meghan searl, ph.d., abpp-cn - home -...

PLEASE SCROLL DOWN FOR ARTICLE

!"#$%&'(#)*+%,&$%-.,/*.&-+-%012%3456%4&'7&'-%5.**+8+9:/2%;<%:)(.0+'%;=>=?))+$$%-+(&#*$2%?))+$$%@+(&#*$2%3$A0$)'#B(#./%/AC0+'%D><E>DFE<9GA0*#$"+'%G$1)".*.81%G'+$$H/I.'C&%6(-%J+8#$(+'+-%#/%K/8*&/-%&/-%L&*+$%J+8#$(+'+-%MAC0+'2%>=F;DNO%J+8#$(+'+-%.II#)+2%P.'(#C+'%4.A$+Q%RFSO>%P.'(#C+'%T('++(Q%6./-./%L>!%RU4Q%VW

!"+%XA&'(+'*1%U.A'/&*%.I%KYB+'#C+/(&*%G$1)".*.81GA0*#)&(#./%-+(&#*$Q%#/)*A-#/8%#/$('A)(#./$%I.'%&A(".'$%&/-%$A0$)'#B(#./%#/I.'C&(#./2"((B2ZZ,,,[#/I.'C&,.'*-[).CZ$CBBZ(#(*+\)./(+/(](F>E>==F=O

^'+&(+'%+II#)#+/)1%#/%&((+/(#./&*%B'.)+$$#/8%'+*&(+-%(.%C#/-IA*/+$$C+-#(&(#./G&A*%?[%P[%7&/%-+/%4A'_&0`%a&0#.%^#.CC#0`%T(&/%5[%^#+*+/&`%?//+%K[%P[%TB+)_+/$)-`%4+/_%G[b&'+/-'+8(0&%@./-+'$%H/$(#(A(+%I.'%b'&#/Q%5.8/#(#./%&/-%b+"&7#.'S5+/(+'%I.'%5.8/#(#7+%M+A'.#C&8#/8Q%J&-0.A-V/#7+'$#(1Q%M#cC+8+/Q%!"+%M+("+'*&/-$%0%a&)A*(1%.I%T)#+/)+Q%J&-0.A-%V/#7+'$#(1Q%M#cC+8+/Q%!"+M+("+'*&/-$%)%@./-+'$%H/$(#(A(+%I.'%b'&#/Q%5.8/#(#./%&/-%b+"&7#.'S5+/(+'%I.'%5.8/#(#7+M+A'.#C&8#/8Q%J&-0.A-%V/#7+'$#(1Q%M#cC+8+/%P+-#)&*%5+/('+Q%M#cC+8+/Q%!"+%M+("+'*&/-$%-

@+B&'(C+/(%.I%G$1)"#&('1Q%J&-0.A-%V/#7+'$#(1Q%M#cC+8+/%P+-#)&*%5+/('+Q%M#cC+8+/Q%!"+%M+("+'*&/-$

a#'$(%BA0*#$"+-%./2%;E%T+B(+C0+'%;==D

!.%)#(+%("#$%?'(#)*+%7&/%-+/%4A'_Q%G&A*%?[%P[%Q%^#.CC#Q%a&0#.%Q%^#+*+/Q%T(&/%5[%Q%TB+)_+/$Q%?//+%K[%P[%&/-%b&'+/-'+8(Q4+/_%G[d;=>=e%f^'+&(+'%+II#)#+/)1%#/%&((+/(#./&*%B'.)+$$#/8%'+*&(+-%(.%C#/-IA*/+$$%C+-#(&(#./fQ%!"+%XA&'(+'*1%U.A'/&*%.IKYB+'#C+/(&*%G$1)".*.81Q%ER2%EQ%>>E<%g%>><=Q%a#'$(%BA0*#$"+-%./2%;E%T+B(+C0+'%;==D%d#a#'$(e!.%*#/_%(.%("#$%?'(#)*+2%@:H2%>=[>=<=Z>FOF=;>=D=R;ODRENVJ62%"((B2ZZ-Y[-.#[.'8Z>=[>=<=Z>FOF=;>=D=R;ODREN

Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf

This article may be used for research, teaching and private study purposes. Any substantial orsystematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply ordistribution in any form to anyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representation that the contentswill be complete or accurate or up to date. The accuracy of any instructions, formulae and drug dosesshould be independently verified with primary sources. The publisher shall not be liable for any loss,actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directlyor indirectly in connection with or arising out of the use of this material.

