the right for the patient with a left frontal lesion This effectwas significant even after Bonferroni correction for multiple comparisons Toexamine the possible effects in this patient further we carried out a second studyto determine whether the findings could be replicated The planned hypothesiswas that he would show a statistically significant effect of intention to the rightbut not to the left or in the baseline intention (ie that the findings would be thesame as before) The left frontal subject was retested using the methodologydescribed for the original study

During testing a printer error (data was always printed after each block of 100trials) occurred at the end of the fifth block of 100 trials during a baseline sessionand the program exited while the bar was still on the screen Although thecomplete data for these 100 trials were still captured a replacement set of 1000baseline and control trials for the entire intention was collected on a separateday However the possible impact of the printer error is considered negligibledue to the small number of trials involved the fact that none of the data werelost and because it occurred during the baseline condition in which the subjectwas asked not to concentrate on moving the arrow On another occasiona baseline session should have preceded a right intention session but the order was

TABLE 3Analysis of Frontal and Normal Subjects (Intention vs Control Conditions)

Subjects Intention t df p-value

Bilateral frontals right 11920 200898 023left 01213 200998 090baseline iexcl13874 199998 017

Left frontal right iexcl31691 5998 00015R

left 00524 7998 096baseline iexcl03656 5998 071

Right frontal right 00375 1998 097left iexcl08754 1998 038baseline iexcl16084 1998 011

Frontals pooled right 06377 208898 052left 00441 210998 096baseline iexcl15794 207998 011

Normals right iexcl08538 95998 039left 11474 91998 025baseline iexcl07036 91998 048

Note R ˆ Direction of effect was to the right

Effects of Frontal Lobe Lesions on Intentionality 659

reversed This resulted in two right intention and two baseline sessions being runconsecutively The right intention sessions were on the same day whereas thebaseline sessions were on different days Another occurrence consisted of thesubject seeing double arrows during a right intention session Finally during a left

TABLE 4Single Subject Analysis

Mean REG Output

Subject Intention Intention Control t p-value Trials (n)

S1 right 9998 10001 04982 062 40000left 9997 10003 08491 040 40000baseline 10003 10004 02000 084 40000

S2 right 9999 10007 08302 041 22900a

left 9988 9991 03145 075 24000baseline 10003 10001 iexcl01708 086 24000

S3 right 9989 10002 17414 008 34000left 10011 10000 iexcl13408 018 33000baseline 10013 10004 iexcl10462 030 32000

S4 right 10002 10002 iexcl00341 097 104000left 9999 10000 02488 080 104000baseline 10003 9997 iexcl13896 016 104000

S5 right 10019 9962 iexcl31691 00015R 6000left 9999 10000 00524 096 8000baseline 10001 9994 iexcl03656 071 6000

S6 right 10007 10008 00375 097 2000left 9991 9964 iexcl08754 038 2000baseline 10009 9959 iexcl16084 011 2000

S7 right 9984 9991 06708 050 16000left 10002 9998 iexcl03481 073 16000baseline 9994 9994 iexcl00022 100 16000

S8 right 10000 10006 07071 048 24000left 10003 10002 iexcl01489 088 24000baseline 10009 9989 iexcl21274 00334R 24000

S9 right 10021 9991 iexcl28790 00040R 18000left 10000 10012 10737 028 16000baseline 9994 9999 04375 066 16000b

S10 right 10001 10004 01845 085 6000left 9988 10014 11721 024 4000baseline 9974 9990 07088 048 4000

S11 right 10002 10002 iexcl00477 096 16000left 9999 10005 06132 054 16000baseline 10005 10004 iexcl01077 091 16000

S12 right 10007 10000 iexcl06514 051 16000left 9995 10007 10216 031 16000baseline 10000 10003 02445 081 16000

Note n ˆ number of intention trials Dagger control trials R ˆ direction of effect was to the righta There were 100 fewer intention trials than control trialsb 500 baseline control trials were collected incorrectly and replaced by trials collected at a different

time Although this introduced a potential bias the difference between the intention and controlconditions was too far from significant for a small number of 500 trials to have had an impact

660 M Freedman et al

intention session the building intercom came on at the end of the fourth block of100 trials The subject stated that this affected his concentration only slightly Inall cases described above the complete set of data was used for analysesincluding the replacement trials that followed the printer error (ie no data werediscarded)

TABLE 5aPseudodata Analysis 1

Mean REG Output

Subject Intention Intention Control t p-value Trials (n)

PS1 right 9997 10005 11636 024 40000left 10011 10003 iexcl10648 029 40000baseline 10007 9995 iexcl17586 008 40000

PS2 right 9995 10010 15873 011 22900left 9998 10004 06576 051 24000baseline 9994 10006 14214 016 24000

PS3 right 9996 10002 08405 040 34000left 9991 10002 13635 017 32000a

baseline 10007 9999 iexcl09442 035 32000PS4 right 10003 9999 iexcl10667 029 104000

left 10006 9999 iexcl17887 007 104000baseline 10001 9999 iexcl02984 077 104000

PS5 right 9982 9993 06322 053 6000left 10020 10000 iexcl12473 021 8000baseline 9988 9993 02343 081 6000

PS6 right 9999 10031 10198 031 2000left 9979 10006 08638 039 2000baseline 9963 10031 21474 00319L 2000

PS7 right 10002 10005 02463 081 16000left 9999 10013 12288 022 16000baseline 10005 9999 iexcl05196 060 16000

PS8 right 10006 9996 iexcl11546 025 24000left 10005 10007 02370 081 24000baseline 10000 9997 iexcl02564 080 24000

PS9 right 9995 10011 15199 013 18000left 9996 10002 05973 055 16000baseline 10005 9995 iexcl08764 038 16000

PS10 right 10007 9993 iexcl07168 047 6000left 9997 10016 08457 040 4000baseline 9981 9987 02792 078 4000

PS11 right 10005 10014 08604 039 16000left 10001 10006 04769 063 16000baseline 10009 9997 iexcl10813 028 16000

PS12 right 9993 10010 15528 012 16000left 10001 10012 09288 035 16000baseline 9998 10013 13483 018 16000

Note n ˆ number of intention trials Dagger control trials L ˆ direction of effect was to the lefta Pseudodata contained 500 fewer intention and control trials as compared to real data

Effects of Frontal Lobe Lesions on Intentionality 661

Results of Replication Study

Figure 3 shows the mean REG output for each intention (right left baseline)and the control output for the left frontal subject There were 19000 intentionand control trials for the right intention and 20000 trials for each of the inten-tion and control conditions for the left and baseline intentions

TABLE 5bPseudodata Analysis 2

Mean REG Output

Subject Intention Intention Control t p-value Trials (n)

PS1 right 9995 10001 08327 041 40000left 9996 9995 iexcl02329 082 40000baseline 10003 10000 iexcl04994 062 40000

PS2 right 10001 9995 iexcl06746 050 22900left 10000 9991 iexcl09721 033 24000baseline 9996 9999 03060 076 24000

PS3 right 10011 10008 iexcl04347 066 34000left 9998 10001 03522 072 32000a

baseline 10005 10000 iexcl06143 054 32000PS4 right 9994 10005 26336 00084L 104000

left 10001 9995 iexcl13071 019 104000baseline 9998 9997 iexcl02664 079 104000

PS5 right 9986 10005 10582 029 6000left 10003 10005 01187 091 8000baseline 10002 9995 iexcl03727 071 6000

PS6 right 9968 9993 07972 043 2000left 9990 9976 iexcl04402 066 2000baseline 10008 9960 iexcl15583 012 2000

PS7 right 9990 9995 04285 067 16000left 10018 10002 iexcl14678 014 16000baseline 9983 9996 11654 024 16000

PS8 right 10008 10005 iexcl02559 080 24000left 10016 9984 iexcl35151 00004R 24000baseline 10002 10002 iexcl00648 095 24000

PS9 right 10001 9998 iexcl02441 081 18000left 10017 9998 iexcl16610 010 16000baseline 9993 9983 iexcl08404 040 16000

PS10 right 10009 10014 02525 080 6000left 10018 10000 iexcl07789 044 4000baseline 9987 10022 15658 012 4000

PS11 right 10006 10010 03433 073 16000left 9992 10000 07129 048 16000baseline 10007 9986 iexcl19258 00541R 16000

PS12 right 9997 9986 iexcl10275 030 16000left 10002 9999 iexcl02967 077 16000baseline 9978 10001 20961 00361L 16000

Note n ˆ number of intention trials Dagger control trials R ˆ direction of effect was to the right L ˆdirection of effect was to the left

a Pseudodata contained 500 fewer intention and control trials as compared to real data

662 M Freedman et al

T-tests were carried out using SAS System for Windows Release 8 todetermine whether there were significant differences between the intention andcontrol conditions for the right left and baseline intentions There wasa significant difference between the intention and control conditions for the rightintention (t[37998]ˆiexcl253 pˆ 00115) but not for the other two intentions (leftt[39998] ˆ 049 p ˆ 06269 and baseline t[39998] ˆ iexcl007 p ˆ 09418) Thesignificant effect for the right intention was in the direction of intention as in theoriginal study

Since the investigation was run in blocks of 1000 trials we examined the datausing blocks as the unit of measurement and the averaged output across trials asthe measure in each block An F-ratio test with condition (intention vs control)as the source of variance of interest and the interaction between condition andblock as the error term showed an effect of condition (p ˆ 00578) This suggeststhat the left frontal subject would show comparable findings on another series of19 pairs of blocks of 1000 trials with 94 confidence

Whereas the mean REG output for the right intention was above the expectedvalue of 100 the output for the control condition was below 100 (Figure 3) Toexamine the possibility that the significant difference between the intention andcontrol condition was due solely to low output during the control condition asopposed to high output during the intention condition we tested whether theoutput for the intention condition was significantly different from a constantvalue of 100 This value is the theoretical mean output of the REG assumingtruly random and unbiased output The control values for the left and baselineconditions are approximately equal to this theoretical mean The REG output forthe right intention was significantly different than 100 and in the direction ofintention (t ˆ 201 p ˆ 0045 two-tailed test) For the control condition themean REG output was less than (although not significantly different from) 100(t ˆ iexcl156 p ˆ 012)

Fig 3 Mean REG output for the left frontal patient in replication study

Effects of Frontal Lobe Lesions on Intentionality 663

Figure 4 (andashc) shows the mean output of the REG broken down byconsecutive 1000 trial sessions for each intention (right left and baseline) andfor the control output For the right intention Figure 4 shows that on most pairsof blocks of 1000 trials the mean REG output follows a fairly consistent patternin which the means are either higher compared to the control condition or aboutthe same In contrast the data for each of the left and baseline conditions showa pattern with less separation between conditions

Fig 4 Mean REG output for each 1000 trial session in replication study (a) Intention Right (b)Intention Left and (c) Intention Baseline

664 M Freedman et al

Discussion

We examined the claims from the PEAR program that an individualrsquosintentions can influence the statistical distribution of random physical phenom-ena Our main focus was to test these claims using well-designed control condi-tions in a population that might maximize the likelihood of detecting any effectsthat may exist Our choice of studying patients with frontal lobe lesions wasbased on the concept that the potential to influence random physical events maybe optimized when attention is diverted from self-awareness (ie such as whenself-awareness is reduced) This state can occur following frontal brain damage(Stuss et al 1986 Stuss 1991 Carver et al 1991) Our findings showed asignificant effect for the right intention in a subject with left frontal braindamage Moreover this effect was in the direction of intention In contrast therewere no significant effects in the other groups (ie bilateral frontal right frontalpooled frontal or normal subjects)

Although the result in the left frontal patient on the right intention wasstatistically significant even after correction for multiple comparisons thisfinding was interpreted with great caution First it was based on a relativelysmall number of intention trials (n ˆ 3000) and was derived from only onesubject Second we found a p-value for an individual lsquolsquosubjectrsquorsquo in the pseu-dodata that would also have met criteria for significance using a Bonferronicorrection and that was even less than the p-value of 00015 obtained fromthe left frontal subject Although the effect of this lsquolsquopseudopatientrsquorsquo was not inthe direction of intention the fact that it was significant raised caution forinterpreting the effect in the left frontal patient However the replication of thefindings in the left frontal patient in a second well-designed study for each of the

Fig 4 (Continued)

Effects of Frontal Lobe Lesions on Intentionality 665

three intentions right left and baseline suggests that the effect in this subjectmay be more than chance occurrence

Additional support for a reliable effect in the left frontal subject comes fromfurther analysis of his data suggesting about a 94 chance of his showinga similar finding if another series of 19 pairs of blocks of 1000 trials were runagain Furthermore the profile of REG output data for the right intentionshowed a fairly consistent separation whereas this was not the case for the left orbaseline intentions

A comment is warranted about the REG output for the left frontal patientbeing lower for the control condition on the right intention or more to the leftas compared to the control conditions for the left or baseline intentions Onemight argue that the significant effect on the right intention was an artifact ofcomparison to control data that was in the left direction and that this widened thedifference between the intention and control data However this argument is nottenable because the effect on the right intention was significant even when theREG output was compared to a theoretical mean of 100mdasha value which is ap-proximately equal to the control means for the left intention (9999985) and thebaseline intention (1000011)

The patientrsquos cognitive deficits and brain lesion have been describedelsewhere (Marras et al 1998) He suffered from a tension pneumocephaluswhich resulted in cognitive deficits and epileptic seizures The cognitive deficitsinclude decreased mental flexibility on the Trail-Making Test poor attentionreduced fluency and impaired spatial planning and visuospatial problem solvingMRI showed an extensive left frontal lesion but the right frontal lobe was intactPsychometric testing and SPECT suggested the addition of right frontal dysfunc-tion The SPECT findings provide a measure of function as opposed to structureand were subtle

As indicated above frontal lesions have been associated with reduced self-awareness (Stuss et al 1986 Stuss 1991 Carver et al 1991) a state that isdifficult for normal individuals to achieve The rationale for studying patientswith damage to the frontal lobes was that decreased self-awareness might facili-tate the effects of intentionality on random physical phenomena Brain regionsthat mediate neurological processes underlying self-awareness include thefrontal lobes bilaterally particularly on the right (Stuss et al 2001a Stuss ampAlexander 2000a Stuss amp Alexander 2000b Stuss et al 2001b) Whereas thepatientrsquos right-sided brain dysfunction may have been insufficient to producea deficit in self-awareness the extensive lesion on the left may have resulted inreduced self-awareness when attention was directed towards the right Howeverit remains unclear why positive results should be found only following damageto the left frontal region and not after bilateral or right frontal lesions Onespeculation is that the effect on random physical events may require reducedself-awareness combined with relatively intact attentional mechanisms Theassociation of frontal lobe lesions especially on the right with impaired at-tention (Stuss amp Levine 2002) may explain the negative findings in the setting

666 M Freedman et al

of bilateral or right frontal damage Whether deficits related to left frontalabnormalities can explain the observed effects on the REG output warrantsfurther study Moreover the question whether normal processes associated withintact frontal lobe function together with preserved self-awareness may serve toinhibit effects of intentionality on random physical phenomena needs to beaddressed

The strength of our conclusions rests largely on a well-designed methodol-ogy and replication of our findings Although our results did not replicate thefindings reported by Jahn and his colleagues in normal subjects (Jahn et al1987a 1997) they support their claims that intentionality can alter the output ofa random event generator Furthermore our findings suggest that patients withfrontal lobe lesions may serve as a good model for future studies of the effects ofconsciousness on random physical events

Acknowledgments

We gratefully acknowledge Dr Donald Stuss for his helpful advice andsupport throughout the study and for reviewing the manuscript and for hishelpful suggestions Dr Louis Siminovitch for helpful advice and supportPatricia Ebert and Dr Donald Stuss for assisting with analysis of lesion local-ization for the CT and MRI scans Janice Pogue for statistical advice LeeFerreira for testing the subjects John Sloan for assistance with data collec-tion Tania Xerri for assistance with data analyses and Tania Xerri MaedeanBrown Wanna Mar and Melissa Ziraldo for preparation of materials for pub-lication We also gratefully acknowledge Professor Robert Jahn BrendaDunne Dr Roger Nelson Dr York Dobyns and John Bradish at the PEARlab for their helpful advice and for providing the portable REG and softwarefor the study This study was supported by the Fetzer Institute Dr Freedmanwas supported by the Saul A Silverman Family Foundation Toronto Canadaas a Canada-International Scientific exchange program (CISEPO) project

References

Alcock J E (1987) ParapsychologyScience of the Anomalous or Search for the Soul Behavioraland Brain Sciences 10 553ndash643

Carver C S amp Scheier M F (1991) Self-Regulation and the Self In J Strauss amp G R Goethals(Eds) The Self Interdisciplinary Approaches (pp 168ndash207) New York Springer-Verlag

Hall J Kim C McElroy B amp Shimony A (1977) Wave-packet reduction as a medium ofcommunication Foundations of Physics 7 759ndash767

Ibison M amp Jeffers S (1998) A Double-Slit Diffraction Experiment to Investigate Claims ofConsciousness-Related Anomalies Journal of Scienti c Exploration 12 543ndash550

Jahn R G amp Dunne B J (1986) On the Quantum Mechanics of Consciousness with Applicationto Anomalous Phenomena Foundations of Physics 16 721ndash772

Jahn R G amp Dunne B J (1987a) Margins of Reality The Role of Consciousness in the PhysicalWorld San Diego Harcourt Brace Jovanovich

Jahn R G Dunne B J amp Nelson R D (1987b) Engineering Anomalies Research Journal ofScienti c Exploration 1 21ndash50

Effects of Frontal Lobe Lesions on Intentionality 667

T-tests were carried out using SAS System for Windows (The SAS Systemfor Windows Release 612 1996) to determine whether there were signifi-cant differences between intention and control conditions for the right left andbaseline intentions Separate analyses were carried out for patients with lesionsin the frontal lobes bilaterally left frontal lobe right frontal lobe all frontalpatients pooled and for the normal subjects This resulted in a total of 15analyses Using a Bonferroni correction for 15 multiple comparisons a p-value 00515 or 0003 would be required for statistical significance As shown in

Fig 2 Mean REG output for patients with brain lesions in bilateral frontal (BF) left frontal (LF)and right frontal (RF) regions frontal patients pooled (FR) and normal subjects (NORM)(a) Intention Right (b) Intention Left and (c) Intention Baseline

656 M Freedman et al

Table 3 there were no significant differences between the intention and controlconditions for the right left or baseline intentions in the bilateral frontal sub-jects right frontal subject or normal subjects In addition there were no signifi-cant differences between the intention and control conditions for the pooledgroup of frontal subjects

For the left frontal patient there was a significant difference between theintention and control conditions for the right intention (p ˆ 00015) The effectwas in the direction of intention and was significant even after Bonferroni cor-rection for multiple comparisons There were no significant differences betweenthe intention and control conditions for the left or baseline intentions in the leftfrontal patient

T-tests were carried out to determine whether there were significantdifferences between the intention and control conditions for the right left orbaseline intentions for the entire sample of frontal and normal subjects pooledThere were no significant differences for any intention (right t[304898] ˆ00480 p ˆ 096 left t[302998] ˆ 06709 p ˆ 050 baseline t[299998] ˆiexcl17053 p ˆ 009)

Secondary Analyses

Secondary analyses were carried out using t-tests to determine whether therewere significant differences between intention and control conditions for eachsubject Table 4 shows the results of the analysis for each subject Significantdifferences at the p ˆ 005 level were found between the intention and controlconditions in three instances intention right (p ˆ 00015) (S5 left frontalpatient) baseline (p ˆ 003) (S8 normal subject) intention right (p ˆ 0004) (S9

Fig 2 (Continued)

Effects of Frontal Lobe Lesions on Intentionality 657

normal subject) The effect was in the intended direction for S5 and S9 Usinga Bonferroni correction for 36 multiple comparisons a p-value 00536 or00014 would be required for statistical significance None of these p-valuesexceeded this threshold

As stated in the methodology section four normal subjects were research staff(S7 S8 S11 S12) and were tested without an examiner in the room In additionsubjects S9 and S10 were relatives of the research staff but were tested with anexaminer present These six subjects were divided into two groups for separateanalysis One group was comprised of the normal subjects who were researchstaff (S7 S8 S11 S12) and the other group was comprised of subjects S9 andS10 T-tests were carried out to determine whether there were significantdifferences between the intention and control conditions for the right left andbaseline intentions for these two groups This resulted in a total of six analysesUsing a Bonferroni correction for six multiple comparisons a p-value of 0056 or 0008 would be required for statistical significance There were no sig-nificant differences between the intention and control conditions for the groupcomprised of subjects S7 S8 S11 and S12 (right p ˆ 069 left p ˆ 061baseline p ˆ 024) For the group comprised of subjects S9 and S10 there wasa difference between the intention and control conditions for the right intention(p ˆ 00164) that was in the direction of intention However this was notsignificant after correction for multiple comparisons There were no significantdifferences between the intention and control conditions for the left (p ˆ 014) orbaseline (p ˆ 048) intentions for the group comprised of S9 and S10

In addition to statistical control for multiple comparisons using a Bonferronicorrection we performed an additional study to experimentally control for thenumerous statistical tests lsquolsquoPseudodatarsquorsquo were created by repeating the entire ex-perimental procedure except that there was no subject or experimenter in theroom during the intention condition Two sets of pseudodata were generated thefirst at the same site where the original data were collected (pseudodata 1) andthe second in a different building (pseudodata 2) The data were labeled to cor-respond to the intentions of the subjects as in the real study

As shown in Tables 5a and 5b the number of statistically significant resultsranged from one (pseudodata 1) to four (pseudodata 2) Moreover one of the

TABLE 2Total Number of Trials (Includes Intention Plus Control Trials)