http://www.informaworld.com/terms-and-conditions-of-access.pdf

http://www.informaworld.com/smpp/title~content=t716100704

http://dx.doi.org/10.1080/17470210903249365

Greater efficiency in attentional processing related tomindfulness meditation

Paul A. M. van den HurkDonders Institute for Brain, Cognition and Behavior–Center for Cognitive Neuroimaging, and Faculty of Science,

Radboud University, Nijmegen, The Netherlands

Fabio GiommiFaculty of Science, Radboud University, Nijmegen, The Netherlands

Stan C. GielenDonders Institute for Brain, Cognition and Behavior–Center for Cognitive Neuroimaging, Radboud University,

Nijmegen, The Netherlands

Anne E. M. SpeckensDonders Institute for Brain, Cognition and Behavior–Center for Cognitive Neuroimaging, and Department of Psychiatry,

Radboud University, Nijmegen Medical Centre, Nijmegen, The Netherlands

Henk P. BarendregtFaculty of Science, Radboud University, Nijmegen, The Netherlands

In this study, attentional processing in relation to mindfulness meditation was investigated. Sincerecent studies have suggested that mindfulness meditation may induce improvements in attentionalprocessing, we have tested 20 expert mindfulness meditators in the attention network test. Theirperformance was compared to that of 20 age- and gender-matched controls. In addition to attentionalnetwork analyses, overall attentional processing was analysed by means of efficiency scores (i.e., accu-racy controlled for reaction time). Better orienting and executive attention (reflected by smaller differ-ences in either reaction time or error score, respectively) were observed in the mindfulness meditationgroup. Furthermore, extensive mindfulness meditation appeared to be related to a reduction of thefraction of errors for responses with the same reaction time. These results provide new insightsinto differences in attentional processing related to mindfulness meditation and suggest the possibilityof increasing the efficiency in attentional processing by extensive mental training.

Keywords: Attention; Attention network test; Mental training; Mindfulness meditation.

Recent studies have shown that attention andthe quality of moment-to-moment awareness areflexible skills that can be trained and improved

through mental training such as meditation (Chan& Woollacott, 2007; Jha, Krompinger, & Baime,2007; Slagter et al., 2007; Tang et al., 2007).

Correspondence should be addressed to Paul van den Hurk, Faculty of Science, Radboud University Nijmegen, Heyendaalseweg135, 6525 AJ Nijmegen, The Netherlands. E-mail: [email protected]

1168 # 2009 The Experimental Psychology Society

http://www.psypress.com/qjep DOI:10.1080/17470210903249365

THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY

2010, 63 (6), 1168–1180

Downloaded By: [HCL Harvard College] At: 02:54 28 October 2010

Over the last few decades, there has been a largeincrease in clinical applications and empiricalresearch on mindfulness-meditation-based inter-ventions, such as the Mindfulness-Based StressReduction programme (MBSR) and Mindfulness-Based Cognitive Therapy (MBCT). The mindful-ness-meditation-based practices have proven to beeffective in reducing the clinical symptoms in a con-siderable number of psychological, psychosomatic,and emotional disturbances (Atin, 1997; Baer,2003; Giommi, 2006; Grossman, Niemann,Schmidt, & Walach, 2004; Kabat-Zinn, 1990,1992, 2003; Kabat-Zinn, Lipworth, & Burney,1985; Kristeller & Hallett, 1999; Speca, Carlson,Goodey, & Angen, 2000; Teasdale, Segal, &Williams, 1995; Teasdale et al., 2000). Despitethe fact that a growing body of scientific literatureabout the clinical applications of mindfulness med-itation seems to reveal promising results of its effec-tiveness (Kabat-Zinn, 1994, 2003), surprisinglyfew studies have addressed the neuropsychologyand the neurophysiology of mindfulness medita-tion, and very little is known about the mechan-ism(s) by which mindfulness meditation exerts itseffect(s) (Bishop et al., 2004; Shapiro, Carlson,Astin, & Freedman, 2006). As mindfulnessmeditation has been described as “a particular wayto pay attention” (Kabat-Zinn, 1990, 2003), aputative candidate mechanism for its effect(s) isa modification in attentional processing. Morespecifically, recent conceptualizations havesuggested mindfulness meditation to improvethe self-regulation of attention (Bishop et al.,2004; Shapiro et al., 2006).