Subjects Right Left Baseline

Bilateral frontals 200900a 201000 200000Left frontal 6000 8000 6000Right frontal 2000 2000 2000Frontals pooled 208900a 211000 208000Normals 96000 92000 92000

a There were 100 fewer intention trials than control trials for the right intention in the bilateralfrontals

658 M Freedman et al

significance levels was greater in the pseudodata than in the real data Howevernone of the effects were in the direction of intention and in three instances thesignificant effects occurred during the baseline intention

Replication Study

The primary analyses showed a statistically significant effect in the direc-tion of intention to the right for the patient with a left frontal lesion This effectwas significant even after Bonferroni correction for multiple comparisons Toexamine the possible effects in this patient further we carried out a second studyto determine whether the findings could be replicated The planned hypothesiswas that he would show a statistically significant effect of intention to the rightbut not to the left or in the baseline intention (ie that the findings would be thesame as before) The left frontal subject was retested using the methodologydescribed for the original study

During testing a printer error (data was always printed after each block of 100trials) occurred at the end of the fifth block of 100 trials during a baseline sessionand the program exited while the bar was still on the screen Although thecomplete data for these 100 trials were still captured a replacement set of 1000baseline and control trials for the entire intention was collected on a separateday However the possible impact of the printer error is considered negligibledue to the small number of trials involved the fact that none of the data werelost and because it occurred during the baseline condition in which the subjectwas asked not to concentrate on moving the arrow On another occasiona baseline session should have preceded a right intention session but the order was

TABLE 3Analysis of Frontal and Normal Subjects (Intention vs Control Conditions)

Subjects Intention t df p-value

Bilateral frontals right 11920 200898 023left 01213 200998 090baseline iexcl13874 199998 017

Left frontal right iexcl31691 5998 00015R

left 00524 7998 096baseline iexcl03656 5998 071

Right frontal right 00375 1998 097left iexcl08754 1998 038baseline iexcl16084 1998 011

Frontals pooled right 06377 208898 052left 00441 210998 096baseline iexcl15794 207998 011

Normals right iexcl08538 95998 039left 11474 91998 025baseline iexcl07036 91998 048

Note R ˆ Direction of effect was to the right

Effects of Frontal Lobe Lesions on Intentionality 659

reversed This resulted in two right intention and two baseline sessions being runconsecutively The right intention sessions were on the same day whereas thebaseline sessions were on different days Another occurrence consisted of thesubject seeing double arrows during a right intention session Finally during a left

TABLE 4Single Subject Analysis

Mean REG Output

Subject Intention Intention Control t p-value Trials (n)

S1 right 9998 10001 04982 062 40000left 9997 10003 08491 040 40000baseline 10003 10004 02000 084 40000

S2 right 9999 10007 08302 041 22900a

left 9988 9991 03145 075 24000baseline 10003 10001 iexcl01708 086 24000

S3 right 9989 10002 17414 008 34000left 10011 10000 iexcl13408 018 33000baseline 10013 10004 iexcl10462 030 32000

S4 right 10002 10002 iexcl00341 097 104000left 9999 10000 02488 080 104000baseline 10003 9997 iexcl13896 016 104000

S5 right 10019 9962 iexcl31691 00015R 6000left 9999 10000 00524 096 8000baseline 10001 9994 iexcl03656 071 6000

S6 right 10007 10008 00375 097 2000left 9991 9964 iexcl08754 038 2000baseline 10009 9959 iexcl16084 011 2000

S7 right 9984 9991 06708 050 16000left 10002 9998 iexcl03481 073 16000baseline 9994 9994 iexcl00022 100 16000

S8 right 10000 10006 07071 048 24000left 10003 10002 iexcl01489 088 24000baseline 10009 9989 iexcl21274 00334R 24000

S9 right 10021 9991 iexcl28790 00040R 18000left 10000 10012 10737 028 16000baseline 9994 9999 04375 066 16000b

S10 right 10001 10004 01845 085 6000left 9988 10014 11721 024 4000baseline 9974 9990 07088 048 4000

S11 right 10002 10002 iexcl00477 096 16000left 9999 10005 06132 054 16000baseline 10005 10004 iexcl01077 091 16000

S12 right 10007 10000 iexcl06514 051 16000left 9995 10007 10216 031 16000baseline 10000 10003 02445 081 16000

Note n ˆ number of intention trials Dagger control trials R ˆ direction of effect was to the righta There were 100 fewer intention trials than control trialsb 500 baseline control trials were collected incorrectly and replaced by trials collected at a different

time Although this introduced a potential bias the difference between the intention and controlconditions was too far from significant for a small number of 500 trials to have had an impact

660 M Freedman et al

intention session the building intercom came on at the end of the fourth block of100 trials The subject stated that this affected his concentration only slightly Inall cases described above the complete set of data was used for analysesincluding the replacement trials that followed the printer error (ie no data werediscarded)

TABLE 5aPseudodata Analysis 1

Mean REG Output

Subject Intention Intention Control t p-value Trials (n)

PS1 right 9997 10005 11636 024 40000left 10011 10003 iexcl10648 029 40000baseline 10007 9995 iexcl17586 008 40000

PS2 right 9995 10010 15873 011 22900left 9998 10004 06576 051 24000baseline 9994 10006 14214 016 24000

PS3 right 9996 10002 08405 040 34000left 9991 10002 13635 017 32000a

baseline 10007 9999 iexcl09442 035 32000PS4 right 10003 9999 iexcl10667 029 104000

left 10006 9999 iexcl17887 007 104000baseline 10001 9999 iexcl02984 077 104000

PS5 right 9982 9993 06322 053 6000left 10020 10000 iexcl12473 021 8000baseline 9988 9993 02343 081 6000

PS6 right 9999 10031 10198 031 2000left 9979 10006 08638 039 2000baseline 9963 10031 21474 00319L 2000

PS7 right 10002 10005 02463 081 16000left 9999 10013 12288 022 16000baseline 10005 9999 iexcl05196 060 16000

PS8 right 10006 9996 iexcl11546 025 24000left 10005 10007 02370 081 24000baseline 10000 9997 iexcl02564 080 24000

PS9 right 9995 10011 15199 013 18000left 9996 10002 05973 055 16000baseline 10005 9995 iexcl08764 038 16000

PS10 right 10007 9993 iexcl07168 047 6000left 9997 10016 08457 040 4000baseline 9981 9987 02792 078 4000

PS11 right 10005 10014 08604 039 16000left 10001 10006 04769 063 16000baseline 10009 9997 iexcl10813 028 16000

PS12 right 9993 10010 15528 012 16000left 10001 10012 09288 035 16000baseline 9998 10013 13483 018 16000

Note n ˆ number of intention trials Dagger control trials L ˆ direction of effect was to the lefta Pseudodata contained 500 fewer intention and control trials as compared to real data

Effects of Frontal Lobe Lesions on Intentionality 661

Results of Replication Study

Figure 3 shows the mean REG output for each intention (right left baseline)and the control output for the left frontal subject There were 19000 intentionand control trials for the right intention and 20000 trials for each of the inten-tion and control conditions for the left and baseline intentions

TABLE 5bPseudodata Analysis 2

Mean REG Output

Subject Intention Intention Control t p-value Trials (n)

PS1 right 9995 10001 08327 041 40000left 9996 9995 iexcl02329 082 40000baseline 10003 10000 iexcl04994 062 40000

PS2 right 10001 9995 iexcl06746 050 22900left 10000 9991 iexcl09721 033 24000baseline 9996 9999 03060 076 24000

PS3 right 10011 10008 iexcl04347 066 34000left 9998 10001 03522 072 32000a

baseline 10005 10000 iexcl06143 054 32000PS4 right 9994 10005 26336 00084L 104000

left 10001 9995 iexcl13071 019 104000baseline 9998 9997 iexcl02664 079 104000

PS5 right 9986 10005 10582 029 6000left 10003 10005 01187 091 8000baseline 10002 9995 iexcl03727 071 6000

PS6 right 9968 9993 07972 043 2000left 9990 9976 iexcl04402 066 2000baseline 10008 9960 iexcl15583 012 2000

PS7 right 9990 9995 04285 067 16000left 10018 10002 iexcl14678 014 16000baseline 9983 9996 11654 024 16000

PS8 right 10008 10005 iexcl02559 080 24000left 10016 9984 iexcl35151 00004R 24000baseline 10002 10002 iexcl00648 095 24000

PS9 right 10001 9998 iexcl02441 081 18000left 10017 9998 iexcl16610 010 16000baseline 9993 9983 iexcl08404 040 16000

PS10 right 10009 10014 02525 080 6000left 10018 10000 iexcl07789 044 4000baseline 9987 10022 15658 012 4000

PS11 right 10006 10010 03433 073 16000left 9992 10000 07129 048 16000baseline 10007 9986 iexcl19258 00541R 16000

PS12 right 9997 9986 iexcl10275 030 16000left 10002 9999 iexcl02967 077 16000baseline 9978 10001 20961 00361L 16000

Note n ˆ number of intention trials Dagger control trials R ˆ direction of effect was to the right L ˆdirection of effect was to the left

a Pseudodata contained 500 fewer intention and control trials as compared to real data

662 M Freedman et al

T-tests were carried out using SAS System for Windows Release 8 todetermine whether there were significant differences between the intention andcontrol conditions for the right left and baseline intentions There wasa significant difference between the intention and control conditions for the rightintention (t[37998]ˆiexcl253 pˆ 00115) but not for the other two intentions (leftt[39998] ˆ 049 p ˆ 06269 and baseline t[39998] ˆ iexcl007 p ˆ 09418) Thesignificant effect for the right intention was in the direction of intention as in theoriginal study

Since the investigation was run in blocks of 1000 trials we examined the datausing blocks as the unit of measurement and the averaged output across trials asthe measure in each block An F-ratio test with condition (intention vs control)as the source of variance of interest and the interaction between condition andblock as the error term showed an effect of condition (p ˆ 00578) This suggeststhat the left frontal subject would show comparable findings on another series of19 pairs of blocks of 1000 trials with 94 confidence

Whereas the mean REG output for the right intention was above the expectedvalue of 100 the output for the control condition was below 100 (Figure 3) Toexamine the possibility that the significant difference between the intention andcontrol condition was due solely to low output during the control condition asopposed to high output during the intention condition we tested whether theoutput for the intention condition was significantly different from a constantvalue of 100 This value is the theoretical mean output of the REG assumingtruly random and unbiased output The control values for the left and baselineconditions are approximately equal to this theoretical mean The REG output forthe right intention was significantly different than 100 and in the direction ofintention (t ˆ 201 p ˆ 0045 two-tailed test) For the control condition themean REG output was less than (although not significantly different from) 100(t ˆ iexcl156 p ˆ 012)

Fig 3 Mean REG output for the left frontal patient in replication study

Effects of Frontal Lobe Lesions on Intentionality 663

Figure 4 (andashc) shows the mean output of the REG broken down byconsecutive 1000 trial sessions for each intention (right left and baseline) andfor the control output For the right intention Figure 4 shows that on most pairsof blocks of 1000 trials the mean REG output follows a fairly consistent patternin which the means are either higher compared to the control condition or aboutthe same In contrast the data for each of the left and baseline conditions showa pattern with less separation between conditions

Fig 4 Mean REG output for each 1000 trial session in replication study (a) Intention Right (b)Intention Left and (c) Intention Baseline

664 M Freedman et al

Discussion

We examined the claims from the PEAR program that an individualrsquosintentions can influence the statistical distribution of random physical phenom-ena Our main focus was to test these claims using well-designed control condi-tions in a population that might maximize the likelihood of detecting any effectsthat may exist Our choice of studying patients with frontal lobe lesions wasbased on the concept that the potential to influence random physical events maybe optimized when attention is diverted from self-awareness (ie such as whenself-awareness is reduced) This state can occur following frontal brain damage(Stuss et al 1986 Stuss 1991 Carver et al 1991) Our findings showed asignificant effect for the right intention in a subject with left frontal braindamage Moreover this effect was in the direction of intention In contrast therewere no significant effects in the other groups (ie bilateral frontal right frontalpooled frontal or normal subjects)

Although the result in the left frontal patient on the right intention wasstatistically significant even after correction for multiple comparisons thisfinding was interpreted with great caution First it was based on a relativelysmall number of intention trials (n ˆ 3000) and was derived from only onesubject Second we found a p-value for an individual lsquolsquosubjectrsquorsquo in the pseu-dodata that would also have met criteria for significance using a Bonferronicorrection and that was even less than the p-value of 00015 obtained fromthe left frontal subject Although the effect of this lsquolsquopseudopatientrsquorsquo was not inthe direction of intention the fact that it was significant raised caution forinterpreting the effect in the left frontal patient However the replication of thefindings in the left frontal patient in a second well-designed study for each of the

Fig 4 (Continued)

Effects of Frontal Lobe Lesions on Intentionality 665

three intentions right left and baseline suggests that the effect in this subjectmay be more than chance occurrence

Additional support for a reliable effect in the left frontal subject comes fromfurther analysis of his data suggesting about a 94 chance of his showinga similar finding if another series of 19 pairs of blocks of 1000 trials were runagain Furthermore the profile of REG output data for the right intentionshowed a fairly consistent separation whereas this was not the case for the left orbaseline intentions

A comment is warranted about the REG output for the left frontal patientbeing lower for the control condition on the right intention or more to the leftas compared to the control conditions for the left or baseline intentions Onemight argue that the significant effect on the right intention was an artifact ofcomparison to control data that was in the left direction and that this widened thedifference between the intention and control data However this argument is nottenable because the effect on the right intention was significant even when theREG output was compared to a theoretical mean of 100mdasha value which is ap-proximately equal to the control means for the left intention (9999985) and thebaseline intention (1000011)

The patientrsquos cognitive deficits and brain lesion have been describedelsewhere (Marras et al 1998) He suffered from a tension pneumocephaluswhich resulted in cognitive deficits and epileptic seizures The cognitive deficitsinclude decreased mental flexibility on the Trail-Making Test poor attentionreduced fluency and impaired spatial planning and visuospatial problem solvingMRI showed an extensive left frontal lesion but the right frontal lobe was intactPsychometric testing and SPECT suggested the addition of right frontal dysfunc-tion The SPECT findings provide a measure of function as opposed to structureand were subtle

As indicated above frontal lesions have been associated with reduced self-awareness (Stuss et al 1986 Stuss 1991 Carver et al 1991) a state that isdifficult for normal individuals to achieve The rationale for studying patientswith damage to the frontal lobes was that decreased self-awareness might facili-tate the effects of intentionality on random physical phenomena Brain regionsthat mediate neurological processes underlying self-awareness include thefrontal lobes bilaterally particularly on the right (Stuss et al 2001a Stuss ampAlexander 2000a Stuss amp Alexander 2000b Stuss et al 2001b) Whereas thepatientrsquos right-sided brain dysfunction may have been insufficient to producea deficit in self-awareness the extensive lesion on the left may have resulted inreduced self-awareness when attention was directed towards the right Howeverit remains unclear why positive results should be found only following damageto the left frontal region and not after bilateral or right frontal lesions Onespeculation is that the effect on random physical events may require reducedself-awareness combined with relatively intact attentional mechanisms Theassociation of frontal lobe lesions especially on the right with impaired at-tention (Stuss amp Levine 2002) may explain the negative findings in the setting

666 M Freedman et al

of bilateral or right frontal damage Whether deficits related to left frontalabnormalities can explain the observed effects on the REG output warrantsfurther study Moreover the question whether normal processes associated withintact frontal lobe function together with preserved self-awareness may serve toinhibit effects of intentionality on random physical phenomena needs to beaddressed

The strength of our conclusions rests largely on a well-designed methodol-ogy and replication of our findings Although our results did not replicate thefindings reported by Jahn and his colleagues in normal subjects (Jahn et al1987a 1997) they support their claims that intentionality can alter the output ofa random event generator Furthermore our findings suggest that patients withfrontal lobe lesions may serve as a good model for future studies of the effects ofconsciousness on random physical events

Acknowledgments

We gratefully acknowledge Dr Donald Stuss for his helpful advice andsupport throughout the study and for reviewing the manuscript and for hishelpful suggestions Dr Louis Siminovitch for helpful advice and supportPatricia Ebert and Dr Donald Stuss for assisting with analysis of lesion local-ization for the CT and MRI scans Janice Pogue for statistical advice LeeFerreira for testing the subjects John Sloan for assistance with data collec-tion Tania Xerri for assistance with data analyses and Tania Xerri MaedeanBrown Wanna Mar and Melissa Ziraldo for preparation of materials for pub-lication We also gratefully acknowledge Professor Robert Jahn BrendaDunne Dr Roger Nelson Dr York Dobyns and John Bradish at the PEARlab for their helpful advice and for providing the portable REG and softwarefor the study This study was supported by the Fetzer Institute Dr Freedmanwas supported by the Saul A Silverman Family Foundation Toronto Canadaas a Canada-International Scientific exchange program (CISEPO) project

References

Alcock J E (1987) ParapsychologyScience of the Anomalous or Search for the Soul Behavioraland Brain Sciences 10 553ndash643

Carver C S amp Scheier M F (1991) Self-Regulation and the Self In J Strauss amp G R Goethals(Eds) The Self Interdisciplinary Approaches (pp 168ndash207) New York Springer-Verlag

Hall J Kim C McElroy B amp Shimony A (1977) Wave-packet reduction as a medium ofcommunication Foundations of Physics 7 759ndash767

Ibison M amp Jeffers S (1998) A Double-Slit Diffraction Experiment to Investigate Claims ofConsciousness-Related Anomalies Journal of Scienti c Exploration 12 543ndash550

Jahn R G amp Dunne B J (1986) On the Quantum Mechanics of Consciousness with Applicationto Anomalous Phenomena Foundations of Physics 16 721ndash772

Jahn R G amp Dunne B J (1987a) Margins of Reality The Role of Consciousness in the PhysicalWorld San Diego Harcourt Brace Jovanovich

Jahn R G Dunne B J amp Nelson R D (1987b) Engineering Anomalies Research Journal ofScienti c Exploration 1 21ndash50

Effects of Frontal Lobe Lesions on Intentionality 667

Table 3 there were no significant differences between the intention and controlconditions for the right left or baseline intentions in the bilateral frontal sub-jects right frontal subject or normal subjects In addition there were no signifi-cant differences between the intention and control conditions for the pooledgroup of frontal subjects

For the left frontal patient there was a significant difference between theintention and control conditions for the right intention (p ˆ 00015) The effectwas in the direction of intention and was significant even after Bonferroni cor-rection for multiple comparisons There were no significant differences betweenthe intention and control conditions for the left or baseline intentions in the leftfrontal patient

T-tests were carried out to determine whether there were significantdifferences between the intention and control conditions for the right left orbaseline intentions for the entire sample of frontal and normal subjects pooledThere were no significant differences for any intention (right t[304898] ˆ00480 p ˆ 096 left t[302998] ˆ 06709 p ˆ 050 baseline t[299998] ˆiexcl17053 p ˆ 009)

Secondary Analyses

Secondary analyses were carried out using t-tests to determine whether therewere significant differences between intention and control conditions for eachsubject Table 4 shows the results of the analysis for each subject Significantdifferences at the p ˆ 005 level were found between the intention and controlconditions in three instances intention right (p ˆ 00015) (S5 left frontalpatient) baseline (p ˆ 003) (S8 normal subject) intention right (p ˆ 0004) (S9

Fig 2 (Continued)

Effects of Frontal Lobe Lesions on Intentionality 657

normal subject) The effect was in the intended direction for S5 and S9 Usinga Bonferroni correction for 36 multiple comparisons a p-value 00536 or00014 would be required for statistical significance None of these p-valuesexceeded this threshold

As stated in the methodology section four normal subjects were research staff(S7 S8 S11 S12) and were tested without an examiner in the room In additionsubjects S9 and S10 were relatives of the research staff but were tested with anexaminer present These six subjects were divided into two groups for separateanalysis One group was comprised of the normal subjects who were researchstaff (S7 S8 S11 S12) and the other group was comprised of subjects S9 andS10 T-tests were carried out to determine whether there were significantdifferences between the intention and control conditions for the right left andbaseline intentions for these two groups This resulted in a total of six analysesUsing a Bonferroni correction for six multiple comparisons a p-value of 0056 or 0008 would be required for statistical significance There were no sig-nificant differences between the intention and control conditions for the groupcomprised of subjects S7 S8 S11 and S12 (right p ˆ 069 left p ˆ 061baseline p ˆ 024) For the group comprised of subjects S9 and S10 there wasa difference between the intention and control conditions for the right intention(p ˆ 00164) that was in the direction of intention However this was notsignificant after correction for multiple comparisons There were no significantdifferences between the intention and control conditions for the left (p ˆ 014) orbaseline (p ˆ 048) intentions for the group comprised of S9 and S10

In addition to statistical control for multiple comparisons using a Bonferronicorrection we performed an additional study to experimentally control for thenumerous statistical tests lsquolsquoPseudodatarsquorsquo were created by repeating the entire ex-perimental procedure except that there was no subject or experimenter in theroom during the intention condition Two sets of pseudodata were generated thefirst at the same site where the original data were collected (pseudodata 1) andthe second in a different building (pseudodata 2) The data were labeled to cor-respond to the intentions of the subjects as in the real study

As shown in Tables 5a and 5b the number of statistically significant resultsranged from one (pseudodata 1) to four (pseudodata 2) Moreover one of the

TABLE 2Total Number of Trials (Includes Intention Plus Control Trials)

Subjects Right Left Baseline

Bilateral frontals 200900a 201000 200000Left frontal 6000 8000 6000Right frontal 2000 2000 2000Frontals pooled 208900a 211000 208000Normals 96000 92000 92000

a There were 100 fewer intention trials than control trials for the right intention in the bilateralfrontals

658 M Freedman et al

significance levels was greater in the pseudodata than in the real data Howevernone of the effects were in the direction of intention and in three instances thesignificant effects occurred during the baseline intention