William James defined attention as: “the takingpossession of the mind in clear and vivid formof one out of what seem several simultaneousobjects or trains of thought” (James, 1890,Vol. 1, pp. 403–404). For about a century there-after, several theoretical models of attention havebeen put forward, one of which is that by Posner

and Petersen (1990). Their tripartite model hasbeen highly influential in modifying the view onattention, considering it as an organ system withits own anatomy and circuitry (Fan, McCandliss,Fossella, Flombaum & Posner, 2005; Fan &Posner, 2004; Posner & Petersen, 1990).1 Intheir model, attention has been conceptualized ascomprising three separate functional componentsor attentional networks: the alerting, orienting,and executive attention networks. The aim of thealerting network is to achieve and maintain avigilant or alert state of preparedness; the orientingnetwork regulates directing and limiting attentionto a subset of possible inputs; and the executiveattention network resolves conflict among multipleresponses.

In 2002, Fan and others (Fan, McCandliss,Sommer, Raz, & Posner, 2002) presented theattention network test to investigate the function-ing of the three different attention networksproposed by Posner and Petersen. Because of thereliability of the scores obtained with the attentionnetwork test and its easy use in a wide variety ofsubjects and patients (Fan et al., 2002), the atten-tion network test has since then been used in manystudies on both normal and clinically impairedattentional processing (Hovels-Gurich et al.,2007; Leskin & White, 2007; Neuhaus et al.,2007; Wang et al., 2005).

In a recent pioneering study, Jha et al. (2007)used the attention network test to examinechanges in attentional processing induced bymind-fulness meditation. They performed a longitudinalstudy with three groups: a retreat, MBSR, andcontrol group.2 During the baseline measurement,they found a significantly better executive attentionnetwork in the retreat group, both in reaction time(RT) and in error score (ES) data, where ES wasdefined as the percentage of incorrect responses.No significant differences were observed on boththe alerting and orienting networks.

1 An organ system is defined as differentiated structures made up of various cells and tissues and adapted for the performance ofsome specific function and grouped with other structures into a system (Posner & Petersen, 1990).

2 A retreat is a period of secluded, continuous, and intensive group practice of meditation, varying from 1 week to 1 month or evenmore (the retreat in the Jha et al. study lasted 1 month). MBSR is an 8-week programme with weekly group sessions of 3 hours anddaily homework.

THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2010, 63 (6) 1169

MEDITATION AND ATTENTIONAL PROCESSING


Although the results obtained by Jha et al.(2007) were very promising, their study raisedseveral questions. First, recent research hasrevealed a positive interaction between the orient-ing and executive network (Callejas, Lupianez,Funes, & Tudela, 2005; Callejas, Lupianez, &Tudela, 2004). Therefore, it was somewhat sur-prising that Jha et al. did not report significantdifferences in the orienting network, while a sig-nificant effect was found in the executivenetwork. In order to investigate this issue, wealso used the attention network test and tested agroup of more experienced meditators than thosetested in the study by Jha et al. (mean 14.5 years,range 0.33–35 years in our group vs. mean 5years, range 0.33–30 years in the study by Jhaet al.). With this sample of more experienced med-itators, we expected to increase the likelihood offinding differences within the orienting and execu-tive network. Secondly, Jha et al. focused theiranalysis on attentional network scores, which isthe standard approach when the attentionnetwork test is used. In our study, we were alsointerested in the analysis of grand mean RT andES data, as these data are informative on overalldifferences in attentional processing between med-itators and controls.

We hypothesized to find the following resultsfor the group of experienced meditators as com-pared to controls: (a) considering the orientingand executive attention, we predicted to observebetter functioning for experienced meditators,reflected by smaller network scores on the orient-ing and executive attention components of theattention network test. To further explain thesehypotheses, we assumed that through mindfulnessmeditation—by which the meditator is trained toget attached to the stimuli as little as possible—amore receptive (“open field”) form of attention isacquired, and, as a consequence, the orienting ofattention is improved—that is, more flexible. Inaddition, since meditators have been trained infocusing attention, we hypothesized that theywould be better in focusing to the relevant infor-mation embedded in the target environmentwhile ignoring the irrelevant, distracting infor-mation. This is thought to be reflected in a

better executive attention. Since only few studiesdid address the alerting network, no hypothesiswas formulated for this attention component. (b)In addition to our interest in the functioning ofspecific components of attention, we were alsointerested in overall differences in attentionalprocessing between meditators and control partici-pants. Therefore, in addition to the attentionnetwork analyses, groups were also comparedregarding grand mean RT and ES data, herebyprobing (efficiency in) attentional processing at amore general and integrated level. This analysismay reveal possible changes in strategy and, more-over, possible differences in overall attentionalprocessing, which cannot just be explained bychanges in strategy. In other words, an attemptis made to dissociate “just” a speed–accuracytrade-off (strategy difference between groups)from a qualitative difference in overall attentionalprocessing between groups). Since Chan andWoollacott (2007) had already suggested thatmindfulness meditation enhances one’s generalperformance on attentional tasks, we hypothesizedthat the meditators would show a better overallperformance on this attention task.

Method

ParticipantsA total of 20 mindfulness meditators (mean age48.1 years, SD 9.0, range 31–60 years; 9 male)and 20 control participants (mean age 48.1 years,SD 9.2, range 30–60 years; 9 male), who werematched to the mindfulness meditators in ageand gender, participated in this study. The meanlevel of mindfulness meditation experience washigh (mean period was 14.5 years, SD 11.1,range from 3 months to 35 years). Regular mind-fulness meditation practice varied from 60 to 420minutes each week. Mindfulness meditation iscomposed of both concentration meditation(samatha) and insight (vipasyana) meditation.Whereas during samatha meditation the prac-titioner is trained to maintain focus on an objectfor a theoretically unlimited period of time,during vipasyana meditation a typical kind ofmeta-awareness is trained.

1170 THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2010, 63 (6)

VAN DEN HURK ET AL.


None of the participants had any knownpsychological or neurological deficits. They allhad normal or corrected-to-normal vision. Asigned informed consent form was obtained fromeach participant before the experiment. Thestudy has been conducted according to the prin-ciples expressed in the Declaration of Helsinki.

Attention network testParticipants were seated in front of a 19-inchcomputer screen at a distance of 65 cm. Stimuliwere presented and responses were collected withPresentation software (Version 10.1 Neurobehav-ioral Systems, Albany, USA). Participants wereinstructed to respond as fast and as accurately aspossible to a target stimulus that was presented inthe centre of a horizontal row with five stimuli(see Figure 1C). The target stimulus was anarrow pointing either to the left or to the rightand was flanked by two flanker stimuli on eachside. Participants were instructed to press the leftmouse button with their left thumb or the rightmouse button with their right thumb as fast aspossible when the target arrow pointed to the leftor right, respectively. The four surroundingflanker stimuli were all arrows pointing in thesame or the opposite direction of the target stimu-lus or were just neutral stripes. The condition inwhich all five arrows pointed in the same directionwas called the congruent target condition. Thecondition in which the flanker arrows pointed inthe direction opposite to the target arrow wasnamed the incongruent target condition. The con-dition when the four flanker stimuli were stripeswas called the neutral target condition. The targetstimulus and the flanker stimuli were presented ata visual angle of 1.18 above or below a fixationcross presented in the middle of the screen.

The target stimulus could be cued in fourdifferent ways. In the first cueing condition, anasterisk was presented at the location of the fix-ation cross (¼ centre cue condition), and thetarget configuration was presented above orbelow the centre of the screen, with equal prob-ability. In the second cueing condition, two aster-isks were presented (¼ double cue condition); thetwo asterisks were presented at the fixed locations

of 1.18 of visual angle above and below the centreof the screen. Since the cue appeared 500 msbefore target onset (see Figure 1A), the cue pro-vided information on the timing of the targetstimulus. In the third cueing condition, an asteriskwas presented at the future location of the targetstimulus above or below the centre of the screen(¼ spatial cue condition). In this case, participantswere informed both on the timing and on thelocation of the target configuration. In the fourthcueing condition, no cue was given, and, as a con-sequence, participants had neither information onthe timing nor on the location of the upcomingtarget symbol.