Replication Study

The primary analyses showed a statistically significant effect in the direc-tion of intention to the right for the patient with a left frontal lesion This effectwas significant even after Bonferroni correction for multiple comparisons Toexamine the possible effects in this patient further we carried out a second studyto determine whether the findings could be replicated The planned hypothesiswas that he would show a statistically significant effect of intention to the rightbut not to the left or in the baseline intention (ie that the findings would be thesame as before) The left frontal subject was retested using the methodologydescribed for the original study

During testing a printer error (data was always printed after each block of 100trials) occurred at the end of the fifth block of 100 trials during a baseline sessionand the program exited while the bar was still on the screen Although thecomplete data for these 100 trials were still captured a replacement set of 1000baseline and control trials for the entire intention was collected on a separateday However the possible impact of the printer error is considered negligibledue to the small number of trials involved the fact that none of the data werelost and because it occurred during the baseline condition in which the subjectwas asked not to concentrate on moving the arrow On another occasiona baseline session should have preceded a right intention session but the order was

TABLE 3Analysis of Frontal and Normal Subjects (Intention vs Control Conditions)

Subjects Intention t df p-value

Bilateral frontals right 11920 200898 023left 01213 200998 090baseline iexcl13874 199998 017

Left frontal right iexcl31691 5998 00015R

left 00524 7998 096baseline iexcl03656 5998 071

Right frontal right 00375 1998 097left iexcl08754 1998 038baseline iexcl16084 1998 011

Frontals pooled right 06377 208898 052left 00441 210998 096baseline iexcl15794 207998 011

Normals right iexcl08538 95998 039left 11474 91998 025baseline iexcl07036 91998 048

Note R ˆ Direction of effect was to the right

Effects of Frontal Lobe Lesions on Intentionality 659

reversed This resulted in two right intention and two baseline sessions being runconsecutively The right intention sessions were on the same day whereas thebaseline sessions were on different days Another occurrence consisted of thesubject seeing double arrows during a right intention session Finally during a left

TABLE 4Single Subject Analysis

Mean REG Output

Subject Intention Intention Control t p-value Trials (n)

S1 right 9998 10001 04982 062 40000left 9997 10003 08491 040 40000baseline 10003 10004 02000 084 40000

S2 right 9999 10007 08302 041 22900a

left 9988 9991 03145 075 24000baseline 10003 10001 iexcl01708 086 24000

S3 right 9989 10002 17414 008 34000left 10011 10000 iexcl13408 018 33000baseline 10013 10004 iexcl10462 030 32000

S4 right 10002 10002 iexcl00341 097 104000left 9999 10000 02488 080 104000baseline 10003 9997 iexcl13896 016 104000

S5 right 10019 9962 iexcl31691 00015R 6000left 9999 10000 00524 096 8000baseline 10001 9994 iexcl03656 071 6000

S6 right 10007 10008 00375 097 2000left 9991 9964 iexcl08754 038 2000baseline 10009 9959 iexcl16084 011 2000

S7 right 9984 9991 06708 050 16000left 10002 9998 iexcl03481 073 16000baseline 9994 9994 iexcl00022 100 16000

S8 right 10000 10006 07071 048 24000left 10003 10002 iexcl01489 088 24000baseline 10009 9989 iexcl21274 00334R 24000

S9 right 10021 9991 iexcl28790 00040R 18000left 10000 10012 10737 028 16000baseline 9994 9999 04375 066 16000b

S10 right 10001 10004 01845 085 6000left 9988 10014 11721 024 4000baseline 9974 9990 07088 048 4000

S11 right 10002 10002 iexcl00477 096 16000left 9999 10005 06132 054 16000baseline 10005 10004 iexcl01077 091 16000

S12 right 10007 10000 iexcl06514 051 16000left 9995 10007 10216 031 16000baseline 10000 10003 02445 081 16000

Note n ˆ number of intention trials Dagger control trials R ˆ direction of effect was to the righta There were 100 fewer intention trials than control trialsb 500 baseline control trials were collected incorrectly and replaced by trials collected at a different

time Although this introduced a potential bias the difference between the intention and controlconditions was too far from significant for a small number of 500 trials to have had an impact

660 M Freedman et al

intention session the building intercom came on at the end of the fourth block of100 trials The subject stated that this affected his concentration only slightly Inall cases described above the complete set of data was used for analysesincluding the replacement trials that followed the printer error (ie no data werediscarded)

TABLE 5aPseudodata Analysis 1

Mean REG Output

Subject Intention Intention Control t p-value Trials (n)

PS1 right 9997 10005 11636 024 40000left 10011 10003 iexcl10648 029 40000baseline 10007 9995 iexcl17586 008 40000

PS2 right 9995 10010 15873 011 22900left 9998 10004 06576 051 24000baseline 9994 10006 14214 016 24000

PS3 right 9996 10002 08405 040 34000left 9991 10002 13635 017 32000a

baseline 10007 9999 iexcl09442 035 32000PS4 right 10003 9999 iexcl10667 029 104000

left 10006 9999 iexcl17887 007 104000baseline 10001 9999 iexcl02984 077 104000

PS5 right 9982 9993 06322 053 6000left 10020 10000 iexcl12473 021 8000baseline 9988 9993 02343 081 6000

PS6 right 9999 10031 10198 031 2000left 9979 10006 08638 039 2000baseline 9963 10031 21474 00319L 2000

PS7 right 10002 10005 02463 081 16000left 9999 10013 12288 022 16000baseline 10005 9999 iexcl05196 060 16000

PS8 right 10006 9996 iexcl11546 025 24000left 10005 10007 02370 081 24000baseline 10000 9997 iexcl02564 080 24000

PS9 right 9995 10011 15199 013 18000left 9996 10002 05973 055 16000baseline 10005 9995 iexcl08764 038 16000

PS10 right 10007 9993 iexcl07168 047 6000left 9997 10016 08457 040 4000baseline 9981 9987 02792 078 4000

PS11 right 10005 10014 08604 039 16000left 10001 10006 04769 063 16000baseline 10009 9997 iexcl10813 028 16000

PS12 right 9993 10010 15528 012 16000left 10001 10012 09288 035 16000baseline 9998 10013 13483 018 16000

Note n ˆ number of intention trials Dagger control trials L ˆ direction of effect was to the lefta Pseudodata contained 500 fewer intention and control trials as compared to real data

Effects of Frontal Lobe Lesions on Intentionality 661

Results of Replication Study

Figure 3 shows the mean REG output for each intention (right left baseline)and the control output for the left frontal subject There were 19000 intentionand control trials for the right intention and 20000 trials for each of the inten-tion and control conditions for the left and baseline intentions

TABLE 5bPseudodata Analysis 2

Mean REG Output

Subject Intention Intention Control t p-value Trials (n)

PS1 right 9995 10001 08327 041 40000left 9996 9995 iexcl02329 082 40000baseline 10003 10000 iexcl04994 062 40000

PS2 right 10001 9995 iexcl06746 050 22900left 10000 9991 iexcl09721 033 24000baseline 9996 9999 03060 076 24000

PS3 right 10011 10008 iexcl04347 066 34000left 9998 10001 03522 072 32000a

baseline 10005 10000 iexcl06143 054 32000PS4 right 9994 10005 26336 00084L 104000

left 10001 9995 iexcl13071 019 104000baseline 9998 9997 iexcl02664 079 104000

PS5 right 9986 10005 10582 029 6000left 10003 10005 01187 091 8000baseline 10002 9995 iexcl03727 071 6000

PS6 right 9968 9993 07972 043 2000left 9990 9976 iexcl04402 066 2000baseline 10008 9960 iexcl15583 012 2000

PS7 right 9990 9995 04285 067 16000left 10018 10002 iexcl14678 014 16000baseline 9983 9996 11654 024 16000

PS8 right 10008 10005 iexcl02559 080 24000left 10016 9984 iexcl35151 00004R 24000baseline 10002 10002 iexcl00648 095 24000

PS9 right 10001 9998 iexcl02441 081 18000left 10017 9998 iexcl16610 010 16000baseline 9993 9983 iexcl08404 040 16000

PS10 right 10009 10014 02525 080 6000left 10018 10000 iexcl07789 044 4000baseline 9987 10022 15658 012 4000

PS11 right 10006 10010 03433 073 16000left 9992 10000 07129 048 16000baseline 10007 9986 iexcl19258 00541R 16000

PS12 right 9997 9986 iexcl10275 030 16000left 10002 9999 iexcl02967 077 16000baseline 9978 10001 20961 00361L 16000

Note n ˆ number of intention trials Dagger control trials R ˆ direction of effect was to the right L ˆdirection of effect was to the left

a Pseudodata contained 500 fewer intention and control trials as compared to real data

662 M Freedman et al

T-tests were carried out using SAS System for Windows Release 8 todetermine whether there were significant differences between the intention andcontrol conditions for the right left and baseline intentions There wasa significant difference between the intention and control conditions for the rightintention (t[37998]ˆiexcl253 pˆ 00115) but not for the other two intentions (leftt[39998] ˆ 049 p ˆ 06269 and baseline t[39998] ˆ iexcl007 p ˆ 09418) Thesignificant effect for the right intention was in the direction of intention as in theoriginal study

Since the investigation was run in blocks of 1000 trials we examined the datausing blocks as the unit of measurement and the averaged output across trials asthe measure in each block An F-ratio test with condition (intention vs control)as the source of variance of interest and the interaction between condition andblock as the error term showed an effect of condition (p ˆ 00578) This suggeststhat the left frontal subject would show comparable findings on another series of19 pairs of blocks of 1000 trials with 94 confidence

Whereas the mean REG output for the right intention was above the expectedvalue of 100 the output for the control condition was below 100 (Figure 3) Toexamine the possibility that the significant difference between the intention andcontrol condition was due solely to low output during the control condition asopposed to high output during the intention condition we tested whether theoutput for the intention condition was significantly different from a constantvalue of 100 This value is the theoretical mean output of the REG assumingtruly random and unbiased output The control values for the left and baselineconditions are approximately equal to this theoretical mean The REG output forthe right intention was significantly different than 100 and in the direction ofintention (t ˆ 201 p ˆ 0045 two-tailed test) For the control condition themean REG output was less than (although not significantly different from) 100(t ˆ iexcl156 p ˆ 012)

Fig 3 Mean REG output for the left frontal patient in replication study

Effects of Frontal Lobe Lesions on Intentionality 663

Figure 4 (andashc) shows the mean output of the REG broken down byconsecutive 1000 trial sessions for each intention (right left and baseline) andfor the control output For the right intention Figure 4 shows that on most pairsof blocks of 1000 trials the mean REG output follows a fairly consistent patternin which the means are either higher compared to the control condition or aboutthe same In contrast the data for each of the left and baseline conditions showa pattern with less separation between conditions

Fig 4 Mean REG output for each 1000 trial session in replication study (a) Intention Right (b)Intention Left and (c) Intention Baseline

664 M Freedman et al

Discussion

We examined the claims from the PEAR program that an individualrsquosintentions can influence the statistical distribution of random physical phenom-ena Our main focus was to test these claims using well-designed control condi-tions in a population that might maximize the likelihood of detecting any effectsthat may exist Our choice of studying patients with frontal lobe lesions wasbased on the concept that the potential to influence random physical events maybe optimized when attention is diverted from self-awareness (ie such as whenself-awareness is reduced) This state can occur following frontal brain damage(Stuss et al 1986 Stuss 1991 Carver et al 1991) Our findings showed asignificant effect for the right intention in a subject with left frontal braindamage Moreover this effect was in the direction of intention In contrast therewere no significant effects in the other groups (ie bilateral frontal right frontalpooled frontal or normal subjects)

Although the result in the left frontal patient on the right intention wasstatistically significant even after correction for multiple comparisons thisfinding was interpreted with great caution First it was based on a relativelysmall number of intention trials (n ˆ 3000) and was derived from only onesubject Second we found a p-value for an individual lsquolsquosubjectrsquorsquo in the pseu-dodata that would also have met criteria for significance using a Bonferronicorrection and that was even less than the p-value of 00015 obtained fromthe left frontal subject Although the effect of this lsquolsquopseudopatientrsquorsquo was not inthe direction of intention the fact that it was significant raised caution forinterpreting the effect in the left frontal patient However the replication of thefindings in the left frontal patient in a second well-designed study for each of the

Fig 4 (Continued)

Effects of Frontal Lobe Lesions on Intentionality 665

three intentions right left and baseline suggests that the effect in this subjectmay be more than chance occurrence

Additional support for a reliable effect in the left frontal subject comes fromfurther analysis of his data suggesting about a 94 chance of his showinga similar finding if another series of 19 pairs of blocks of 1000 trials were runagain Furthermore the profile of REG output data for the right intentionshowed a fairly consistent separation whereas this was not the case for the left orbaseline intentions

A comment is warranted about the REG output for the left frontal patientbeing lower for the control condition on the right intention or more to the leftas compared to the control conditions for the left or baseline intentions Onemight argue that the significant effect on the right intention was an artifact ofcomparison to control data that was in the left direction and that this widened thedifference between the intention and control data However this argument is nottenable because the effect on the right intention was significant even when theREG output was compared to a theoretical mean of 100mdasha value which is ap-proximately equal to the control means for the left intention (9999985) and thebaseline intention (1000011)

The patientrsquos cognitive deficits and brain lesion have been describedelsewhere (Marras et al 1998) He suffered from a tension pneumocephaluswhich resulted in cognitive deficits and epileptic seizures The cognitive deficitsinclude decreased mental flexibility on the Trail-Making Test poor attentionreduced fluency and impaired spatial planning and visuospatial problem solvingMRI showed an extensive left frontal lesion but the right frontal lobe was intactPsychometric testing and SPECT suggested the addition of right frontal dysfunc-tion The SPECT findings provide a measure of function as opposed to structureand were subtle

As indicated above frontal lesions have been associated with reduced self-awareness (Stuss et al 1986 Stuss 1991 Carver et al 1991) a state that isdifficult for normal individuals to achieve The rationale for studying patientswith damage to the frontal lobes was that decreased self-awareness might facili-tate the effects of intentionality on random physical phenomena Brain regionsthat mediate neurological processes underlying self-awareness include thefrontal lobes bilaterally particularly on the right (Stuss et al 2001a Stuss ampAlexander 2000a Stuss amp Alexander 2000b Stuss et al 2001b) Whereas thepatientrsquos right-sided brain dysfunction may have been insufficient to producea deficit in self-awareness the extensive lesion on the left may have resulted inreduced self-awareness when attention was directed towards the right Howeverit remains unclear why positive results should be found only following damageto the left frontal region and not after bilateral or right frontal lesions Onespeculation is that the effect on random physical events may require reducedself-awareness combined with relatively intact attentional mechanisms Theassociation of frontal lobe lesions especially on the right with impaired at-tention (Stuss amp Levine 2002) may explain the negative findings in the setting

666 M Freedman et al

of bilateral or right frontal damage Whether deficits related to left frontalabnormalities can explain the observed effects on the REG output warrantsfurther study Moreover the question whether normal processes associated withintact frontal lobe function together with preserved self-awareness may serve toinhibit effects of intentionality on random physical phenomena needs to beaddressed

The strength of our conclusions rests largely on a well-designed methodol-ogy and replication of our findings Although our results did not replicate thefindings reported by Jahn and his colleagues in normal subjects (Jahn et al1987a 1997) they support their claims that intentionality can alter the output ofa random event generator Furthermore our findings suggest that patients withfrontal lobe lesions may serve as a good model for future studies of the effects ofconsciousness on random physical events

Acknowledgments

We gratefully acknowledge Dr Donald Stuss for his helpful advice andsupport throughout the study and for reviewing the manuscript and for hishelpful suggestions Dr Louis Siminovitch for helpful advice and supportPatricia Ebert and Dr Donald Stuss for assisting with analysis of lesion local-ization for the CT and MRI scans Janice Pogue for statistical advice LeeFerreira for testing the subjects John Sloan for assistance with data collec-tion Tania Xerri for assistance with data analyses and Tania Xerri MaedeanBrown Wanna Mar and Melissa Ziraldo for preparation of materials for pub-lication We also gratefully acknowledge Professor Robert Jahn BrendaDunne Dr Roger Nelson Dr York Dobyns and John Bradish at the PEARlab for their helpful advice and for providing the portable REG and softwarefor the study This study was supported by the Fetzer Institute Dr Freedmanwas supported by the Saul A Silverman Family Foundation Toronto Canadaas a Canada-International Scientific exchange program (CISEPO) project

References

Alcock J E (1987) ParapsychologyScience of the Anomalous or Search for the Soul Behavioraland Brain Sciences 10 553ndash643

Carver C S amp Scheier M F (1991) Self-Regulation and the Self In J Strauss amp G R Goethals(Eds) The Self Interdisciplinary Approaches (pp 168ndash207) New York Springer-Verlag

Hall J Kim C McElroy B amp Shimony A (1977) Wave-packet reduction as a medium ofcommunication Foundations of Physics 7 759ndash767

Ibison M amp Jeffers S (1998) A Double-Slit Diffraction Experiment to Investigate Claims ofConsciousness-Related Anomalies Journal of Scienti c Exploration 12 543ndash550

Jahn R G amp Dunne B J (1986) On the Quantum Mechanics of Consciousness with Applicationto Anomalous Phenomena Foundations of Physics 16 721ndash772

Jahn R G amp Dunne B J (1987a) Margins of Reality The Role of Consciousness in the PhysicalWorld San Diego Harcourt Brace Jovanovich

Jahn R G Dunne B J amp Nelson R D (1987b) Engineering Anomalies Research Journal ofScienti c Exploration 1 21ndash50

Effects of Frontal Lobe Lesions on Intentionality 667

normal subject) The effect was in the intended direction for S5 and S9 Usinga Bonferroni correction for 36 multiple comparisons a p-value 00536 or00014 would be required for statistical significance None of these p-valuesexceeded this threshold

As stated in the methodology section four normal subjects were research staff(S7 S8 S11 S12) and were tested without an examiner in the room In additionsubjects S9 and S10 were relatives of the research staff but were tested with anexaminer present These six subjects were divided into two groups for separateanalysis One group was comprised of the normal subjects who were researchstaff (S7 S8 S11 S12) and the other group was comprised of subjects S9 andS10 T-tests were carried out to determine whether there were significantdifferences between the intention and control conditions for the right left andbaseline intentions for these two groups This resulted in a total of six analysesUsing a Bonferroni correction for six multiple comparisons a p-value of 0056 or 0008 would be required for statistical significance There were no sig-nificant differences between the intention and control conditions for the groupcomprised of subjects S7 S8 S11 and S12 (right p ˆ 069 left p ˆ 061baseline p ˆ 024) For the group comprised of subjects S9 and S10 there wasa difference between the intention and control conditions for the right intention(p ˆ 00164) that was in the direction of intention However this was notsignificant after correction for multiple comparisons There were no significantdifferences between the intention and control conditions for the left (p ˆ 014) orbaseline (p ˆ 048) intentions for the group comprised of S9 and S10

In addition to statistical control for multiple comparisons using a Bonferronicorrection we performed an additional study to experimentally control for thenumerous statistical tests lsquolsquoPseudodatarsquorsquo were created by repeating the entire ex-perimental procedure except that there was no subject or experimenter in theroom during the intention condition Two sets of pseudodata were generated thefirst at the same site where the original data were collected (pseudodata 1) andthe second in a different building (pseudodata 2) The data were labeled to cor-respond to the intentions of the subjects as in the real study

As shown in Tables 5a and 5b the number of statistically significant resultsranged from one (pseudodata 1) to four (pseudodata 2) Moreover one of the

TABLE 2Total Number of Trials (Includes Intention Plus Control Trials)

Subjects Right Left Baseline

Bilateral frontals 200900a 201000 200000Left frontal 6000 8000 6000Right frontal 2000 2000 2000Frontals pooled 208900a 211000 208000Normals 96000 92000 92000

a There were 100 fewer intention trials than control trials for the right intention in the bilateralfrontals

658 M Freedman et al

significance levels was greater in the pseudodata than in the real data Howevernone of the effects were in the direction of intention and in three instances thesignificant effects occurred during the baseline intention

Replication Study

The primary analyses showed a statistically significant effect in the direc-tion of intention to the right for the patient with a left frontal lesion This effectwas significant even after Bonferroni correction for multiple comparisons Toexamine the possible effects in this patient further we carried out a second studyto determine whether the findings could be replicated The planned hypothesiswas that he would show a statistically significant effect of intention to the rightbut not to the left or in the baseline intention (ie that the findings would be thesame as before) The left frontal subject was retested using the methodologydescribed for the original study

During testing a printer error (data was always printed after each block of 100trials) occurred at the end of the fifth block of 100 trials during a baseline sessionand the program exited while the bar was still on the screen Although thecomplete data for these 100 trials were still captured a replacement set of 1000baseline and control trials for the entire intention was collected on a separateday However the possible impact of the printer error is considered negligibledue to the small number of trials involved the fact that none of the data werelost and because it occurred during the baseline condition in which the subjectwas asked not to concentrate on moving the arrow On another occasiona baseline session should have preceded a right intention session but the order was

TABLE 3Analysis of Frontal and Normal Subjects (Intention vs Control Conditions)

Subjects Intention t df p-value

Bilateral frontals right 11920 200898 023left 01213 200998 090baseline iexcl13874 199998 017

Left frontal right iexcl31691 5998 00015R

left 00524 7998 096baseline iexcl03656 5998 071

Right frontal right 00375 1998 097left iexcl08754 1998 038baseline iexcl16084 1998 011

Frontals pooled right 06377 208898 052left 00441 210998 096baseline iexcl15794 207998 011

Normals right iexcl08538 95998 039left 11474 91998 025baseline iexcl07036 91998 048