The attention network test consisted of onetraining block with 24 trials and three test blockswith 94 trials each. After the first and secondblocks, participants took a break of a fewminutes, before starting the next block of theattention network test. A single trial consisted ofthe following: during a variable interval (VI, seeFigure 1A), ranging from 400–1,600 ms, a fix-ation cross was presented in the middle of thescreen. Then, depending on the cue condition, acue could be presented for 100 ms. Thereafter, acentral fixation cross was presented for 400 ms,followed by the target stimulus, which was pre-sented for 1,700 ms, or shorter if a response wasgiven within 1,700 ms. Finally, a fixation crosswas presented during a variable delay. Thelength of this delay was determined by subtractingthe RT plus 400 ms from the constant trial dur-ation that was 3,500 ms (see Figure 1A). All 12combinations of cueing (4) and target (3) con-ditions were presented in random order withineach block. Both RT and ES were measured.

Attention networks calculations and analysesFor each participant, all reaction times (RT) of aspecific condition, outside the range of the meanRT + 4 standard deviations of that specific con-dition, and RTs shorter than 100 ms were excludedfrom analysis. Then, the average RT for each of the12 conditions (4 cue conditions ! 3 targetconditions) was recalculated. Determination ofthe network effects was based on these RT data.Error scores (ES; percentage of incorrect




responses) of the 12 conditions were calculated bydividing the number of incorrect responses by thetotal number of responses for the specific conditionand multiplying this number by 100.

To calculate the alerting network effect meanRT and ES of the double cue condition were sub-tracted from the mean RT and ES of the no cuecondition. In such a way, the potentially beneficialeffect of an alerting cue on RT and ES was probed.The orienting network effect was calculated bysubtraction of the mean RT and ES of theresponses in the spatial cue condition from themean RT and ES of the responses in the centre

cue condition. This allowed us to probe the ben-eficial effect of spatial information, in addition totiming information. The executive network effectwas calculated by subtracting the mean RT andES in the congruent target condition from themean RT and ES in the incongruent target con-dition. In this way, the advantage of congruenceover incongruence in the target condition wasdetermined. Mean RT and ES of the threetarget conditions were averaged to calculate theorienting and alerting network effects, whereasmean RT and ES of the four cueing conditionswere averaged to calculate the executive network

Figure 1. Attentional network test paradigm. (A) During a variable interval (VI ¼ 400–1,600 ms) a central fixation cross is presented, andthe participant is instructed to look at it. Then a cue can be presented for 100 ms. (B) The four cue conditions are shown: no cue, central cue,double cue, and spatial cue. After presentation of the cue, a central fixation cross is shown for 400 ms, followed by the target stimulus. (C) Thethree different target configurations are shown: the neutral, congruent, and incongruent target configurations. The target is visible until theparticipant responds with a maximum of 1,700 ms. If reaction time (RT) is shorter than 1,700 ms, the stimulus is replaced by the centralfixation cross.


VAN DEN HURK ET AL.


effect. Normalized network effects were deter-mined by dividing the raw network effect by themean of the two conditions involved in thecalculation of the network effect: raw effect,R ¼ RTA–RTB; normalized effect, N¼ R/[(RTA þ RTB)/2], with RTx representing meanRT for condition x.3

In all tests, we verified the normal distributionof data. When a normal distribution was violated,we used nonparametric tests. The specific test isdescribed in each case. In all other cases we haveused paired t tests to investigate whether medita-tors differed from their matched controls. Inthose cases, where the hypothesis was very specific,as to the differences of responses of both groups forthe orienting and executive network scores, a one-tailed paired t test was used.

General behavioural analysesIn order to analyse differences between meditatorsand their matched controls on overall RT and ESdata, two paired tests were run. Grand mean RT/ES data of each participant were used in thesetests.

Results

Attention networks analysesA one-tailed paired t test on normalized RTnetwork data revealed no significant differencebetween meditators and controls in the executivenetwork, t(19) ¼ –0.779, p ¼ .22. However, a sig-nificant difference in the orienting networkt(19) ¼ –1.746, p , .05 (see Figure 2 andFootnote 3) was observed.

A one-tailed paired t test on ES network datashowed no significant difference in the orientingnetwork t(19) ¼ 1.185, p ¼ .13, whereas a trendtowards significance was observed in the executivenetwork, t(19) ¼ –1.560, p ¼ .07 (see Figure 3).This trend could be explained by the fact thatmeditators make significantly fewer errors in theincongruent target condition, Z(80) ¼ –2.403,

p , .05 (see Figure 4; nonparametric WilcoxonSigned Ranks Test was used to compare ES ofeach cue condition of each meditator in the incon-gruent target condition with the corresponding ESof a matched control; this yields a distribution of80 (4 cue ! 20 participants) paired data points).No significant differences were observed inboth the congruent, Z(80) ¼ –1.004, p ¼ .316,and neutral target condition, Z(80) ¼ –1.060,p ¼ .289 (analyses performed here are similar tothe incongruent condition analysis). Presumably,we are dealing with floor effects in these twotarget conditions, as ES approach 0% (seeFigure 4).

In conclusion, meditators showed a signifi-cantly smaller orienting network effect on RT

Figure 2.Mean normalized reaction time (RT) difference scores forall three networks. A significant difference between the results of themeditators and controls exists in the orienting network (p, .05).No significant difference was found for the executive and alertingnetworks. Error bars indicate 95% confidence intervals.

3 Normalized network data were calculated by dividing the absolute network effect by the mean RT of the two conditions used tocalculate the network effect. See Wang et al. (2005) for similar analyses of RT network scores.




data than did controls. And a trend towardssignificance in the same direction on ES data inthe executive network was observed.

General behavioural analysesA trend towards a significant difference betweengroups on grand mean RT data was observed,t(19) ¼ –2.008, p ¼ .059. Mindfulness meditatorsgenerally tended to respond slower (see Figure 5).

Also a trend towards a significant differencebetween groups on grand mean ES data wasobserved, t(19) ¼ –1.824, p ¼ .084. Mindfulnessmeditators generally tended to respond moreaccurately (see Figure 4).

At first, one is tempted to explain these behav-ioural differences by referring to the well-knowntrade-off function between speed and accuracy inbehavioural experiments (Posner, 1978). If so,

meditators would simply have put more emphasison accuracy (i.e., smaller ES), accepting longerRTs, and would therefore show a speed–accuracydistribution that could be obtained by extrapol-ation from the distribution of controls. Instead,a different, vertically shifted speed–accuracydistribution would imply a difference in the effi-ciency/functioning of attentional processing (seeFigure 6).

Overall attentional processing efficiencyRT-bin analysis. In order to investigate which ofthese two explanations fits our data, we dividedthe whole RT time range of both meditators andcontrols in small time windows. Within eachtime window, the responses of both groups canbe regarded as “equally” fast. For all windows,the total number of correct and incorrect responses

Figure 3. Mean error score (ES; %) for all three networks. A trendtowards a significant difference between the results of the meditatorsand controls is observed in the executive attention network. Nosignificant difference was found for the orienting and alertingnetworks. Error bars indicate 95% confidence intervals.

Figure 4. Mean error score (ES; %) for meditators (dashed line)and controls (solid line) in all 12 (4 cue ! 3 target) conditions.Meditators make fewer errors than controls. This effect is mostobvious in the incongruent condition. Error bars indicate 95%confidence intervals.


VAN DEN HURK ET AL.


and also the total number of responses of each ofthe two groups were calculated. Then, the percen-tage of incorrect responses for each group wasdetermined within each time window. Thisyields one data point (percentage of incorrectresponses) for each group for each time bin.These data points of all time bins of each grouprender a distribution of error percentage scoresfor that specific group.

This analysis was repeated with time windowsof 20, 40, and 60 ms, respectively, in order tocheck the robustness of the result. Only timewindows in which, at the group level, more than30 responses were given were included. Then,the error percentage distribution of the meditatorswas compared to the error percentage distributionof the controls (see Figure 7). Irrespective of thesize of the time window, a Wilcoxon Signed

Rank test revealed a significantly differentspeed–accuracy distribution of meditators fromthat of controls Z(10) ¼ –2.090, p, .05;Z(13) ¼ –2.510, p , .05; Z(23) ¼ –2.728,p , .01 (for 60-, 40-, and 20-ms time windows,respectively). For all three time windows, thedistribution of error percentages included at least94.6% of all data from both the control and themeditator group. This means that these resultsare based on a large amount of overlapping RTdata (see Figure 8 for RT distributions of bothgroups). Thus, considering the fact that meditatorsappear more accurate with equal RT, this resultsuggests a higher efficiency in attentional proces-sing in mindfulness meditators.

Logistic regression analysis. Another way of lookingat both RT and ES and a possible different

Figure 5. Mean reaction time (RT; ms) for meditators (dashedline) and controls (solid line) in all 12 (4 cue ! 3 target)conditions. Meditators have longer RTs than controls. Error barsindicate 95% confidence intervals.