Note R ˆ Direction of effect was to the right

Effects of Frontal Lobe Lesions on Intentionality 659

reversed This resulted in two right intention and two baseline sessions being runconsecutively The right intention sessions were on the same day whereas thebaseline sessions were on different days Another occurrence consisted of thesubject seeing double arrows during a right intention session Finally during a left

TABLE 4Single Subject Analysis

Mean REG Output

Subject Intention Intention Control t p-value Trials (n)

S1 right 9998 10001 04982 062 40000left 9997 10003 08491 040 40000baseline 10003 10004 02000 084 40000

S2 right 9999 10007 08302 041 22900a

left 9988 9991 03145 075 24000baseline 10003 10001 iexcl01708 086 24000

S3 right 9989 10002 17414 008 34000left 10011 10000 iexcl13408 018 33000baseline 10013 10004 iexcl10462 030 32000

S4 right 10002 10002 iexcl00341 097 104000left 9999 10000 02488 080 104000baseline 10003 9997 iexcl13896 016 104000

S5 right 10019 9962 iexcl31691 00015R 6000left 9999 10000 00524 096 8000baseline 10001 9994 iexcl03656 071 6000

S6 right 10007 10008 00375 097 2000left 9991 9964 iexcl08754 038 2000baseline 10009 9959 iexcl16084 011 2000

S7 right 9984 9991 06708 050 16000left 10002 9998 iexcl03481 073 16000baseline 9994 9994 iexcl00022 100 16000

S8 right 10000 10006 07071 048 24000left 10003 10002 iexcl01489 088 24000baseline 10009 9989 iexcl21274 00334R 24000

S9 right 10021 9991 iexcl28790 00040R 18000left 10000 10012 10737 028 16000baseline 9994 9999 04375 066 16000b

S10 right 10001 10004 01845 085 6000left 9988 10014 11721 024 4000baseline 9974 9990 07088 048 4000

S11 right 10002 10002 iexcl00477 096 16000left 9999 10005 06132 054 16000baseline 10005 10004 iexcl01077 091 16000

S12 right 10007 10000 iexcl06514 051 16000left 9995 10007 10216 031 16000baseline 10000 10003 02445 081 16000

Note n ˆ number of intention trials Dagger control trials R ˆ direction of effect was to the righta There were 100 fewer intention trials than control trialsb 500 baseline control trials were collected incorrectly and replaced by trials collected at a different

time Although this introduced a potential bias the difference between the intention and controlconditions was too far from significant for a small number of 500 trials to have had an impact

660 M Freedman et al

intention session the building intercom came on at the end of the fourth block of100 trials The subject stated that this affected his concentration only slightly Inall cases described above the complete set of data was used for analysesincluding the replacement trials that followed the printer error (ie no data werediscarded)

TABLE 5aPseudodata Analysis 1

Mean REG Output

Subject Intention Intention Control t p-value Trials (n)

PS1 right 9997 10005 11636 024 40000left 10011 10003 iexcl10648 029 40000baseline 10007 9995 iexcl17586 008 40000

PS2 right 9995 10010 15873 011 22900left 9998 10004 06576 051 24000baseline 9994 10006 14214 016 24000

PS3 right 9996 10002 08405 040 34000left 9991 10002 13635 017 32000a

baseline 10007 9999 iexcl09442 035 32000PS4 right 10003 9999 iexcl10667 029 104000

left 10006 9999 iexcl17887 007 104000baseline 10001 9999 iexcl02984 077 104000

PS5 right 9982 9993 06322 053 6000left 10020 10000 iexcl12473 021 8000baseline 9988 9993 02343 081 6000

PS6 right 9999 10031 10198 031 2000left 9979 10006 08638 039 2000baseline 9963 10031 21474 00319L 2000

PS7 right 10002 10005 02463 081 16000left 9999 10013 12288 022 16000baseline 10005 9999 iexcl05196 060 16000

PS8 right 10006 9996 iexcl11546 025 24000left 10005 10007 02370 081 24000baseline 10000 9997 iexcl02564 080 24000

PS9 right 9995 10011 15199 013 18000left 9996 10002 05973 055 16000baseline 10005 9995 iexcl08764 038 16000

PS10 right 10007 9993 iexcl07168 047 6000left 9997 10016 08457 040 4000baseline 9981 9987 02792 078 4000

PS11 right 10005 10014 08604 039 16000left 10001 10006 04769 063 16000baseline 10009 9997 iexcl10813 028 16000

PS12 right 9993 10010 15528 012 16000left 10001 10012 09288 035 16000baseline 9998 10013 13483 018 16000

Note n ˆ number of intention trials Dagger control trials L ˆ direction of effect was to the lefta Pseudodata contained 500 fewer intention and control trials as compared to real data

Effects of Frontal Lobe Lesions on Intentionality 661

Results of Replication Study

Figure 3 shows the mean REG output for each intention (right left baseline)and the control output for the left frontal subject There were 19000 intentionand control trials for the right intention and 20000 trials for each of the inten-tion and control conditions for the left and baseline intentions

TABLE 5bPseudodata Analysis 2

Mean REG Output

Subject Intention Intention Control t p-value Trials (n)

PS1 right 9995 10001 08327 041 40000left 9996 9995 iexcl02329 082 40000baseline 10003 10000 iexcl04994 062 40000

PS2 right 10001 9995 iexcl06746 050 22900left 10000 9991 iexcl09721 033 24000baseline 9996 9999 03060 076 24000

PS3 right 10011 10008 iexcl04347 066 34000left 9998 10001 03522 072 32000a

baseline 10005 10000 iexcl06143 054 32000PS4 right 9994 10005 26336 00084L 104000

left 10001 9995 iexcl13071 019 104000baseline 9998 9997 iexcl02664 079 104000

PS5 right 9986 10005 10582 029 6000left 10003 10005 01187 091 8000baseline 10002 9995 iexcl03727 071 6000

PS6 right 9968 9993 07972 043 2000left 9990 9976 iexcl04402 066 2000baseline 10008 9960 iexcl15583 012 2000

PS7 right 9990 9995 04285 067 16000left 10018 10002 iexcl14678 014 16000baseline 9983 9996 11654 024 16000

PS8 right 10008 10005 iexcl02559 080 24000left 10016 9984 iexcl35151 00004R 24000baseline 10002 10002 iexcl00648 095 24000

PS9 right 10001 9998 iexcl02441 081 18000left 10017 9998 iexcl16610 010 16000baseline 9993 9983 iexcl08404 040 16000

PS10 right 10009 10014 02525 080 6000left 10018 10000 iexcl07789 044 4000baseline 9987 10022 15658 012 4000

PS11 right 10006 10010 03433 073 16000left 9992 10000 07129 048 16000baseline 10007 9986 iexcl19258 00541R 16000

PS12 right 9997 9986 iexcl10275 030 16000left 10002 9999 iexcl02967 077 16000baseline 9978 10001 20961 00361L 16000

Note n ˆ number of intention trials Dagger control trials R ˆ direction of effect was to the right L ˆdirection of effect was to the left

a Pseudodata contained 500 fewer intention and control trials as compared to real data

662 M Freedman et al

T-tests were carried out using SAS System for Windows Release 8 todetermine whether there were significant differences between the intention andcontrol conditions for the right left and baseline intentions There wasa significant difference between the intention and control conditions for the rightintention (t[37998]ˆiexcl253 pˆ 00115) but not for the other two intentions (leftt[39998] ˆ 049 p ˆ 06269 and baseline t[39998] ˆ iexcl007 p ˆ 09418) Thesignificant effect for the right intention was in the direction of intention as in theoriginal study

Since the investigation was run in blocks of 1000 trials we examined the datausing blocks as the unit of measurement and the averaged output across trials asthe measure in each block An F-ratio test with condition (intention vs control)as the source of variance of interest and the interaction between condition andblock as the error term showed an effect of condition (p ˆ 00578) This suggeststhat the left frontal subject would show comparable findings on another series of19 pairs of blocks of 1000 trials with 94 confidence

Whereas the mean REG output for the right intention was above the expectedvalue of 100 the output for the control condition was below 100 (Figure 3) Toexamine the possibility that the significant difference between the intention andcontrol condition was due solely to low output during the control condition asopposed to high output during the intention condition we tested whether theoutput for the intention condition was significantly different from a constantvalue of 100 This value is the theoretical mean output of the REG assumingtruly random and unbiased output The control values for the left and baselineconditions are approximately equal to this theoretical mean The REG output forthe right intention was significantly different than 100 and in the direction ofintention (t ˆ 201 p ˆ 0045 two-tailed test) For the control condition themean REG output was less than (although not significantly different from) 100(t ˆ iexcl156 p ˆ 012)

Fig 3 Mean REG output for the left frontal patient in replication study

Effects of Frontal Lobe Lesions on Intentionality 663

Figure 4 (andashc) shows the mean output of the REG broken down byconsecutive 1000 trial sessions for each intention (right left and baseline) andfor the control output For the right intention Figure 4 shows that on most pairsof blocks of 1000 trials the mean REG output follows a fairly consistent patternin which the means are either higher compared to the control condition or aboutthe same In contrast the data for each of the left and baseline conditions showa pattern with less separation between conditions

Fig 4 Mean REG output for each 1000 trial session in replication study (a) Intention Right (b)Intention Left and (c) Intention Baseline

664 M Freedman et al

Discussion

We examined the claims from the PEAR program that an individualrsquosintentions can influence the statistical distribution of random physical phenom-ena Our main focus was to test these claims using well-designed control condi-tions in a population that might maximize the likelihood of detecting any effectsthat may exist Our choice of studying patients with frontal lobe lesions wasbased on the concept that the potential to influence random physical events maybe optimized when attention is diverted from self-awareness (ie such as whenself-awareness is reduced) This state can occur following frontal brain damage(Stuss et al 1986 Stuss 1991 Carver et al 1991) Our findings showed asignificant effect for the right intention in a subject with left frontal braindamage Moreover this effect was in the direction of intention In contrast therewere no significant effects in the other groups (ie bilateral frontal right frontalpooled frontal or normal subjects)

Although the result in the left frontal patient on the right intention wasstatistically significant even after correction for multiple comparisons thisfinding was interpreted with great caution First it was based on a relativelysmall number of intention trials (n ˆ 3000) and was derived from only onesubject Second we found a p-value for an individual lsquolsquosubjectrsquorsquo in the pseu-dodata that would also have met criteria for significance using a Bonferronicorrection and that was even less than the p-value of 00015 obtained fromthe left frontal subject Although the effect of this lsquolsquopseudopatientrsquorsquo was not inthe direction of intention the fact that it was significant raised caution forinterpreting the effect in the left frontal patient However the replication of thefindings in the left frontal patient in a second well-designed study for each of the

Fig 4 (Continued)

Effects of Frontal Lobe Lesions on Intentionality 665

three intentions right left and baseline suggests that the effect in this subjectmay be more than chance occurrence

Additional support for a reliable effect in the left frontal subject comes fromfurther analysis of his data suggesting about a 94 chance of his showinga similar finding if another series of 19 pairs of blocks of 1000 trials were runagain Furthermore the profile of REG output data for the right intentionshowed a fairly consistent separation whereas this was not the case for the left orbaseline intentions

A comment is warranted about the REG output for the left frontal patientbeing lower for the control condition on the right intention or more to the leftas compared to the control conditions for the left or baseline intentions Onemight argue that the significant effect on the right intention was an artifact ofcomparison to control data that was in the left direction and that this widened thedifference between the intention and control data However this argument is nottenable because the effect on the right intention was significant even when theREG output was compared to a theoretical mean of 100mdasha value which is ap-proximately equal to the control means for the left intention (9999985) and thebaseline intention (1000011)

The patientrsquos cognitive deficits and brain lesion have been describedelsewhere (Marras et al 1998) He suffered from a tension pneumocephaluswhich resulted in cognitive deficits and epileptic seizures The cognitive deficitsinclude decreased mental flexibility on the Trail-Making Test poor attentionreduced fluency and impaired spatial planning and visuospatial problem solvingMRI showed an extensive left frontal lesion but the right frontal lobe was intactPsychometric testing and SPECT suggested the addition of right frontal dysfunc-tion The SPECT findings provide a measure of function as opposed to structureand were subtle

As indicated above frontal lesions have been associated with reduced self-awareness (Stuss et al 1986 Stuss 1991 Carver et al 1991) a state that isdifficult for normal individuals to achieve The rationale for studying patientswith damage to the frontal lobes was that decreased self-awareness might facili-tate the effects of intentionality on random physical phenomena Brain regionsthat mediate neurological processes underlying self-awareness include thefrontal lobes bilaterally particularly on the right (Stuss et al 2001a Stuss ampAlexander 2000a Stuss amp Alexander 2000b Stuss et al 2001b) Whereas thepatientrsquos right-sided brain dysfunction may have been insufficient to producea deficit in self-awareness the extensive lesion on the left may have resulted inreduced self-awareness when attention was directed towards the right Howeverit remains unclear why positive results should be found only following damageto the left frontal region and not after bilateral or right frontal lesions Onespeculation is that the effect on random physical events may require reducedself-awareness combined with relatively intact attentional mechanisms Theassociation of frontal lobe lesions especially on the right with impaired at-tention (Stuss amp Levine 2002) may explain the negative findings in the setting

666 M Freedman et al

of bilateral or right frontal damage Whether deficits related to left frontalabnormalities can explain the observed effects on the REG output warrantsfurther study Moreover the question whether normal processes associated withintact frontal lobe function together with preserved self-awareness may serve toinhibit effects of intentionality on random physical phenomena needs to beaddressed

The strength of our conclusions rests largely on a well-designed methodol-ogy and replication of our findings Although our results did not replicate thefindings reported by Jahn and his colleagues in normal subjects (Jahn et al1987a 1997) they support their claims that intentionality can alter the output ofa random event generator Furthermore our findings suggest that patients withfrontal lobe lesions may serve as a good model for future studies of the effects ofconsciousness on random physical events

Acknowledgments

We gratefully acknowledge Dr Donald Stuss for his helpful advice andsupport throughout the study and for reviewing the manuscript and for hishelpful suggestions Dr Louis Siminovitch for helpful advice and supportPatricia Ebert and Dr Donald Stuss for assisting with analysis of lesion local-ization for the CT and MRI scans Janice Pogue for statistical advice LeeFerreira for testing the subjects John Sloan for assistance with data collec-tion Tania Xerri for assistance with data analyses and Tania Xerri MaedeanBrown Wanna Mar and Melissa Ziraldo for preparation of materials for pub-lication We also gratefully acknowledge Professor Robert Jahn BrendaDunne Dr Roger Nelson Dr York Dobyns and John Bradish at the PEARlab for their helpful advice and for providing the portable REG and softwarefor the study This study was supported by the Fetzer Institute Dr Freedmanwas supported by the Saul A Silverman Family Foundation Toronto Canadaas a Canada-International Scientific exchange program (CISEPO) project

References

Alcock J E (1987) ParapsychologyScience of the Anomalous or Search for the Soul Behavioraland Brain Sciences 10 553ndash643

Carver C S amp Scheier M F (1991) Self-Regulation and the Self In J Strauss amp G R Goethals(Eds) The Self Interdisciplinary Approaches (pp 168ndash207) New York Springer-Verlag

Hall J Kim C McElroy B amp Shimony A (1977) Wave-packet reduction as a medium ofcommunication Foundations of Physics 7 759ndash767

Ibison M amp Jeffers S (1998) A Double-Slit Diffraction Experiment to Investigate Claims ofConsciousness-Related Anomalies Journal of Scienti c Exploration 12 543ndash550

Jahn R G amp Dunne B J (1986) On the Quantum Mechanics of Consciousness with Applicationto Anomalous Phenomena Foundations of Physics 16 721ndash772

Jahn R G amp Dunne B J (1987a) Margins of Reality The Role of Consciousness in the PhysicalWorld San Diego Harcourt Brace Jovanovich

Jahn R G Dunne B J amp Nelson R D (1987b) Engineering Anomalies Research Journal ofScienti c Exploration 1 21ndash50

Effects of Frontal Lobe Lesions on Intentionality 667

significance levels was greater in the pseudodata than in the real data Howevernone of the effects were in the direction of intention and in three instances thesignificant effects occurred during the baseline intention

Replication Study

The primary analyses showed a statistically significant effect in the direc-tion of intention to the right for the patient with a left frontal lesion This effectwas significant even after Bonferroni correction for multiple comparisons Toexamine the possible effects in this patient further we carried out a second studyto determine whether the findings could be replicated The planned hypothesiswas that he would show a statistically significant effect of intention to the rightbut not to the left or in the baseline intention (ie that the findings would be thesame as before) The left frontal subject was retested using the methodologydescribed for the original study

During testing a printer error (data was always printed after each block of 100trials) occurred at the end of the fifth block of 100 trials during a baseline sessionand the program exited while the bar was still on the screen Although thecomplete data for these 100 trials were still captured a replacement set of 1000baseline and control trials for the entire intention was collected on a separateday However the possible impact of the printer error is considered negligibledue to the small number of trials involved the fact that none of the data werelost and because it occurred during the baseline condition in which the subjectwas asked not to concentrate on moving the arrow On another occasiona baseline session should have preceded a right intention session but the order was

TABLE 3Analysis of Frontal and Normal Subjects (Intention vs Control Conditions)

Subjects Intention t df p-value

Bilateral frontals right 11920 200898 023left 01213 200998 090baseline iexcl13874 199998 017

Left frontal right iexcl31691 5998 00015R

left 00524 7998 096baseline iexcl03656 5998 071

Right frontal right 00375 1998 097left iexcl08754 1998 038baseline iexcl16084 1998 011

Frontals pooled right 06377 208898 052left 00441 210998 096baseline iexcl15794 207998 011

Normals right iexcl08538 95998 039left 11474 91998 025baseline iexcl07036 91998 048

Note R ˆ Direction of effect was to the right

Effects of Frontal Lobe Lesions on Intentionality 659

reversed This resulted in two right intention and two baseline sessions being runconsecutively The right intention sessions were on the same day whereas thebaseline sessions were on different days Another occurrence consisted of thesubject seeing double arrows during a right intention session Finally during a left

TABLE 4Single Subject Analysis

Mean REG Output

Subject Intention Intention Control t p-value Trials (n)

S1 right 9998 10001 04982 062 40000left 9997 10003 08491 040 40000baseline 10003 10004 02000 084 40000

S2 right 9999 10007 08302 041 22900a

left 9988 9991 03145 075 24000baseline 10003 10001 iexcl01708 086 24000

S3 right 9989 10002 17414 008 34000left 10011 10000 iexcl13408 018 33000baseline 10013 10004 iexcl10462 030 32000

S4 right 10002 10002 iexcl00341 097 104000left 9999 10000 02488 080 104000baseline 10003 9997 iexcl13896 016 104000

S5 right 10019 9962 iexcl31691 00015R 6000left 9999 10000 00524 096 8000baseline 10001 9994 iexcl03656 071 6000

S6 right 10007 10008 00375 097 2000left 9991 9964 iexcl08754 038 2000baseline 10009 9959 iexcl16084 011 2000

S7 right 9984 9991 06708 050 16000left 10002 9998 iexcl03481 073 16000baseline 9994 9994 iexcl00022 100 16000

S8 right 10000 10006 07071 048 24000left 10003 10002 iexcl01489 088 24000baseline 10009 9989 iexcl21274 00334R 24000

S9 right 10021 9991 iexcl28790 00040R 18000left 10000 10012 10737 028 16000baseline 9994 9999 04375 066 16000b

S10 right 10001 10004 01845 085 6000left 9988 10014 11721 024 4000baseline 9974 9990 07088 048 4000

S11 right 10002 10002 iexcl00477 096 16000left 9999 10005 06132 054 16000baseline 10005 10004 iexcl01077 091 16000

S12 right 10007 10000 iexcl06514 051 16000left 9995 10007 10216 031 16000baseline 10000 10003 02445 081 16000

Note n ˆ number of intention trials Dagger control trials R ˆ direction of effect was to the righta There were 100 fewer intention trials than control trialsb 500 baseline control trials were collected incorrectly and replaced by trials collected at a different

time Although this introduced a potential bias the difference between the intention and controlconditions was too far from significant for a small number of 500 trials to have had an impact

660 M Freedman et al

intention session the building intercom came on at the end of the fourth block of100 trials The subject stated that this affected his concentration only slightly Inall cases described above the complete set of data was used for analysesincluding the replacement trials that followed the printer error (ie no data werediscarded)

TABLE 5aPseudodata Analysis 1

Mean REG Output

Subject Intention Intention Control t p-value Trials (n)

PS1 right 9997 10005 11636 024 40000left 10011 10003 iexcl10648 029 40000baseline 10007 9995 iexcl17586 008 40000

PS2 right 9995 10010 15873 011 22900left 9998 10004 06576 051 24000baseline 9994 10006 14214 016 24000

PS3 right 9996 10002 08405 040 34000left 9991 10002 13635 017 32000a

baseline 10007 9999 iexcl09442 035 32000PS4 right 10003 9999 iexcl10667 029 104000

left 10006 9999 iexcl17887 007 104000baseline 10001 9999 iexcl02984 077 104000

PS5 right 9982 9993 06322 053 6000left 10020 10000 iexcl12473 021 8000baseline 9988 9993 02343 081 6000

PS6 right 9999 10031 10198 031 2000left 9979 10006 08638 039 2000baseline 9963 10031 21474 00319L 2000

PS7 right 10002 10005 02463 081 16000left 9999 10013 12288 022 16000baseline 10005 9999 iexcl05196 060 16000

PS8 right 10006 9996 iexcl11546 025 24000left 10005 10007 02370 081 24000baseline 10000 9997 iexcl02564 080 24000

PS9 right 9995 10011 15199 013 18000left 9996 10002 05973 055 16000baseline 10005 9995 iexcl08764 038 16000

PS10 right 10007 9993 iexcl07168 047 6000left 9997 10016 08457 040 4000baseline 9981 9987 02792 078 4000

PS11 right 10005 10014 08604 039 16000left 10001 10006 04769 063 16000baseline 10009 9997 iexcl10813 028 16000

PS12 right 9993 10010 15528 012 16000left 10001 10012 09288 035 16000baseline 9998 10013 13483 018 16000

Note n ˆ number of intention trials Dagger control trials L ˆ direction of effect was to the lefta Pseudodata contained 500 fewer intention and control trials as compared to real data

Effects of Frontal Lobe Lesions on Intentionality 661

Results of Replication Study

Figure 3 shows the mean REG output for each intention (right left baseline)and the control output for the left frontal subject There were 19000 intentionand control trials for the right intention and 20000 trials for each of the inten-tion and control conditions for the left and baseline intentions

TABLE 5bPseudodata Analysis 2

Mean REG Output

Subject Intention Intention Control t p-value Trials (n)

PS1 right 9995 10001 08327 041 40000left 9996 9995 iexcl02329 082 40000baseline 10003 10000 iexcl04994 062 40000

PS2 right 10001 9995 iexcl06746 050 22900left 10000 9991 iexcl09721 033 24000baseline 9996 9999 03060 076 24000

PS3 right 10011 10008 iexcl04347 066 34000left 9998 10001 03522 072 32000a

baseline 10005 10000 iexcl06143 054 32000PS4 right 9994 10005 26336 00084L 104000

left 10001 9995 iexcl13071 019 104000baseline 9998 9997 iexcl02664 079 104000

PS5 right 9986 10005 10582 029 6000left 10003 10005 01187 091 8000baseline 10002 9995 iexcl03727 071 6000

PS6 right 9968 9993 07972 043 2000left 9990 9976 iexcl04402 066 2000baseline 10008 9960 iexcl15583 012 2000

PS7 right 9990 9995 04285 067 16000left 10018 10002 iexcl14678 014 16000baseline 9983 9996 11654 024 16000

PS8 right 10008 10005 iexcl02559 080 24000left 10016 9984 iexcl35151 00004R 24000baseline 10002 10002 iexcl00648 095 24000

PS9 right 10001 9998 iexcl02441 081 18000left 10017 9998 iexcl16610 010 16000baseline 9993 9983 iexcl08404 040 16000

PS10 right 10009 10014 02525 080 6000left 10018 10000 iexcl07789 044 4000baseline 9987 10022 15658 012 4000

PS11 right 10006 10010 03433 073 16000left 9992 10000 07129 048 16000baseline 10007 9986 iexcl19258 00541R 16000

PS12 right 9997 9986 iexcl10275 030 16000left 10002 9999 iexcl02967 077 16000baseline 9978 10001 20961 00361L 16000

Note n ˆ number of intention trials Dagger control trials R ˆ direction of effect was to the right L ˆdirection of effect was to the left

a Pseudodata contained 500 fewer intention and control trials as compared to real data

662 M Freedman et al

T-tests were carried out using SAS System for Windows Release 8 todetermine whether there were significant differences between the intention andcontrol conditions for the right left and baseline intentions There wasa significant difference between the intention and control conditions for the rightintention (t[37998]ˆiexcl253 pˆ 00115) but not for the other two intentions (leftt[39998] ˆ 049 p ˆ 06269 and baseline t[39998] ˆ iexcl007 p ˆ 09418) Thesignificant effect for the right intention was in the direction of intention as in theoriginal study

Since the investigation was run in blocks of 1000 trials we examined the datausing blocks as the unit of measurement and the averaged output across trials asthe measure in each block An F-ratio test with condition (intention vs control)as the source of variance of interest and the interaction between condition andblock as the error term showed an effect of condition (p ˆ 00578) This suggeststhat the left frontal subject would show comparable findings on another series of19 pairs of blocks of 1000 trials with 94 confidence

Whereas the mean REG output for the right intention was above the expectedvalue of 100 the output for the control condition was below 100 (Figure 3) Toexamine the possibility that the significant difference between the intention andcontrol condition was due solely to low output during the control condition asopposed to high output during the intention condition we tested whether theoutput for the intention condition was significantly different from a constantvalue of 100 This value is the theoretical mean output of the REG assumingtruly random and unbiased output The control values for the left and baselineconditions are approximately equal to this theoretical mean The REG output forthe right intention was significantly different than 100 and in the direction ofintention (t ˆ 201 p ˆ 0045 two-tailed test) For the control condition themean REG output was less than (although not significantly different from) 100(t ˆ iexcl156 p ˆ 012)

Fig 3 Mean REG output for the left frontal patient in replication study

Effects of Frontal Lobe Lesions on Intentionality 663

Figure 4 (andashc) shows the mean output of the REG broken down byconsecutive 1000 trial sessions for each intention (right left and baseline) andfor the control output For the right intention Figure 4 shows that on most pairsof blocks of 1000 trials the mean REG output follows a fairly consistent patternin which the means are either higher compared to the control condition or aboutthe same In contrast the data for each of the left and baseline conditions showa pattern with less separation between conditions

Fig 4 Mean REG output for each 1000 trial session in replication study (a) Intention Right (b)Intention Left and (c) Intention Baseline

664 M Freedman et al

Discussion

We examined the claims from the PEAR program that an individualrsquosintentions can influence the statistical distribution of random physical phenom-ena Our main focus was to test these claims using well-designed control condi-tions in a population that might maximize the likelihood of detecting any effectsthat may exist Our choice of studying patients with frontal lobe lesions wasbased on the concept that the potential to influence random physical events maybe optimized when attention is diverted from self-awareness (ie such as whenself-awareness is reduced) This state can occur following frontal brain damage(Stuss et al 1986 Stuss 1991 Carver et al 1991) Our findings showed asignificant effect for the right intention in a subject with left frontal braindamage Moreover this effect was in the direction of intention In contrast therewere no significant effects in the other groups (ie bilateral frontal right frontalpooled frontal or normal subjects)

Although the result in the left frontal patient on the right intention wasstatistically significant even after correction for multiple comparisons thisfinding was interpreted with great caution First it was based on a relativelysmall number of intention trials (n ˆ 3000) and was derived from only onesubject Second we found a p-value for an individual lsquolsquosubjectrsquorsquo in the pseu-dodata that would also have met criteria for significance using a Bonferronicorrection and that was even less than the p-value of 00015 obtained fromthe left frontal subject Although the effect of this lsquolsquopseudopatientrsquorsquo was not inthe direction of intention the fact that it was significant raised caution forinterpreting the effect in the left frontal patient However the replication of thefindings in the left frontal patient in a second well-designed study for each of the

Fig 4 (Continued)

Effects of Frontal Lobe Lesions on Intentionality 665

three intentions right left and baseline suggests that the effect in this subjectmay be more than chance occurrence

Additional support for a reliable effect in the left frontal subject comes fromfurther analysis of his data suggesting about a 94 chance of his showinga similar finding if another series of 19 pairs of blocks of 1000 trials were runagain Furthermore the profile of REG output data for the right intentionshowed a fairly consistent separation whereas this was not the case for the left orbaseline intentions

A comment is warranted about the REG output for the left frontal patientbeing lower for the control condition on the right intention or more to the leftas compared to the control conditions for the left or baseline intentions Onemight argue that the significant effect on the right intention was an artifact ofcomparison to control data that was in the left direction and that this widened thedifference between the intention and control data However this argument is nottenable because the effect on the right intention was significant even when theREG output was compared to a theoretical mean of 100mdasha value which is ap-proximately equal to the control means for the left intention (9999985) and thebaseline intention (1000011)

The patientrsquos cognitive deficits and brain lesion have been describedelsewhere (Marras et al 1998) He suffered from a tension pneumocephaluswhich resulted in cognitive deficits and epileptic seizures The cognitive deficitsinclude decreased mental flexibility on the Trail-Making Test poor attentionreduced fluency and impaired spatial planning and visuospatial problem solvingMRI showed an extensive left frontal lesion but the right frontal lobe was intactPsychometric testing and SPECT suggested the addition of right frontal dysfunc-tion The SPECT findings provide a measure of function as opposed to structureand were subtle

As indicated above frontal lesions have been associated with reduced self-awareness (Stuss et al 1986 Stuss 1991 Carver et al 1991) a state that isdifficult for normal individuals to achieve The rationale for studying patientswith damage to the frontal lobes was that decreased self-awareness might facili-tate the effects of intentionality on random physical phenomena Brain regionsthat mediate neurological processes underlying self-awareness include thefrontal lobes bilaterally particularly on the right (Stuss et al 2001a Stuss ampAlexander 2000a Stuss amp Alexander 2000b Stuss et al 2001b) Whereas thepatientrsquos right-sided brain dysfunction may have been insufficient to producea deficit in self-awareness the extensive lesion on the left may have resulted inreduced self-awareness when attention was directed towards the right Howeverit remains unclear why positive results should be found only following damageto the left frontal region and not after bilateral or right frontal lesions Onespeculation is that the effect on random physical events may require reducedself-awareness combined with relatively intact attentional mechanisms Theassociation of frontal lobe lesions especially on the right with impaired at-tention (Stuss amp Levine 2002) may explain the negative findings in the setting

666 M Freedman et al

of bilateral or right frontal damage Whether deficits related to left frontalabnormalities can explain the observed effects on the REG output warrantsfurther study Moreover the question whether normal processes associated withintact frontal lobe function together with preserved self-awareness may serve toinhibit effects of intentionality on random physical phenomena needs to beaddressed

The strength of our conclusions rests largely on a well-designed methodol-ogy and replication of our findings Although our results did not replicate thefindings reported by Jahn and his colleagues in normal subjects (Jahn et al1987a 1997) they support their claims that intentionality can alter the output ofa random event generator Furthermore our findings suggest that patients withfrontal lobe lesions may serve as a good model for future studies of the effects ofconsciousness on random physical events

Acknowledgments

We gratefully acknowledge Dr Donald Stuss for his helpful advice andsupport throughout the study and for reviewing the manuscript and for hishelpful suggestions Dr Louis Siminovitch for helpful advice and supportPatricia Ebert and Dr Donald Stuss for assisting with analysis of lesion local-ization for the CT and MRI scans Janice Pogue for statistical advice LeeFerreira for testing the subjects John Sloan for assistance with data collec-tion Tania Xerri for assistance with data analyses and Tania Xerri MaedeanBrown Wanna Mar and Melissa Ziraldo for preparation of materials for pub-lication We also gratefully acknowledge Professor Robert Jahn BrendaDunne Dr Roger Nelson Dr York Dobyns and John Bradish at the PEARlab for their helpful advice and for providing the portable REG and softwarefor the study This study was supported by the Fetzer Institute Dr Freedmanwas supported by the Saul A Silverman Family Foundation Toronto Canadaas a Canada-International Scientific exchange program (CISEPO) project

References

Alcock J E (1987) ParapsychologyScience of the Anomalous or Search for the Soul Behavioraland Brain Sciences 10 553ndash643

Carver C S amp Scheier M F (1991) Self-Regulation and the Self In J Strauss amp G R Goethals(Eds) The Self Interdisciplinary Approaches (pp 168ndash207) New York Springer-Verlag

Hall J Kim C McElroy B amp Shimony A (1977) Wave-packet reduction as a medium ofcommunication Foundations of Physics 7 759ndash767

Ibison M amp Jeffers S (1998) A Double-Slit Diffraction Experiment to Investigate Claims ofConsciousness-Related Anomalies Journal of Scienti c Exploration 12 543ndash550

Jahn R G amp Dunne B J (1986) On the Quantum Mechanics of Consciousness with Applicationto Anomalous Phenomena Foundations of Physics 16 721ndash772

Jahn R G amp Dunne B J (1987a) Margins of Reality The Role of Consciousness in the PhysicalWorld San Diego Harcourt Brace Jovanovich

Jahn R G Dunne B J amp Nelson R D (1987b) Engineering Anomalies Research Journal ofScienti c Exploration 1 21ndash50

Effects of Frontal Lobe Lesions on Intentionality 667

reversed This resulted in two right intention and two baseline sessions being runconsecutively The right intention sessions were on the same day whereas thebaseline sessions were on different days Another occurrence consisted of thesubject seeing double arrows during a right intention session Finally during a left

TABLE 4Single Subject Analysis

Mean REG Output

Subject Intention Intention Control t p-value Trials (n)

S1 right 9998 10001 04982 062 40000left 9997 10003 08491 040 40000baseline 10003 10004 02000 084 40000

S2 right 9999 10007 08302 041 22900a

left 9988 9991 03145 075 24000baseline 10003 10001 iexcl01708 086 24000

S3 right 9989 10002 17414 008 34000left 10011 10000 iexcl13408 018 33000baseline 10013 10004 iexcl10462 030 32000

S4 right 10002 10002 iexcl00341 097 104000left 9999 10000 02488 080 104000baseline 10003 9997 iexcl13896 016 104000

S5 right 10019 9962 iexcl31691 00015R 6000left 9999 10000 00524 096 8000baseline 10001 9994 iexcl03656 071 6000

S6 right 10007 10008 00375 097 2000left 9991 9964 iexcl08754 038 2000baseline 10009 9959 iexcl16084 011 2000

S7 right 9984 9991 06708 050 16000left 10002 9998 iexcl03481 073 16000baseline 9994 9994 iexcl00022 100 16000

S8 right 10000 10006 07071 048 24000left 10003 10002 iexcl01489 088 24000baseline 10009 9989 iexcl21274 00334R 24000

S9 right 10021 9991 iexcl28790 00040R 18000left 10000 10012 10737 028 16000baseline 9994 9999 04375 066 16000b

S10 right 10001 10004 01845 085 6000left 9988 10014 11721 024 4000baseline 9974 9990 07088 048 4000

S11 right 10002 10002 iexcl00477 096 16000left 9999 10005 06132 054 16000baseline 10005 10004 iexcl01077 091 16000

S12 right 10007 10000 iexcl06514 051 16000left 9995 10007 10216 031 16000baseline 10000 10003 02445 081 16000

Note n ˆ number of intention trials Dagger control trials R ˆ direction of effect was to the righta There were 100 fewer intention trials than control trialsb 500 baseline control trials were collected incorrectly and replaced by trials collected at a different

time Although this introduced a potential bias the difference between the intention and controlconditions was too far from significant for a small number of 500 trials to have had an impact

660 M Freedman et al

intention session the building intercom came on at the end of the fourth block of100 trials The subject stated that this affected his concentration only slightly Inall cases described above the complete set of data was used for analysesincluding the replacement trials that followed the printer error (ie no data werediscarded)

TABLE 5aPseudodata Analysis 1

Mean REG Output

Subject Intention Intention Control t p-value Trials (n)

PS1 right 9997 10005 11636 024 40000left 10011 10003 iexcl10648 029 40000baseline 10007 9995 iexcl17586 008 40000

PS2 right 9995 10010 15873 011 22900left 9998 10004 06576 051 24000baseline 9994 10006 14214 016 24000

PS3 right 9996 10002 08405 040 34000left 9991 10002 13635 017 32000a

baseline 10007 9999 iexcl09442 035 32000PS4 right 10003 9999 iexcl10667 029 104000

left 10006 9999 iexcl17887 007 104000baseline 10001 9999 iexcl02984 077 104000

PS5 right 9982 9993 06322 053 6000left 10020 10000 iexcl12473 021 8000baseline 9988 9993 02343 081 6000

PS6 right 9999 10031 10198 031 2000left 9979 10006 08638 039 2000baseline 9963 10031 21474 00319L 2000

PS7 right 10002 10005 02463 081 16000left 9999 10013 12288 022 16000baseline 10005 9999 iexcl05196 060 16000

PS8 right 10006 9996 iexcl11546 025 24000left 10005 10007 02370 081 24000baseline 10000 9997 iexcl02564 080 24000

PS9 right 9995 10011 15199 013 18000left 9996 10002 05973 055 16000baseline 10005 9995 iexcl08764 038 16000

PS10 right 10007 9993 iexcl07168 047 6000left 9997 10016 08457 040 4000baseline 9981 9987 02792 078 4000

PS11 right 10005 10014 08604 039 16000left 10001 10006 04769 063 16000baseline 10009 9997 iexcl10813 028 16000

PS12 right 9993 10010 15528 012 16000left 10001 10012 09288 035 16000baseline 9998 10013 13483 018 16000

Note n ˆ number of intention trials Dagger control trials L ˆ direction of effect was to the lefta Pseudodata contained 500 fewer intention and control trials as compared to real data

Effects of Frontal Lobe Lesions on Intentionality 661

Results of Replication Study

Figure 3 shows the mean REG output for each intention (right left baseline)and the control output for the left frontal subject There were 19000 intentionand control trials for the right intention and 20000 trials for each of the inten-tion and control conditions for the left and baseline intentions

TABLE 5bPseudodata Analysis 2

Mean REG Output

Subject Intention Intention Control t p-value Trials (n)

PS1 right 9995 10001 08327 041 40000left 9996 9995 iexcl02329 082 40000baseline 10003 10000 iexcl04994 062 40000

PS2 right 10001 9995 iexcl06746 050 22900left 10000 9991 iexcl09721 033 24000baseline 9996 9999 03060 076 24000

PS3 right 10011 10008 iexcl04347 066 34000left 9998 10001 03522 072 32000a

baseline 10005 10000 iexcl06143 054 32000PS4 right 9994 10005 26336 00084L 104000

left 10001 9995 iexcl13071 019 104000baseline 9998 9997 iexcl02664 079 104000

PS5 right 9986 10005 10582 029 6000left 10003 10005 01187 091 8000baseline 10002 9995 iexcl03727 071 6000

PS6 right 9968 9993 07972 043 2000left 9990 9976 iexcl04402 066 2000baseline 10008 9960 iexcl15583 012 2000

PS7 right 9990 9995 04285 067 16000left 10018 10002 iexcl14678 014 16000baseline 9983 9996 11654 024 16000

PS8 right 10008 10005 iexcl02559 080 24000left 10016 9984 iexcl35151 00004R 24000baseline 10002 10002 iexcl00648 095 24000

PS9 right 10001 9998 iexcl02441 081 18000left 10017 9998 iexcl16610 010 16000baseline 9993 9983 iexcl08404 040 16000

PS10 right 10009 10014 02525 080 6000left 10018 10000 iexcl07789 044 4000baseline 9987 10022 15658 012 4000

PS11 right 10006 10010 03433 073 16000left 9992 10000 07129 048 16000baseline 10007 9986 iexcl19258 00541R 16000

PS12 right 9997 9986 iexcl10275 030 16000left 10002 9999 iexcl02967 077 16000baseline 9978 10001 20961 00361L 16000

Note n ˆ number of intention trials Dagger control trials R ˆ direction of effect was to the right L ˆdirection of effect was to the left

a Pseudodata contained 500 fewer intention and control trials as compared to real data

662 M Freedman et al

T-tests were carried out using SAS System for Windows Release 8 todetermine whether there were significant differences between the intention andcontrol conditions for the right left and baseline intentions There wasa significant difference between the intention and control conditions for the rightintention (t[37998]ˆiexcl253 pˆ 00115) but not for the other two intentions (leftt[39998] ˆ 049 p ˆ 06269 and baseline t[39998] ˆ iexcl007 p ˆ 09418) Thesignificant effect for the right intention was in the direction of intention as in theoriginal study

Since the investigation was run in blocks of 1000 trials we examined the datausing blocks as the unit of measurement and the averaged output across trials asthe measure in each block An F-ratio test with condition (intention vs control)as the source of variance of interest and the interaction between condition andblock as the error term showed an effect of condition (p ˆ 00578) This suggeststhat the left frontal subject would show comparable findings on another series of19 pairs of blocks of 1000 trials with 94 confidence

Whereas the mean REG output for the right intention was above the expectedvalue of 100 the output for the control condition was below 100 (Figure 3) Toexamine the possibility that the significant difference between the intention andcontrol condition was due solely to low output during the control condition asopposed to high output during the intention condition we tested whether theoutput for the intention condition was significantly different from a constantvalue of 100 This value is the theoretical mean output of the REG assumingtruly random and unbiased output The control values for the left and baselineconditions are approximately equal to this theoretical mean The REG output forthe right intention was significantly different than 100 and in the direction ofintention (t ˆ 201 p ˆ 0045 two-tailed test) For the control condition themean REG output was less than (although not significantly different from) 100(t ˆ iexcl156 p ˆ 012)

Fig 3 Mean REG output for the left frontal patient in replication study

Effects of Frontal Lobe Lesions on Intentionality 663

Figure 4 (andashc) shows the mean output of the REG broken down byconsecutive 1000 trial sessions for each intention (right left and baseline) andfor the control output For the right intention Figure 4 shows that on most pairsof blocks of 1000 trials the mean REG output follows a fairly consistent patternin which the means are either higher compared to the control condition or aboutthe same In contrast the data for each of the left and baseline conditions showa pattern with less separation between conditions

Fig 4 Mean REG output for each 1000 trial session in replication study (a) Intention Right (b)Intention Left and (c) Intention Baseline

664 M Freedman et al

Discussion

We examined the claims from the PEAR program that an individualrsquosintentions can influence the statistical distribution of random physical phenom-ena Our main focus was to test these claims using well-designed control condi-tions in a population that might maximize the likelihood of detecting any effectsthat may exist Our choice of studying patients with frontal lobe lesions wasbased on the concept that the potential to influence random physical events maybe optimized when attention is diverted from self-awareness (ie such as whenself-awareness is reduced) This state can occur following frontal brain damage(Stuss et al 1986 Stuss 1991 Carver et al 1991) Our findings showed asignificant effect for the right intention in a subject with left frontal braindamage Moreover this effect was in the direction of intention In contrast therewere no significant effects in the other groups (ie bilateral frontal right frontalpooled frontal or normal subjects)