Figure 6. Two hypothetical data distributions: (A) Meditatorsopted for a different strategy: longer reaction time (RT) andlower error score (ES). Distribution of meditators can be obtainedby extrapolation of data of controls. (B) Meditators differ inefficiency of attentional processing: They show a vertically shifteddistribution—that is, with identical RT meditators have lowerES. In this case, meditators would have a different distribution,which cannot be obtained by extrapolation.




speed–accuracy distribution between groups is thefollowing. One could see RT as a predictor ofthe probability to give a correct response, wherethe probability of a correct response increases withlonger RT (a view inspired by the well-knownspeed–accuracy trade-off).Next, groupmembershipcould also be important for the probability of acorrect response. Now, if only RT would appear tobe a predictor of the probability of a correct response,only a difference between the two groups in theperformance strategy could explain our generalbehavioural results. However, if, on top of thepredictive value of RT, group membership wouldalso appear predictive on the probability of acorrect response, intrinsic differences betweengroups in attentional processing must be present.

In order to distinguish between these two poss-ible explanations for the observed differences

between meditators and controls in RT and ES,we used a hierarchical logistic regression analysis.The type of response (correct vs. incorrect) wasdesignated as the dependent variable, and RTand group membership (meditator vs. control)were entered as predictors (RT in Block 1 andgroup in Block 2). Interestingly, group member-ship appeared a significant predictor in additionto RT: bRT ¼ –.009, p, .001, exp(B) ¼ 0.991;bgroup ¼ .284, p , .01, exp(B) ¼ 1.329. Thisresult points to an intrinsic difference betweenmeditators and controls in attentional processing.Therefore, results cannot be explained by “just”referring to a difference in performance strategybetween the two groups, but rather reflect amore efficient use of resources by the meditators,since meditators perform more accurately whenidentical RTs are considered.

Discussion

Attention networks analysisIn this study experienced mindfulness meditatorsshowed a better orienting of attention than theirmatched controls, as reflected by lower scoreson the orienting network. This decrease waspresent in RT data, but not in ES data (seeFigures 2 and 3).

Figure 7. Mean error score (ES; %) for meditators (dashed bars)and controls (white bars) for reaction times (RTs) within awindow of 40 ms in the range between 380 and 900 ms.Meditators make significantly fewer errors than controls forresponses with the same RT.

Figure 8. Distribution of reaction times (RTs) for meditators(black) and controls (white). Notice the large overlap of thedistributions. In general, meditators tend to have slightly longerRTs than controls.


VAN DEN HURK ET AL.


Our interpretation of this smaller orientingnetwork effect as reflecting amore flexible orientingof attention is based on the remarkmade by Fan andPosner: “For the orienting condition we generallyassume that larger orienting times arise because ofa difficulty in disengaging from the centre cue,where no target appears” (Fan & Posner, 2004,p. 212). Thus, it seems that meditators alreadyperform closer to their optimum in the centre cuecondition, since additional spatial informationdoes not seem to reduce their RT to the sameextent as in controls. However, one cannot comple-tely rule out the possibility that meditators justcannot benefit as much as controls from additionalspatial information, which should then be due tosome orienting attention impairment(s). In ourview, the presence of such a large number of partici-pants having orienting attention impairments in themeditator group, to produce such a systematic shiftfrom the control group, would be quite unlikely.

It seems more plausible that mindfulness med-itators, in general, show a more flexible orientingnetwork. This is because an important instructionduring the practice of mindfulness meditation is tocontinue to successively detach–attach–detachattention to all objects passing by in the receptivefields, thereby presumably improving the flexibilityof the orienting of attention. Jha et al. recentlynoted that “the practice of repeatedly engaging,disengaging and moving, instantiates the orientingor ‘shift’ operation of attention” (Jha et al., 2007;Posner & Badgaiyan, 1998; Posner & Gilbert,1999)—that is, a statement that supports ourinterpretation.

Besides the better orienting of attention, also atrend towards a significantly better executive atten-tion was revealed in our study, as meditatorsshowed smaller executive network ES. No differ-ence in executive attention was found consideringRT data. The trend in the executive network canbe explained by the fact that controls make signifi-cantly more errors in the incongruent condition(see Figure 4). Notably, the incongruent targetcondition appears to be the only target conditionin which the results of the two groups are signifi-cantly different from each other considering ES;we are probably dealing with floor effects in the

other two target conditions, as ES approaches 0%(see Figure 4) in both groups. Similar flooreffects have also been obtained by Fan et al. (2002).