Although the result in the left frontal patient on the right intention wasstatistically significant even after correction for multiple comparisons thisfinding was interpreted with great caution First it was based on a relativelysmall number of intention trials (n ˆ 3000) and was derived from only onesubject Second we found a p-value for an individual lsquolsquosubjectrsquorsquo in the pseu-dodata that would also have met criteria for significance using a Bonferronicorrection and that was even less than the p-value of 00015 obtained fromthe left frontal subject Although the effect of this lsquolsquopseudopatientrsquorsquo was not inthe direction of intention the fact that it was significant raised caution forinterpreting the effect in the left frontal patient However the replication of thefindings in the left frontal patient in a second well-designed study for each of the

Fig 4 (Continued)

Effects of Frontal Lobe Lesions on Intentionality 665

three intentions right left and baseline suggests that the effect in this subjectmay be more than chance occurrence

Additional support for a reliable effect in the left frontal subject comes fromfurther analysis of his data suggesting about a 94 chance of his showinga similar finding if another series of 19 pairs of blocks of 1000 trials were runagain Furthermore the profile of REG output data for the right intentionshowed a fairly consistent separation whereas this was not the case for the left orbaseline intentions

A comment is warranted about the REG output for the left frontal patientbeing lower for the control condition on the right intention or more to the leftas compared to the control conditions for the left or baseline intentions Onemight argue that the significant effect on the right intention was an artifact ofcomparison to control data that was in the left direction and that this widened thedifference between the intention and control data However this argument is nottenable because the effect on the right intention was significant even when theREG output was compared to a theoretical mean of 100mdasha value which is ap-proximately equal to the control means for the left intention (9999985) and thebaseline intention (1000011)

The patientrsquos cognitive deficits and brain lesion have been describedelsewhere (Marras et al 1998) He suffered from a tension pneumocephaluswhich resulted in cognitive deficits and epileptic seizures The cognitive deficitsinclude decreased mental flexibility on the Trail-Making Test poor attentionreduced fluency and impaired spatial planning and visuospatial problem solvingMRI showed an extensive left frontal lesion but the right frontal lobe was intactPsychometric testing and SPECT suggested the addition of right frontal dysfunc-tion The SPECT findings provide a measure of function as opposed to structureand were subtle

As indicated above frontal lesions have been associated with reduced self-awareness (Stuss et al 1986 Stuss 1991 Carver et al 1991) a state that isdifficult for normal individuals to achieve The rationale for studying patientswith damage to the frontal lobes was that decreased self-awareness might facili-tate the effects of intentionality on random physical phenomena Brain regionsthat mediate neurological processes underlying self-awareness include thefrontal lobes bilaterally particularly on the right (Stuss et al 2001a Stuss ampAlexander 2000a Stuss amp Alexander 2000b Stuss et al 2001b) Whereas thepatientrsquos right-sided brain dysfunction may have been insufficient to producea deficit in self-awareness the extensive lesion on the left may have resulted inreduced self-awareness when attention was directed towards the right Howeverit remains unclear why positive results should be found only following damageto the left frontal region and not after bilateral or right frontal lesions Onespeculation is that the effect on random physical events may require reducedself-awareness combined with relatively intact attentional mechanisms Theassociation of frontal lobe lesions especially on the right with impaired at-tention (Stuss amp Levine 2002) may explain the negative findings in the setting

666 M Freedman et al

of bilateral or right frontal damage Whether deficits related to left frontalabnormalities can explain the observed effects on the REG output warrantsfurther study Moreover the question whether normal processes associated withintact frontal lobe function together with preserved self-awareness may serve toinhibit effects of intentionality on random physical phenomena needs to beaddressed

The strength of our conclusions rests largely on a well-designed methodol-ogy and replication of our findings Although our results did not replicate thefindings reported by Jahn and his colleagues in normal subjects (Jahn et al1987a 1997) they support their claims that intentionality can alter the output ofa random event generator Furthermore our findings suggest that patients withfrontal lobe lesions may serve as a good model for future studies of the effects ofconsciousness on random physical events

Acknowledgments

We gratefully acknowledge Dr Donald Stuss for his helpful advice andsupport throughout the study and for reviewing the manuscript and for hishelpful suggestions Dr Louis Siminovitch for helpful advice and supportPatricia Ebert and Dr Donald Stuss for assisting with analysis of lesion local-ization for the CT and MRI scans Janice Pogue for statistical advice LeeFerreira for testing the subjects John Sloan for assistance with data collec-tion Tania Xerri for assistance with data analyses and Tania Xerri MaedeanBrown Wanna Mar and Melissa Ziraldo for preparation of materials for pub-lication We also gratefully acknowledge Professor Robert Jahn BrendaDunne Dr Roger Nelson Dr York Dobyns and John Bradish at the PEARlab for their helpful advice and for providing the portable REG and softwarefor the study This study was supported by the Fetzer Institute Dr Freedmanwas supported by the Saul A Silverman Family Foundation Toronto Canadaas a Canada-International Scientific exchange program (CISEPO) project

References

Alcock J E (1987) ParapsychologyScience of the Anomalous or Search for the Soul Behavioraland Brain Sciences 10 553ndash643

Carver C S amp Scheier M F (1991) Self-Regulation and the Self In J Strauss amp G R Goethals(Eds) The Self Interdisciplinary Approaches (pp 168ndash207) New York Springer-Verlag

Hall J Kim C McElroy B amp Shimony A (1977) Wave-packet reduction as a medium ofcommunication Foundations of Physics 7 759ndash767

Ibison M amp Jeffers S (1998) A Double-Slit Diffraction Experiment to Investigate Claims ofConsciousness-Related Anomalies Journal of Scienti c Exploration 12 543ndash550

Jahn R G amp Dunne B J (1986) On the Quantum Mechanics of Consciousness with Applicationto Anomalous Phenomena Foundations of Physics 16 721ndash772

Jahn R G amp Dunne B J (1987a) Margins of Reality The Role of Consciousness in the PhysicalWorld San Diego Harcourt Brace Jovanovich

Jahn R G Dunne B J amp Nelson R D (1987b) Engineering Anomalies Research Journal ofScienti c Exploration 1 21ndash50

Effects of Frontal Lobe Lesions on Intentionality 667

intention session the building intercom came on at the end of the fourth block of100 trials The subject stated that this affected his concentration only slightly Inall cases described above the complete set of data was used for analysesincluding the replacement trials that followed the printer error (ie no data werediscarded)

TABLE 5aPseudodata Analysis 1

Mean REG Output

Subject Intention Intention Control t p-value Trials (n)

PS1 right 9997 10005 11636 024 40000left 10011 10003 iexcl10648 029 40000baseline 10007 9995 iexcl17586 008 40000

PS2 right 9995 10010 15873 011 22900left 9998 10004 06576 051 24000baseline 9994 10006 14214 016 24000

PS3 right 9996 10002 08405 040 34000left 9991 10002 13635 017 32000a

baseline 10007 9999 iexcl09442 035 32000PS4 right 10003 9999 iexcl10667 029 104000

left 10006 9999 iexcl17887 007 104000baseline 10001 9999 iexcl02984 077 104000

PS5 right 9982 9993 06322 053 6000left 10020 10000 iexcl12473 021 8000baseline 9988 9993 02343 081 6000

PS6 right 9999 10031 10198 031 2000left 9979 10006 08638 039 2000baseline 9963 10031 21474 00319L 2000

PS7 right 10002 10005 02463 081 16000left 9999 10013 12288 022 16000baseline 10005 9999 iexcl05196 060 16000

PS8 right 10006 9996 iexcl11546 025 24000left 10005 10007 02370 081 24000baseline 10000 9997 iexcl02564 080 24000

PS9 right 9995 10011 15199 013 18000left 9996 10002 05973 055 16000baseline 10005 9995 iexcl08764 038 16000

PS10 right 10007 9993 iexcl07168 047 6000left 9997 10016 08457 040 4000baseline 9981 9987 02792 078 4000

PS11 right 10005 10014 08604 039 16000left 10001 10006 04769 063 16000baseline 10009 9997 iexcl10813 028 16000

PS12 right 9993 10010 15528 012 16000left 10001 10012 09288 035 16000baseline 9998 10013 13483 018 16000

Note n ˆ number of intention trials Dagger control trials L ˆ direction of effect was to the lefta Pseudodata contained 500 fewer intention and control trials as compared to real data

Effects of Frontal Lobe Lesions on Intentionality 661

Results of Replication Study

Figure 3 shows the mean REG output for each intention (right left baseline)and the control output for the left frontal subject There were 19000 intentionand control trials for the right intention and 20000 trials for each of the inten-tion and control conditions for the left and baseline intentions

TABLE 5bPseudodata Analysis 2

Mean REG Output

Subject Intention Intention Control t p-value Trials (n)

PS1 right 9995 10001 08327 041 40000left 9996 9995 iexcl02329 082 40000baseline 10003 10000 iexcl04994 062 40000

PS2 right 10001 9995 iexcl06746 050 22900left 10000 9991 iexcl09721 033 24000baseline 9996 9999 03060 076 24000

PS3 right 10011 10008 iexcl04347 066 34000left 9998 10001 03522 072 32000a

baseline 10005 10000 iexcl06143 054 32000PS4 right 9994 10005 26336 00084L 104000

left 10001 9995 iexcl13071 019 104000baseline 9998 9997 iexcl02664 079 104000

PS5 right 9986 10005 10582 029 6000left 10003 10005 01187 091 8000baseline 10002 9995 iexcl03727 071 6000

PS6 right 9968 9993 07972 043 2000left 9990 9976 iexcl04402 066 2000baseline 10008 9960 iexcl15583 012 2000

PS7 right 9990 9995 04285 067 16000left 10018 10002 iexcl14678 014 16000baseline 9983 9996 11654 024 16000

PS8 right 10008 10005 iexcl02559 080 24000left 10016 9984 iexcl35151 00004R 24000baseline 10002 10002 iexcl00648 095 24000

PS9 right 10001 9998 iexcl02441 081 18000left 10017 9998 iexcl16610 010 16000baseline 9993 9983 iexcl08404 040 16000

PS10 right 10009 10014 02525 080 6000left 10018 10000 iexcl07789 044 4000baseline 9987 10022 15658 012 4000

PS11 right 10006 10010 03433 073 16000left 9992 10000 07129 048 16000baseline 10007 9986 iexcl19258 00541R 16000

PS12 right 9997 9986 iexcl10275 030 16000left 10002 9999 iexcl02967 077 16000baseline 9978 10001 20961 00361L 16000

Note n ˆ number of intention trials Dagger control trials R ˆ direction of effect was to the right L ˆdirection of effect was to the left

a Pseudodata contained 500 fewer intention and control trials as compared to real data

662 M Freedman et al

T-tests were carried out using SAS System for Windows Release 8 todetermine whether there were significant differences between the intention andcontrol conditions for the right left and baseline intentions There wasa significant difference between the intention and control conditions for the rightintention (t[37998]ˆiexcl253 pˆ 00115) but not for the other two intentions (leftt[39998] ˆ 049 p ˆ 06269 and baseline t[39998] ˆ iexcl007 p ˆ 09418) Thesignificant effect for the right intention was in the direction of intention as in theoriginal study

Since the investigation was run in blocks of 1000 trials we examined the datausing blocks as the unit of measurement and the averaged output across trials asthe measure in each block An F-ratio test with condition (intention vs control)as the source of variance of interest and the interaction between condition andblock as the error term showed an effect of condition (p ˆ 00578) This suggeststhat the left frontal subject would show comparable findings on another series of19 pairs of blocks of 1000 trials with 94 confidence

Whereas the mean REG output for the right intention was above the expectedvalue of 100 the output for the control condition was below 100 (Figure 3) Toexamine the possibility that the significant difference between the intention andcontrol condition was due solely to low output during the control condition asopposed to high output during the intention condition we tested whether theoutput for the intention condition was significantly different from a constantvalue of 100 This value is the theoretical mean output of the REG assumingtruly random and unbiased output The control values for the left and baselineconditions are approximately equal to this theoretical mean The REG output forthe right intention was significantly different than 100 and in the direction ofintention (t ˆ 201 p ˆ 0045 two-tailed test) For the control condition themean REG output was less than (although not significantly different from) 100(t ˆ iexcl156 p ˆ 012)

Fig 3 Mean REG output for the left frontal patient in replication study

Effects of Frontal Lobe Lesions on Intentionality 663

Figure 4 (andashc) shows the mean output of the REG broken down byconsecutive 1000 trial sessions for each intention (right left and baseline) andfor the control output For the right intention Figure 4 shows that on most pairsof blocks of 1000 trials the mean REG output follows a fairly consistent patternin which the means are either higher compared to the control condition or aboutthe same In contrast the data for each of the left and baseline conditions showa pattern with less separation between conditions

Fig 4 Mean REG output for each 1000 trial session in replication study (a) Intention Right (b)Intention Left and (c) Intention Baseline

664 M Freedman et al

Discussion

We examined the claims from the PEAR program that an individualrsquosintentions can influence the statistical distribution of random physical phenom-ena Our main focus was to test these claims using well-designed control condi-tions in a population that might maximize the likelihood of detecting any effectsthat may exist Our choice of studying patients with frontal lobe lesions wasbased on the concept that the potential to influence random physical events maybe optimized when attention is diverted from self-awareness (ie such as whenself-awareness is reduced) This state can occur following frontal brain damage(Stuss et al 1986 Stuss 1991 Carver et al 1991) Our findings showed asignificant effect for the right intention in a subject with left frontal braindamage Moreover this effect was in the direction of intention In contrast therewere no significant effects in the other groups (ie bilateral frontal right frontalpooled frontal or normal subjects)

Although the result in the left frontal patient on the right intention wasstatistically significant even after correction for multiple comparisons thisfinding was interpreted with great caution First it was based on a relativelysmall number of intention trials (n ˆ 3000) and was derived from only onesubject Second we found a p-value for an individual lsquolsquosubjectrsquorsquo in the pseu-dodata that would also have met criteria for significance using a Bonferronicorrection and that was even less than the p-value of 00015 obtained fromthe left frontal subject Although the effect of this lsquolsquopseudopatientrsquorsquo was not inthe direction of intention the fact that it was significant raised caution forinterpreting the effect in the left frontal patient However the replication of thefindings in the left frontal patient in a second well-designed study for each of the

Fig 4 (Continued)

Effects of Frontal Lobe Lesions on Intentionality 665

three intentions right left and baseline suggests that the effect in this subjectmay be more than chance occurrence

Additional support for a reliable effect in the left frontal subject comes fromfurther analysis of his data suggesting about a 94 chance of his showinga similar finding if another series of 19 pairs of blocks of 1000 trials were runagain Furthermore the profile of REG output data for the right intentionshowed a fairly consistent separation whereas this was not the case for the left orbaseline intentions

A comment is warranted about the REG output for the left frontal patientbeing lower for the control condition on the right intention or more to the leftas compared to the control conditions for the left or baseline intentions Onemight argue that the significant effect on the right intention was an artifact ofcomparison to control data that was in the left direction and that this widened thedifference between the intention and control data However this argument is nottenable because the effect on the right intention was significant even when theREG output was compared to a theoretical mean of 100mdasha value which is ap-proximately equal to the control means for the left intention (9999985) and thebaseline intention (1000011)

The patientrsquos cognitive deficits and brain lesion have been describedelsewhere (Marras et al 1998) He suffered from a tension pneumocephaluswhich resulted in cognitive deficits and epileptic seizures The cognitive deficitsinclude decreased mental flexibility on the Trail-Making Test poor attentionreduced fluency and impaired spatial planning and visuospatial problem solvingMRI showed an extensive left frontal lesion but the right frontal lobe was intactPsychometric testing and SPECT suggested the addition of right frontal dysfunc-tion The SPECT findings provide a measure of function as opposed to structureand were subtle

As indicated above frontal lesions have been associated with reduced self-awareness (Stuss et al 1986 Stuss 1991 Carver et al 1991) a state that isdifficult for normal individuals to achieve The rationale for studying patientswith damage to the frontal lobes was that decreased self-awareness might facili-tate the effects of intentionality on random physical phenomena Brain regionsthat mediate neurological processes underlying self-awareness include thefrontal lobes bilaterally particularly on the right (Stuss et al 2001a Stuss ampAlexander 2000a Stuss amp Alexander 2000b Stuss et al 2001b) Whereas thepatientrsquos right-sided brain dysfunction may have been insufficient to producea deficit in self-awareness the extensive lesion on the left may have resulted inreduced self-awareness when attention was directed towards the right Howeverit remains unclear why positive results should be found only following damageto the left frontal region and not after bilateral or right frontal lesions Onespeculation is that the effect on random physical events may require reducedself-awareness combined with relatively intact attentional mechanisms Theassociation of frontal lobe lesions especially on the right with impaired at-tention (Stuss amp Levine 2002) may explain the negative findings in the setting

666 M Freedman et al

of bilateral or right frontal damage Whether deficits related to left frontalabnormalities can explain the observed effects on the REG output warrantsfurther study Moreover the question whether normal processes associated withintact frontal lobe function together with preserved self-awareness may serve toinhibit effects of intentionality on random physical phenomena needs to beaddressed

The strength of our conclusions rests largely on a well-designed methodol-ogy and replication of our findings Although our results did not replicate thefindings reported by Jahn and his colleagues in normal subjects (Jahn et al1987a 1997) they support their claims that intentionality can alter the output ofa random event generator Furthermore our findings suggest that patients withfrontal lobe lesions may serve as a good model for future studies of the effects ofconsciousness on random physical events

Acknowledgments

We gratefully acknowledge Dr Donald Stuss for his helpful advice andsupport throughout the study and for reviewing the manuscript and for hishelpful suggestions Dr Louis Siminovitch for helpful advice and supportPatricia Ebert and Dr Donald Stuss for assisting with analysis of lesion local-ization for the CT and MRI scans Janice Pogue for statistical advice LeeFerreira for testing the subjects John Sloan for assistance with data collec-tion Tania Xerri for assistance with data analyses and Tania Xerri MaedeanBrown Wanna Mar and Melissa Ziraldo for preparation of materials for pub-lication We also gratefully acknowledge Professor Robert Jahn BrendaDunne Dr Roger Nelson Dr York Dobyns and John Bradish at the PEARlab for their helpful advice and for providing the portable REG and softwarefor the study This study was supported by the Fetzer Institute Dr Freedmanwas supported by the Saul A Silverman Family Foundation Toronto Canadaas a Canada-International Scientific exchange program (CISEPO) project

References

Alcock J E (1987) ParapsychologyScience of the Anomalous or Search for the Soul Behavioraland Brain Sciences 10 553ndash643

Carver C S amp Scheier M F (1991) Self-Regulation and the Self In J Strauss amp G R Goethals(Eds) The Self Interdisciplinary Approaches (pp 168ndash207) New York Springer-Verlag

Hall J Kim C McElroy B amp Shimony A (1977) Wave-packet reduction as a medium ofcommunication Foundations of Physics 7 759ndash767

Ibison M amp Jeffers S (1998) A Double-Slit Diffraction Experiment to Investigate Claims ofConsciousness-Related Anomalies Journal of Scienti c Exploration 12 543ndash550

Jahn R G amp Dunne B J (1986) On the Quantum Mechanics of Consciousness with Applicationto Anomalous Phenomena Foundations of Physics 16 721ndash772

Jahn R G amp Dunne B J (1987a) Margins of Reality The Role of Consciousness in the PhysicalWorld San Diego Harcourt Brace Jovanovich

Jahn R G Dunne B J amp Nelson R D (1987b) Engineering Anomalies Research Journal ofScienti c Exploration 1 21ndash50

Effects of Frontal Lobe Lesions on Intentionality 667

Results of Replication Study

Figure 3 shows the mean REG output for each intention (right left baseline)and the control output for the left frontal subject There were 19000 intentionand control trials for the right intention and 20000 trials for each of the inten-tion and control conditions for the left and baseline intentions

TABLE 5bPseudodata Analysis 2

Mean REG Output

Subject Intention Intention Control t p-value Trials (n)

PS1 right 9995 10001 08327 041 40000left 9996 9995 iexcl02329 082 40000baseline 10003 10000 iexcl04994 062 40000

PS2 right 10001 9995 iexcl06746 050 22900left 10000 9991 iexcl09721 033 24000baseline 9996 9999 03060 076 24000

PS3 right 10011 10008 iexcl04347 066 34000left 9998 10001 03522 072 32000a

baseline 10005 10000 iexcl06143 054 32000PS4 right 9994 10005 26336 00084L 104000

left 10001 9995 iexcl13071 019 104000baseline 9998 9997 iexcl02664 079 104000

PS5 right 9986 10005 10582 029 6000left 10003 10005 01187 091 8000baseline 10002 9995 iexcl03727 071 6000

PS6 right 9968 9993 07972 043 2000left 9990 9976 iexcl04402 066 2000baseline 10008 9960 iexcl15583 012 2000

PS7 right 9990 9995 04285 067 16000left 10018 10002 iexcl14678 014 16000baseline 9983 9996 11654 024 16000

PS8 right 10008 10005 iexcl02559 080 24000left 10016 9984 iexcl35151 00004R 24000baseline 10002 10002 iexcl00648 095 24000

PS9 right 10001 9998 iexcl02441 081 18000left 10017 9998 iexcl16610 010 16000baseline 9993 9983 iexcl08404 040 16000

PS10 right 10009 10014 02525 080 6000left 10018 10000 iexcl07789 044 4000baseline 9987 10022 15658 012 4000

PS11 right 10006 10010 03433 073 16000left 9992 10000 07129 048 16000baseline 10007 9986 iexcl19258 00541R 16000

PS12 right 9997 9986 iexcl10275 030 16000left 10002 9999 iexcl02967 077 16000baseline 9978 10001 20961 00361L 16000

Note n ˆ number of intention trials Dagger control trials R ˆ direction of effect was to the right L ˆdirection of effect was to the left

a Pseudodata contained 500 fewer intention and control trials as compared to real data

662 M Freedman et al

T-tests were carried out using SAS System for Windows Release 8 todetermine whether there were significant differences between the intention andcontrol conditions for the right left and baseline intentions There wasa significant difference between the intention and control conditions for the rightintention (t[37998]ˆiexcl253 pˆ 00115) but not for the other two intentions (leftt[39998] ˆ 049 p ˆ 06269 and baseline t[39998] ˆ iexcl007 p ˆ 09418) Thesignificant effect for the right intention was in the direction of intention as in theoriginal study

Since the investigation was run in blocks of 1000 trials we examined the datausing blocks as the unit of measurement and the averaged output across trials asthe measure in each block An F-ratio test with condition (intention vs control)as the source of variance of interest and the interaction between condition andblock as the error term showed an effect of condition (p ˆ 00578) This suggeststhat the left frontal subject would show comparable findings on another series of19 pairs of blocks of 1000 trials with 94 confidence

Whereas the mean REG output for the right intention was above the expectedvalue of 100 the output for the control condition was below 100 (Figure 3) Toexamine the possibility that the significant difference between the intention andcontrol condition was due solely to low output during the control condition asopposed to high output during the intention condition we tested whether theoutput for the intention condition was significantly different from a constantvalue of 100 This value is the theoretical mean output of the REG assumingtruly random and unbiased output The control values for the left and baselineconditions are approximately equal to this theoretical mean The REG output forthe right intention was significantly different than 100 and in the direction ofintention (t ˆ 201 p ˆ 0045 two-tailed test) For the control condition themean REG output was less than (although not significantly different from) 100(t ˆ iexcl156 p ˆ 012)

Fig 3 Mean REG output for the left frontal patient in replication study

Effects of Frontal Lobe Lesions on Intentionality 663

Figure 4 (andashc) shows the mean output of the REG broken down byconsecutive 1000 trial sessions for each intention (right left and baseline) andfor the control output For the right intention Figure 4 shows that on most pairsof blocks of 1000 trials the mean REG output follows a fairly consistent patternin which the means are either higher compared to the control condition or aboutthe same In contrast the data for each of the left and baseline conditions showa pattern with less separation between conditions

Fig 4 Mean REG output for each 1000 trial session in replication study (a) Intention Right (b)Intention Left and (c) Intention Baseline

664 M Freedman et al

Discussion

We examined the claims from the PEAR program that an individualrsquosintentions can influence the statistical distribution of random physical phenom-ena Our main focus was to test these claims using well-designed control condi-tions in a population that might maximize the likelihood of detecting any effectsthat may exist Our choice of studying patients with frontal lobe lesions wasbased on the concept that the potential to influence random physical events maybe optimized when attention is diverted from self-awareness (ie such as whenself-awareness is reduced) This state can occur following frontal brain damage(Stuss et al 1986 Stuss 1991 Carver et al 1991) Our findings showed asignificant effect for the right intention in a subject with left frontal braindamage Moreover this effect was in the direction of intention In contrast therewere no significant effects in the other groups (ie bilateral frontal right frontalpooled frontal or normal subjects)

Although the result in the left frontal patient on the right intention wasstatistically significant even after correction for multiple comparisons thisfinding was interpreted with great caution First it was based on a relativelysmall number of intention trials (n ˆ 3000) and was derived from only onesubject Second we found a p-value for an individual lsquolsquosubjectrsquorsquo in the pseu-dodata that would also have met criteria for significance using a Bonferronicorrection and that was even less than the p-value of 00015 obtained fromthe left frontal subject Although the effect of this lsquolsquopseudopatientrsquorsquo was not inthe direction of intention the fact that it was significant raised caution forinterpreting the effect in the left frontal patient However the replication of thefindings in the left frontal patient in a second well-designed study for each of the

Fig 4 (Continued)

Effects of Frontal Lobe Lesions on Intentionality 665

three intentions right left and baseline suggests that the effect in this subjectmay be more than chance occurrence

Additional support for a reliable effect in the left frontal subject comes fromfurther analysis of his data suggesting about a 94 chance of his showinga similar finding if another series of 19 pairs of blocks of 1000 trials were runagain Furthermore the profile of REG output data for the right intentionshowed a fairly consistent separation whereas this was not the case for the left orbaseline intentions

A comment is warranted about the REG output for the left frontal patientbeing lower for the control condition on the right intention or more to the leftas compared to the control conditions for the left or baseline intentions Onemight argue that the significant effect on the right intention was an artifact ofcomparison to control data that was in the left direction and that this widened thedifference between the intention and control data However this argument is nottenable because the effect on the right intention was significant even when theREG output was compared to a theoretical mean of 100mdasha value which is ap-proximately equal to the control means for the left intention (9999985) and thebaseline intention (1000011)

The patientrsquos cognitive deficits and brain lesion have been describedelsewhere (Marras et al 1998) He suffered from a tension pneumocephaluswhich resulted in cognitive deficits and epileptic seizures The cognitive deficitsinclude decreased mental flexibility on the Trail-Making Test poor attentionreduced fluency and impaired spatial planning and visuospatial problem solvingMRI showed an extensive left frontal lesion but the right frontal lobe was intactPsychometric testing and SPECT suggested the addition of right frontal dysfunc-tion The SPECT findings provide a measure of function as opposed to structureand were subtle

As indicated above frontal lesions have been associated with reduced self-awareness (Stuss et al 1986 Stuss 1991 Carver et al 1991) a state that isdifficult for normal individuals to achieve The rationale for studying patientswith damage to the frontal lobes was that decreased self-awareness might facili-tate the effects of intentionality on random physical phenomena Brain regionsthat mediate neurological processes underlying self-awareness include thefrontal lobes bilaterally particularly on the right (Stuss et al 2001a Stuss ampAlexander 2000a Stuss amp Alexander 2000b Stuss et al 2001b) Whereas thepatientrsquos right-sided brain dysfunction may have been insufficient to producea deficit in self-awareness the extensive lesion on the left may have resulted inreduced self-awareness when attention was directed towards the right Howeverit remains unclear why positive results should be found only following damageto the left frontal region and not after bilateral or right frontal lesions Onespeculation is that the effect on random physical events may require reducedself-awareness combined with relatively intact attentional mechanisms Theassociation of frontal lobe lesions especially on the right with impaired at-tention (Stuss amp Levine 2002) may explain the negative findings in the setting

666 M Freedman et al

of bilateral or right frontal damage Whether deficits related to left frontalabnormalities can explain the observed effects on the REG output warrantsfurther study Moreover the question whether normal processes associated withintact frontal lobe function together with preserved self-awareness may serve toinhibit effects of intentionality on random physical phenomena needs to beaddressed

The strength of our conclusions rests largely on a well-designed methodol-ogy and replication of our findings Although our results did not replicate thefindings reported by Jahn and his colleagues in normal subjects (Jahn et al1987a 1997) they support their claims that intentionality can alter the output ofa random event generator Furthermore our findings suggest that patients withfrontal lobe lesions may serve as a good model for future studies of the effects ofconsciousness on random physical events

Acknowledgments

We gratefully acknowledge Dr Donald Stuss for his helpful advice andsupport throughout the study and for reviewing the manuscript and for hishelpful suggestions Dr Louis Siminovitch for helpful advice and supportPatricia Ebert and Dr Donald Stuss for assisting with analysis of lesion local-ization for the CT and MRI scans Janice Pogue for statistical advice LeeFerreira for testing the subjects John Sloan for assistance with data collec-tion Tania Xerri for assistance with data analyses and Tania Xerri MaedeanBrown Wanna Mar and Melissa Ziraldo for preparation of materials for pub-lication We also gratefully acknowledge Professor Robert Jahn BrendaDunne Dr Roger Nelson Dr York Dobyns and John Bradish at the PEARlab for their helpful advice and for providing the portable REG and softwarefor the study This study was supported by the Fetzer Institute Dr Freedmanwas supported by the Saul A Silverman Family Foundation Toronto Canadaas a Canada-International Scientific exchange program (CISEPO) project

References

Alcock J E (1987) ParapsychologyScience of the Anomalous or Search for the Soul Behavioraland Brain Sciences 10 553ndash643

Carver C S amp Scheier M F (1991) Self-Regulation and the Self In J Strauss amp G R Goethals(Eds) The Self Interdisciplinary Approaches (pp 168ndash207) New York Springer-Verlag

Hall J Kim C McElroy B amp Shimony A (1977) Wave-packet reduction as a medium ofcommunication Foundations of Physics 7 759ndash767

Ibison M amp Jeffers S (1998) A Double-Slit Diffraction Experiment to Investigate Claims ofConsciousness-Related Anomalies Journal of Scienti c Exploration 12 543ndash550

Jahn R G amp Dunne B J (1986) On the Quantum Mechanics of Consciousness with Applicationto Anomalous Phenomena Foundations of Physics 16 721ndash772

Jahn R G amp Dunne B J (1987a) Margins of Reality The Role of Consciousness in the PhysicalWorld San Diego Harcourt Brace Jovanovich

Jahn R G Dunne B J amp Nelson R D (1987b) Engineering Anomalies Research Journal ofScienti c Exploration 1 21ndash50

Effects of Frontal Lobe Lesions on Intentionality 667

T-tests were carried out using SAS System for Windows Release 8 todetermine whether there were significant differences between the intention andcontrol conditions for the right left and baseline intentions There wasa significant difference between the intention and control conditions for the rightintention (t[37998]ˆiexcl253 pˆ 00115) but not for the other two intentions (leftt[39998] ˆ 049 p ˆ 06269 and baseline t[39998] ˆ iexcl007 p ˆ 09418) Thesignificant effect for the right intention was in the direction of intention as in theoriginal study

Since the investigation was run in blocks of 1000 trials we examined the datausing blocks as the unit of measurement and the averaged output across trials asthe measure in each block An F-ratio test with condition (intention vs control)as the source of variance of interest and the interaction between condition andblock as the error term showed an effect of condition (p ˆ 00578) This suggeststhat the left frontal subject would show comparable findings on another series of19 pairs of blocks of 1000 trials with 94 confidence

Whereas the mean REG output for the right intention was above the expectedvalue of 100 the output for the control condition was below 100 (Figure 3) Toexamine the possibility that the significant difference between the intention andcontrol condition was due solely to low output during the control condition asopposed to high output during the intention condition we tested whether theoutput for the intention condition was significantly different from a constantvalue of 100 This value is the theoretical mean output of the REG assumingtruly random and unbiased output The control values for the left and baselineconditions are approximately equal to this theoretical mean The REG output forthe right intention was significantly different than 100 and in the direction ofintention (t ˆ 201 p ˆ 0045 two-tailed test) For the control condition themean REG output was less than (although not significantly different from) 100(t ˆ iexcl156 p ˆ 012)

Fig 3 Mean REG output for the left frontal patient in replication study

Effects of Frontal Lobe Lesions on Intentionality 663

Figure 4 (andashc) shows the mean output of the REG broken down byconsecutive 1000 trial sessions for each intention (right left and baseline) andfor the control output For the right intention Figure 4 shows that on most pairsof blocks of 1000 trials the mean REG output follows a fairly consistent patternin which the means are either higher compared to the control condition or aboutthe same In contrast the data for each of the left and baseline conditions showa pattern with less separation between conditions

Fig 4 Mean REG output for each 1000 trial session in replication study (a) Intention Right (b)Intention Left and (c) Intention Baseline

664 M Freedman et al

Discussion

We examined the claims from the PEAR program that an individualrsquosintentions can influence the statistical distribution of random physical phenom-ena Our main focus was to test these claims using well-designed control condi-tions in a population that might maximize the likelihood of detecting any effectsthat may exist Our choice of studying patients with frontal lobe lesions wasbased on the concept that the potential to influence random physical events maybe optimized when attention is diverted from self-awareness (ie such as whenself-awareness is reduced) This state can occur following frontal brain damage(Stuss et al 1986 Stuss 1991 Carver et al 1991) Our findings showed asignificant effect for the right intention in a subject with left frontal braindamage Moreover this effect was in the direction of intention In contrast therewere no significant effects in the other groups (ie bilateral frontal right frontalpooled frontal or normal subjects)

Although the result in the left frontal patient on the right intention wasstatistically significant even after correction for multiple comparisons thisfinding was interpreted with great caution First it was based on a relativelysmall number of intention trials (n ˆ 3000) and was derived from only onesubject Second we found a p-value for an individual lsquolsquosubjectrsquorsquo in the pseu-dodata that would also have met criteria for significance using a Bonferronicorrection and that was even less than the p-value of 00015 obtained fromthe left frontal subject Although the effect of this lsquolsquopseudopatientrsquorsquo was not inthe direction of intention the fact that it was significant raised caution forinterpreting the effect in the left frontal patient However the replication of thefindings in the left frontal patient in a second well-designed study for each of the

Fig 4 (Continued)

Effects of Frontal Lobe Lesions on Intentionality 665

three intentions right left and baseline suggests that the effect in this subjectmay be more than chance occurrence

Additional support for a reliable effect in the left frontal subject comes fromfurther analysis of his data suggesting about a 94 chance of his showinga similar finding if another series of 19 pairs of blocks of 1000 trials were runagain Furthermore the profile of REG output data for the right intentionshowed a fairly consistent separation whereas this was not the case for the left orbaseline intentions

A comment is warranted about the REG output for the left frontal patientbeing lower for the control condition on the right intention or more to the leftas compared to the control conditions for the left or baseline intentions Onemight argue that the significant effect on the right intention was an artifact ofcomparison to control data that was in the left direction and that this widened thedifference between the intention and control data However this argument is nottenable because the effect on the right intention was significant even when theREG output was compared to a theoretical mean of 100mdasha value which is ap-proximately equal to the control means for the left intention (9999985) and thebaseline intention (1000011)

The patientrsquos cognitive deficits and brain lesion have been describedelsewhere (Marras et al 1998) He suffered from a tension pneumocephaluswhich resulted in cognitive deficits and epileptic seizures The cognitive deficitsinclude decreased mental flexibility on the Trail-Making Test poor attentionreduced fluency and impaired spatial planning and visuospatial problem solvingMRI showed an extensive left frontal lesion but the right frontal lobe was intactPsychometric testing and SPECT suggested the addition of right frontal dysfunc-tion The SPECT findings provide a measure of function as opposed to structureand were subtle

As indicated above frontal lesions have been associated with reduced self-awareness (Stuss et al 1986 Stuss 1991 Carver et al 1991) a state that isdifficult for normal individuals to achieve The rationale for studying patientswith damage to the frontal lobes was that decreased self-awareness might facili-tate the effects of intentionality on random physical phenomena Brain regionsthat mediate neurological processes underlying self-awareness include thefrontal lobes bilaterally particularly on the right (Stuss et al 2001a Stuss ampAlexander 2000a Stuss amp Alexander 2000b Stuss et al 2001b) Whereas thepatientrsquos right-sided brain dysfunction may have been insufficient to producea deficit in self-awareness the extensive lesion on the left may have resulted inreduced self-awareness when attention was directed towards the right Howeverit remains unclear why positive results should be found only following damageto the left frontal region and not after bilateral or right frontal lesions Onespeculation is that the effect on random physical events may require reducedself-awareness combined with relatively intact attentional mechanisms Theassociation of frontal lobe lesions especially on the right with impaired at-tention (Stuss amp Levine 2002) may explain the negative findings in the setting

666 M Freedman et al

of bilateral or right frontal damage Whether deficits related to left frontalabnormalities can explain the observed effects on the REG output warrantsfurther study Moreover the question whether normal processes associated withintact frontal lobe function together with preserved self-awareness may serve toinhibit effects of intentionality on random physical phenomena needs to beaddressed

The strength of our conclusions rests largely on a well-designed methodol-ogy and replication of our findings Although our results did not replicate thefindings reported by Jahn and his colleagues in normal subjects (Jahn et al1987a 1997) they support their claims that intentionality can alter the output ofa random event generator Furthermore our findings suggest that patients withfrontal lobe lesions may serve as a good model for future studies of the effects ofconsciousness on random physical events

Acknowledgments

We gratefully acknowledge Dr Donald Stuss for his helpful advice andsupport throughout the study and for reviewing the manuscript and for hishelpful suggestions Dr Louis Siminovitch for helpful advice and supportPatricia Ebert and Dr Donald Stuss for assisting with analysis of lesion local-ization for the CT and MRI scans Janice Pogue for statistical advice LeeFerreira for testing the subjects John Sloan for assistance with data collec-tion Tania Xerri for assistance with data analyses and Tania Xerri MaedeanBrown Wanna Mar and Melissa Ziraldo for preparation of materials for pub-lication We also gratefully acknowledge Professor Robert Jahn BrendaDunne Dr Roger Nelson Dr York Dobyns and John Bradish at the PEARlab for their helpful advice and for providing the portable REG and softwarefor the study This study was supported by the Fetzer Institute Dr Freedmanwas supported by the Saul A Silverman Family Foundation Toronto Canadaas a Canada-International Scientific exchange program (CISEPO) project

References

Alcock J E (1987) ParapsychologyScience of the Anomalous or Search for the Soul Behavioraland Brain Sciences 10 553ndash643

Carver C S amp Scheier M F (1991) Self-Regulation and the Self In J Strauss amp G R Goethals(Eds) The Self Interdisciplinary Approaches (pp 168ndash207) New York Springer-Verlag

Hall J Kim C McElroy B amp Shimony A (1977) Wave-packet reduction as a medium ofcommunication Foundations of Physics 7 759ndash767

Ibison M amp Jeffers S (1998) A Double-Slit Diffraction Experiment to Investigate Claims ofConsciousness-Related Anomalies Journal of Scienti c Exploration 12 543ndash550

Jahn R G amp Dunne B J (1986) On the Quantum Mechanics of Consciousness with Applicationto Anomalous Phenomena Foundations of Physics 16 721ndash772

Jahn R G amp Dunne B J (1987a) Margins of Reality The Role of Consciousness in the PhysicalWorld San Diego Harcourt Brace Jovanovich

Jahn R G Dunne B J amp Nelson R D (1987b) Engineering Anomalies Research Journal ofScienti c Exploration 1 21ndash50

Effects of Frontal Lobe Lesions on Intentionality 667

Figure 4 (andashc) shows the mean output of the REG broken down byconsecutive 1000 trial sessions for each intention (right left and baseline) andfor the control output For the right intention Figure 4 shows that on most pairsof blocks of 1000 trials the mean REG output follows a fairly consistent patternin which the means are either higher compared to the control condition or aboutthe same In contrast the data for each of the left and baseline conditions showa pattern with less separation between conditions

Fig 4 Mean REG output for each 1000 trial session in replication study (a) Intention Right (b)Intention Left and (c) Intention Baseline

664 M Freedman et al

Discussion

We examined the claims from the PEAR program that an individualrsquosintentions can influence the statistical distribution of random physical phenom-ena Our main focus was to test these claims using well-designed control condi-tions in a population that might maximize the likelihood of detecting any effectsthat may exist Our choice of studying patients with frontal lobe lesions wasbased on the concept that the potential to influence random physical events maybe optimized when attention is diverted from self-awareness (ie such as whenself-awareness is reduced) This state can occur following frontal brain damage(Stuss et al 1986 Stuss 1991 Carver et al 1991) Our findings showed asignificant effect for the right intention in a subject with left frontal braindamage Moreover this effect was in the direction of intention In contrast therewere no significant effects in the other groups (ie bilateral frontal right frontalpooled frontal or normal subjects)

Although the result in the left frontal patient on the right intention wasstatistically significant even after correction for multiple comparisons thisfinding was interpreted with great caution First it was based on a relativelysmall number of intention trials (n ˆ 3000) and was derived from only onesubject Second we found a p-value for an individual lsquolsquosubjectrsquorsquo in the pseu-dodata that would also have met criteria for significance using a Bonferronicorrection and that was even less than the p-value of 00015 obtained fromthe left frontal subject Although the effect of this lsquolsquopseudopatientrsquorsquo was not inthe direction of intention the fact that it was significant raised caution forinterpreting the effect in the left frontal patient However the replication of thefindings in the left frontal patient in a second well-designed study for each of the

Fig 4 (Continued)

Effects of Frontal Lobe Lesions on Intentionality 665

three intentions right left and baseline suggests that the effect in this subjectmay be more than chance occurrence

Additional support for a reliable effect in the left frontal subject comes fromfurther analysis of his data suggesting about a 94 chance of his showinga similar finding if another series of 19 pairs of blocks of 1000 trials were runagain Furthermore the profile of REG output data for the right intentionshowed a fairly consistent separation whereas this was not the case for the left orbaseline intentions

A comment is warranted about the REG output for the left frontal patientbeing lower for the control condition on the right intention or more to the leftas compared to the control conditions for the left or baseline intentions Onemight argue that the significant effect on the right intention was an artifact ofcomparison to control data that was in the left direction and that this widened thedifference between the intention and control data However this argument is nottenable because the effect on the right intention was significant even when theREG output was compared to a theoretical mean of 100mdasha value which is ap-proximately equal to the control means for the left intention (9999985) and thebaseline intention (1000011)

The patientrsquos cognitive deficits and brain lesion have been describedelsewhere (Marras et al 1998) He suffered from a tension pneumocephaluswhich resulted in cognitive deficits and epileptic seizures The cognitive deficitsinclude decreased mental flexibility on the Trail-Making Test poor attentionreduced fluency and impaired spatial planning and visuospatial problem solvingMRI showed an extensive left frontal lesion but the right frontal lobe was intactPsychometric testing and SPECT suggested the addition of right frontal dysfunc-tion The SPECT findings provide a measure of function as opposed to structureand were subtle

As indicated above frontal lesions have been associated with reduced self-awareness (Stuss et al 1986 Stuss 1991 Carver et al 1991) a state that isdifficult for normal individuals to achieve The rationale for studying patientswith damage to the frontal lobes was that decreased self-awareness might facili-tate the effects of intentionality on random physical phenomena Brain regionsthat mediate neurological processes underlying self-awareness include thefrontal lobes bilaterally particularly on the right (Stuss et al 2001a Stuss ampAlexander 2000a Stuss amp Alexander 2000b Stuss et al 2001b) Whereas thepatientrsquos right-sided brain dysfunction may have been insufficient to producea deficit in self-awareness the extensive lesion on the left may have resulted inreduced self-awareness when attention was directed towards the right Howeverit remains unclear why positive results should be found only following damageto the left frontal region and not after bilateral or right frontal lesions Onespeculation is that the effect on random physical events may require reducedself-awareness combined with relatively intact attentional mechanisms Theassociation of frontal lobe lesions especially on the right with impaired at-tention (Stuss amp Levine 2002) may explain the negative findings in the setting

666 M Freedman et al

of bilateral or right frontal damage Whether deficits related to left frontalabnormalities can explain the observed effects on the REG output warrantsfurther study Moreover the question whether normal processes associated withintact frontal lobe function together with preserved self-awareness may serve toinhibit effects of intentionality on random physical phenomena needs to beaddressed

The strength of our conclusions rests largely on a well-designed methodol-ogy and replication of our findings Although our results did not replicate thefindings reported by Jahn and his colleagues in normal subjects (Jahn et al1987a 1997) they support their claims that intentionality can alter the output ofa random event generator Furthermore our findings suggest that patients withfrontal lobe lesions may serve as a good model for future studies of the effects ofconsciousness on random physical events

Acknowledgments

We gratefully acknowledge Dr Donald Stuss for his helpful advice andsupport throughout the study and for reviewing the manuscript and for hishelpful suggestions Dr Louis Siminovitch for helpful advice and supportPatricia Ebert and Dr Donald Stuss for assisting with analysis of lesion local-ization for the CT and MRI scans Janice Pogue for statistical advice LeeFerreira for testing the subjects John Sloan for assistance with data collec-tion Tania Xerri for assistance with data analyses and Tania Xerri MaedeanBrown Wanna Mar and Melissa Ziraldo for preparation of materials for pub-lication We also gratefully acknowledge Professor Robert Jahn BrendaDunne Dr Roger Nelson Dr York Dobyns and John Bradish at the PEARlab for their helpful advice and for providing the portable REG and softwarefor the study This study was supported by the Fetzer Institute Dr Freedmanwas supported by the Saul A Silverman Family Foundation Toronto Canadaas a Canada-International Scientific exchange program (CISEPO) project

References

Alcock J E (1987) ParapsychologyScience of the Anomalous or Search for the Soul Behavioraland Brain Sciences 10 553ndash643

Carver C S amp Scheier M F (1991) Self-Regulation and the Self In J Strauss amp G R Goethals(Eds) The Self Interdisciplinary Approaches (pp 168ndash207) New York Springer-Verlag

Hall J Kim C McElroy B amp Shimony A (1977) Wave-packet reduction as a medium ofcommunication Foundations of Physics 7 759ndash767

Ibison M amp Jeffers S (1998) A Double-Slit Diffraction Experiment to Investigate Claims ofConsciousness-Related Anomalies Journal of Scienti c Exploration 12 543ndash550

Jahn R G amp Dunne B J (1986) On the Quantum Mechanics of Consciousness with Applicationto Anomalous Phenomena Foundations of Physics 16 721ndash772

Jahn R G amp Dunne B J (1987a) Margins of Reality The Role of Consciousness in the PhysicalWorld San Diego Harcourt Brace Jovanovich

Jahn R G Dunne B J amp Nelson R D (1987b) Engineering Anomalies Research Journal ofScienti c Exploration 1 21ndash50

Effects of Frontal Lobe Lesions on Intentionality 667

Discussion

We examined the claims from the PEAR program that an individualrsquosintentions can influence the statistical distribution of random physical phenom-ena Our main focus was to test these claims using well-designed control condi-tions in a population that might maximize the likelihood of detecting any effectsthat may exist Our choice of studying patients with frontal lobe lesions wasbased on the concept that the potential to influence random physical events maybe optimized when attention is diverted from self-awareness (ie such as whenself-awareness is reduced) This state can occur following frontal brain damage(Stuss et al 1986 Stuss 1991 Carver et al 1991) Our findings showed asignificant effect for the right intention in a subject with left frontal braindamage Moreover this effect was in the direction of intention In contrast therewere no significant effects in the other groups (ie bilateral frontal right frontalpooled frontal or normal subjects)

Although the result in the left frontal patient on the right intention wasstatistically significant even after correction for multiple comparisons thisfinding was interpreted with great caution First it was based on a relativelysmall number of intention trials (n ˆ 3000) and was derived from only onesubject Second we found a p-value for an individual lsquolsquosubjectrsquorsquo in the pseu-dodata that would also have met criteria for significance using a Bonferronicorrection and that was even less than the p-value of 00015 obtained fromthe left frontal subject Although the effect of this lsquolsquopseudopatientrsquorsquo was not inthe direction of intention the fact that it was significant raised caution forinterpreting the effect in the left frontal patient However the replication of thefindings in the left frontal patient in a second well-designed study for each of the

Fig 4 (Continued)

Effects of Frontal Lobe Lesions on Intentionality 665

three intentions right left and baseline suggests that the effect in this subjectmay be more than chance occurrence

Additional support for a reliable effect in the left frontal subject comes fromfurther analysis of his data suggesting about a 94 chance of his showinga similar finding if another series of 19 pairs of blocks of 1000 trials were runagain Furthermore the profile of REG output data for the right intentionshowed a fairly consistent separation whereas this was not the case for the left orbaseline intentions

A comment is warranted about the REG output for the left frontal patientbeing lower for the control condition on the right intention or more to the leftas compared to the control conditions for the left or baseline intentions Onemight argue that the significant effect on the right intention was an artifact ofcomparison to control data that was in the left direction and that this widened thedifference between the intention and control data However this argument is nottenable because the effect on the right intention was significant even when theREG output was compared to a theoretical mean of 100mdasha value which is ap-proximately equal to the control means for the left intention (9999985) and thebaseline intention (1000011)

The patientrsquos cognitive deficits and brain lesion have been describedelsewhere (Marras et al 1998) He suffered from a tension pneumocephaluswhich resulted in cognitive deficits and epileptic seizures The cognitive deficitsinclude decreased mental flexibility on the Trail-Making Test poor attentionreduced fluency and impaired spatial planning and visuospatial problem solvingMRI showed an extensive left frontal lesion but the right frontal lobe was intactPsychometric testing and SPECT suggested the addition of right frontal dysfunc-tion The SPECT findings provide a measure of function as opposed to structureand were subtle

As indicated above frontal lesions have been associated with reduced self-awareness (Stuss et al 1986 Stuss 1991 Carver et al 1991) a state that isdifficult for normal individuals to achieve The rationale for studying patientswith damage to the frontal lobes was that decreased self-awareness might facili-tate the effects of intentionality on random physical phenomena Brain regionsthat mediate neurological processes underlying self-awareness include thefrontal lobes bilaterally particularly on the right (Stuss et al 2001a Stuss ampAlexander 2000a Stuss amp Alexander 2000b Stuss et al 2001b) Whereas thepatientrsquos right-sided brain dysfunction may have been insufficient to producea deficit in self-awareness the extensive lesion on the left may have resulted inreduced self-awareness when attention was directed towards the right Howeverit remains unclear why positive results should be found only following damageto the left frontal region and not after bilateral or right frontal lesions Onespeculation is that the effect on random physical events may require reducedself-awareness combined with relatively intact attentional mechanisms Theassociation of frontal lobe lesions especially on the right with impaired at-tention (Stuss amp Levine 2002) may explain the negative findings in the setting

666 M Freedman et al

of bilateral or right frontal damage Whether deficits related to left frontalabnormalities can explain the observed effects on the REG output warrantsfurther study Moreover the question whether normal processes associated withintact frontal lobe function together with preserved self-awareness may serve toinhibit effects of intentionality on random physical phenomena needs to beaddressed

The strength of our conclusions rests largely on a well-designed methodol-ogy and replication of our findings Although our results did not replicate thefindings reported by Jahn and his colleagues in normal subjects (Jahn et al1987a 1997) they support their claims that intentionality can alter the output ofa random event generator Furthermore our findings suggest that patients withfrontal lobe lesions may serve as a good model for future studies of the effects ofconsciousness on random physical events

Acknowledgments

We gratefully acknowledge Dr Donald Stuss for his helpful advice andsupport throughout the study and for reviewing the manuscript and for hishelpful suggestions Dr Louis Siminovitch for helpful advice and supportPatricia Ebert and Dr Donald Stuss for assisting with analysis of lesion local-ization for the CT and MRI scans Janice Pogue for statistical advice LeeFerreira for testing the subjects John Sloan for assistance with data collec-tion Tania Xerri for assistance with data analyses and Tania Xerri MaedeanBrown Wanna Mar and Melissa Ziraldo for preparation of materials for pub-lication We also gratefully acknowledge Professor Robert Jahn BrendaDunne Dr Roger Nelson Dr York Dobyns and John Bradish at the PEARlab for their helpful advice and for providing the portable REG and softwarefor the study This study was supported by the Fetzer Institute Dr Freedmanwas supported by the Saul A Silverman Family Foundation Toronto Canadaas a Canada-International Scientific exchange program (CISEPO) project

References

Alcock J E (1987) ParapsychologyScience of the Anomalous or Search for the Soul Behavioraland Brain Sciences 10 553ndash643

Carver C S amp Scheier M F (1991) Self-Regulation and the Self In J Strauss amp G R Goethals(Eds) The Self Interdisciplinary Approaches (pp 168ndash207) New York Springer-Verlag

Hall J Kim C McElroy B amp Shimony A (1977) Wave-packet reduction as a medium ofcommunication Foundations of Physics 7 759ndash767

Ibison M amp Jeffers S (1998) A Double-Slit Diffraction Experiment to Investigate Claims ofConsciousness-Related Anomalies Journal of Scienti c Exploration 12 543ndash550

Jahn R G amp Dunne B J (1986) On the Quantum Mechanics of Consciousness with Applicationto Anomalous Phenomena Foundations of Physics 16 721ndash772

Jahn R G amp Dunne B J (1987a) Margins of Reality The Role of Consciousness in the PhysicalWorld San Diego Harcourt Brace Jovanovich

Jahn R G Dunne B J amp Nelson R D (1987b) Engineering Anomalies Research Journal ofScienti c Exploration 1 21ndash50

Effects of Frontal Lobe Lesions on Intentionality 667

three intentions right left and baseline suggests that the effect in this subjectmay be more than chance occurrence

Additional support for a reliable effect in the left frontal subject comes fromfurther analysis of his data suggesting about a 94 chance of his showinga similar finding if another series of 19 pairs of blocks of 1000 trials were runagain Furthermore the profile of REG output data for the right intentionshowed a fairly consistent separation whereas this was not the case for the left orbaseline intentions

A comment is warranted about the REG output for the left frontal patientbeing lower for the control condition on the right intention or more to the leftas compared to the control conditions for the left or baseline intentions Onemight argue that the significant effect on the right intention was an artifact ofcomparison to control data that was in the left direction and that this widened thedifference between the intention and control data However this argument is nottenable because the effect on the right intention was significant even when theREG output was compared to a theoretical mean of 100mdasha value which is ap-proximately equal to the control means for the left intention (9999985) and thebaseline intention (1000011)

The patientrsquos cognitive deficits and brain lesion have been describedelsewhere (Marras et al 1998) He suffered from a tension pneumocephaluswhich resulted in cognitive deficits and epileptic seizures The cognitive deficitsinclude decreased mental flexibility on the Trail-Making Test poor attentionreduced fluency and impaired spatial planning and visuospatial problem solvingMRI showed an extensive left frontal lesion but the right frontal lobe was intactPsychometric testing and SPECT suggested the addition of right frontal dysfunc-tion The SPECT findings provide a measure of function as opposed to structureand were subtle

As indicated above frontal lesions have been associated with reduced self-awareness (Stuss et al 1986 Stuss 1991 Carver et al 1991) a state that isdifficult for normal individuals to achieve The rationale for studying patientswith damage to the frontal lobes was that decreased self-awareness might facili-tate the effects of intentionality on random physical phenomena Brain regionsthat mediate neurological processes underlying self-awareness include thefrontal lobes bilaterally particularly on the right (Stuss et al 2001a Stuss ampAlexander 2000a Stuss amp Alexander 2000b Stuss et al 2001b) Whereas thepatientrsquos right-sided brain dysfunction may have been insufficient to producea deficit in self-awareness the extensive lesion on the left may have resulted inreduced self-awareness when attention was directed towards the right Howeverit remains unclear why positive results should be found only following damageto the left frontal region and not after bilateral or right frontal lesions Onespeculation is that the effect on random physical events may require reducedself-awareness combined with relatively intact attentional mechanisms Theassociation of frontal lobe lesions especially on the right with impaired at-tention (Stuss amp Levine 2002) may explain the negative findings in the setting

666 M Freedman et al

of bilateral or right frontal damage Whether deficits related to left frontalabnormalities can explain the observed effects on the REG output warrantsfurther study Moreover the question whether normal processes associated withintact frontal lobe function together with preserved self-awareness may serve toinhibit effects of intentionality on random physical phenomena needs to beaddressed

The strength of our conclusions rests largely on a well-designed methodol-ogy and replication of our findings Although our results did not replicate thefindings reported by Jahn and his colleagues in normal subjects (Jahn et al1987a 1997) they support their claims that intentionality can alter the output ofa random event generator Furthermore our findings suggest that patients withfrontal lobe lesions may serve as a good model for future studies of the effects ofconsciousness on random physical events

Acknowledgments

We gratefully acknowledge Dr Donald Stuss for his helpful advice andsupport throughout the study and for reviewing the manuscript and for hishelpful suggestions Dr Louis Siminovitch for helpful advice and supportPatricia Ebert and Dr Donald Stuss for assisting with analysis of lesion local-ization for the CT and MRI scans Janice Pogue for statistical advice LeeFerreira for testing the subjects John Sloan for assistance with data collec-tion Tania Xerri for assistance with data analyses and Tania Xerri MaedeanBrown Wanna Mar and Melissa Ziraldo for preparation of materials for pub-lication We also gratefully acknowledge Professor Robert Jahn BrendaDunne Dr Roger Nelson Dr York Dobyns and John Bradish at the PEARlab for their helpful advice and for providing the portable REG and softwarefor the study This study was supported by the Fetzer Institute Dr Freedmanwas supported by the Saul A Silverman Family Foundation Toronto Canadaas a Canada-International Scientific exchange program (CISEPO) project

References

Alcock J E (1987) ParapsychologyScience of the Anomalous or Search for the Soul Behavioraland Brain Sciences 10 553ndash643

Carver C S amp Scheier M F (1991) Self-Regulation and the Self In J Strauss amp G R Goethals(Eds) The Self Interdisciplinary Approaches (pp 168ndash207) New York Springer-Verlag

Hall J Kim C McElroy B amp Shimony A (1977) Wave-packet reduction as a medium ofcommunication Foundations of Physics 7 759ndash767

Ibison M amp Jeffers S (1998) A Double-Slit Diffraction Experiment to Investigate Claims ofConsciousness-Related Anomalies Journal of Scienti c Exploration 12 543ndash550

Jahn R G amp Dunne B J (1986) On the Quantum Mechanics of Consciousness with Applicationto Anomalous Phenomena Foundations of Physics 16 721ndash772

Jahn R G amp Dunne B J (1987a) Margins of Reality The Role of Consciousness in the PhysicalWorld San Diego Harcourt Brace Jovanovich

Jahn R G Dunne B J amp Nelson R D (1987b) Engineering Anomalies Research Journal ofScienti c Exploration 1 21ndash50

Effects of Frontal Lobe Lesions on Intentionality 667

of bilateral or right frontal damage Whether deficits related to left frontalabnormalities can explain the observed effects on the REG output warrantsfurther study Moreover the question whether normal processes associated withintact frontal lobe function together with preserved self-awareness may serve toinhibit effects of intentionality on random physical phenomena needs to beaddressed

The strength of our conclusions rests largely on a well-designed methodol-ogy and replication of our findings Although our results did not replicate thefindings reported by Jahn and his colleagues in normal subjects (Jahn et al1987a 1997) they support their claims that intentionality can alter the output ofa random event generator Furthermore our findings suggest that patients withfrontal lobe lesions may serve as a good model for future studies of the effects ofconsciousness on random physical events

Acknowledgments

We gratefully acknowledge Dr Donald Stuss for his helpful advice andsupport throughout the study and for reviewing the manuscript and for hishelpful suggestions Dr Louis Siminovitch for helpful advice and supportPatricia Ebert and Dr Donald Stuss for assisting with analysis of lesion local-ization for the CT and MRI scans Janice Pogue for statistical advice LeeFerreira for testing the subjects John Sloan for assistance with data collec-tion Tania Xerri for assistance with data analyses and Tania Xerri MaedeanBrown Wanna Mar and Melissa Ziraldo for preparation of materials for pub-lication We also gratefully acknowledge Professor Robert Jahn BrendaDunne Dr Roger Nelson Dr York Dobyns and John Bradish at the PEARlab for their helpful advice and for providing the portable REG and softwarefor the study This study was supported by the Fetzer Institute Dr Freedmanwas supported by the Saul A Silverman Family Foundation Toronto Canadaas a Canada-International Scientific exchange program (CISEPO) project

References

Alcock J E (1987) ParapsychologyScience of the Anomalous or Search for the Soul Behavioraland Brain Sciences 10 553ndash643

Carver C S amp Scheier M F (1991) Self-Regulation and the Self In J Strauss amp G R Goethals(Eds) The Self Interdisciplinary Approaches (pp 168ndash207) New York Springer-Verlag

Hall J Kim C McElroy B amp Shimony A (1977) Wave-packet reduction as a medium ofcommunication Foundations of Physics 7 759ndash767

Ibison M amp Jeffers S (1998) A Double-Slit Diffraction Experiment to Investigate Claims ofConsciousness-Related Anomalies Journal of Scienti c Exploration 12 543ndash550

Jahn R G amp Dunne B J (1986) On the Quantum Mechanics of Consciousness with Applicationto Anomalous Phenomena Foundations of Physics 16 721ndash772

Jahn R G amp Dunne B J (1987a) Margins of Reality The Role of Consciousness in the PhysicalWorld San Diego Harcourt Brace Jovanovich

Jahn R G Dunne B J amp Nelson R D (1987b) Engineering Anomalies Research Journal ofScienti c Exploration 1 21ndash50

Effects of Frontal Lobe Lesions on Intentionality 667

Jahn R G Dunne B J Nelson R D Dobyns Y H amp Bradish G J (1997) Correlations ofrandom binary sequences with pre-stated operator intention A review of a 12-year programJournal of Scienti c Exploration 11 345ndash367

Jeffers S (in press) Physics and Claims for Anomalous Effects Related to Consciousness Journal ofConsciousness Studies

Jeffers S amp Sloan J (1992) A low light level diffraction experiment for anomalies research Journalof Scienti c Exploration 6 333ndash352

Marras L C Kalaparambath T P Black S E amp Rowed D W (1998) Severe tensionpneumocephalus complicating frontal sinus osteoma Canadian Journal of Neurological Sciences25 79ndash81

Nelson R D Braddish G J amp Dobyns Y H (1989) Random Event Generator Quali cationcalibration and analysis (Technical Note PEAR 89001) Princeton NJ Princeton EngineeringAnomalies Research Princeton University School of EngineeringApplied Science

Nelson R D Dunne B J amp Jahn R G (1984) An REG experiment with large database capabilityIII Operator-related anomalies (Technical Note PEAR 84003) Princeton NJ PrincetonEngineering Anomalies Research Princeton University School of EngineeringApplied Science

Radin D I amp Nelson R D (1989) Evidence for consciousness-related anomalies in randomphysical systems Foundations of Physics 19 1499ndash1514

The SAS System for Windows Release 612 (1996) Cary NC SAS Institute IncSchmidt H (1969) Precognition of a quantum process Journal of Parapsychology 33 99ndash108Schmidt H amp Pantas L (1972) PSI tests with internally different machines Journal of

Parapsychology 36 222ndash232Stuss D T (1991) Self Awareness and the Frontal Lobes A Neuropsychological Perspective In

J Strauss amp G R Goethals (Eds) The Self Interdisciplinary Approaches (pp 255ndash278) NewYork Springer-Verlag

Stuss D T amp Alexander M P (2000a) Affectively burnt in A proposed role of the right frontallobe In E Tulving (Ed) Memory consciousness and the brain The Tallin Conference (pp 215ndash227) Philadelphia Psychology Press

Stuss D T amp Alexander M P (2000b) The anatomical basis of affective behavior emotion andself-awareness A speci c role of the right frontal lobe In G Hatano N Okada amp H Tanabe(Eds) Affective Minds The 13th Toyota Conference (pp 13ndash25) Amsterdam Elsevier

Stuss D T amp Benson D F (1986) The Frontal Lobes New York Raven PressStuss D T amp Levine B (2002) Adult clinical neuropsychologyLessons from studies of the frontal

lobes Annual Review of Psychology 53 401ndash433Stuss D T Gallup G G Jr amp Alexander M P (2001a) The frontal lobes are necessary for

lsquotheory of mindrsquo Brain 124 279ndash286Stuss D T Picton T W amp Alexander M P (2001b) Consciousness self-awareness and the

frontal lobes In S P Salloway P F Malloy amp J D Duffy (Eds) The Frontal Lobes andNeuropsychiatric Illness (pp101ndash109) Washington DC American Psychiatric Publishing

668 M Freedman et al