It is of interest that we found a better orientingand a better executive attention as this had notbeen found in the study by Jha et al. (2007; whofound no difference in the orienting of attention).This difference might be explained by the fact thatin the present study more experienced meditatorswere tested and that there was a better matchbetween meditators and control participants.

Over the past years, several studies have relatedthe functioning of the executive network to thefunctioning of the orienting network (Callejaset al., 2005; Callejas et al., 2004; Funes &Lupianez, 2003). The orienting network appearsto exert a positive influence on the executive atten-tion network, as the flanker effect is reduced inspatially cued trials in comparison to uncuedtrials (Callejas et al., 2005; Callejas et al., 2004).Considering this positive interaction, the betterexecutive attention in the meditators can partiallybe explained by the better functioning of theirorienting network. In addition, meditators alsomight profit from an increased ability to focusattention to the relevant information embeddedin the target condition, while ignoring or detach-ing faster from the irrelevant information signalledby the target configuration.

Better executive attention in meditators has alsobeen reported by Jha et al. (2007) and Chan andWoollacott (2007). Chan and Woollacottexplained the effect of meditation on attention byreferring to the possibility of an increased abilityto focus attention and an improved inhibition ofautomatic responses (for example, inhibiting theshift of attention to distracting externally gener-ated stimuli or internally generated thoughts) infavour of the desired response (remaining focusedon the desired object or task). This explanation issimilar to our interpretation given above.

As there is a large body of neuropsychologicalliterature showing that (impaired) executive atten-tion is the key element for self-regulation of cog-nition and emotion (Fernandez-Duque, Baird, &Posner, 2000; Posner & Rothbart, 1998; Tanget al., 2007), it would be interesting for future




studies to explore possible changes in orientingand executive attention, induced by mindfulnessmeditation, in clinical populations. As attentionalbiases in information processing have beendemonstrated to be important in patients withanxiety and depression (Williams, Watts,MacLeod, & Mathews, 1997), the effectivenessof mindfulness-meditation-based practices mightbe mediated by improvements in attentional pro-cessing, especially in executive attention.

Overall analysis of attentional processingIf we consider the grand mean RT and ES, theattention network test allows a distinctionbetween the performance of meditators and con-trols at a more general and integrated level ofattentional processing, as meditators appeared tohave systematically longer RT and, at the sametime, showed a systematic increase in accuracy(decrease in ES) as compared to controls (seeFigures 4 and 5). These systematic shifts acrossalmost all conditions cannot be explained, in ourview, by specific changes limited to the alerting,orienting or executive component of attention.They can better be explained as the result of amore general and persistent change in attentionalprocessing, which requires an integrated analysis.

At first sight, one could interpret the differ-ences in RT and ES as a good example of thewell known speed–accuracy trade-off effect(Posner, 1978). In that case, the differencescould be explained by the fact that meditatorstake more time to respond and consequentlybecome more accurate. This result by itselfwould already be of interest, showing mindfulnessmeditation to be related to a systematic change inthe performance strategy for such a low-level task.

Alternatively, the differences could actuallyreflect a higher “efficiency” in attentional proces-sing in the mindfulness meditation group. Wewould like to refer to the term “efficiency” whenwe want to take into consideration the overallperformance on this attention task—that is, toconsider both RT and ES at the same time.In our view, these two performance dimensionstogether are very informative on the efficiencyin the use of mental resources, as the efficiency in

functioning of a mental operation is reflected byscores on both these dimensions.

Without any additional analyses, no inferencecould be made to distinguish the performance ofmeditators from that of controls, considering theefficiency of attentional processing. To introducea measure of efficiency in which both RT andES are considered, we opted for the strategy toanalyse ES for responses with the same RT (seeRT-bin analysis and logistic regression analysisin the Results section for details).

Both the RT-bin analysis and logisticregression analysis showed meditators to be moreaccurate when identical RTs are considered. Thisgripping result, in our view, reflects a generalincrease in efficiency in attentional processing inmeditators (see Figure 7). As such, it provides evi-dence for the suggestion by Chan and Woollacott(2007) that mindfulness meditation enhancesone’s general performance on attentional tasks.

Conclusions

Wehave found that (a)meditators showa significantbetter orienting of attention and a trend towards asignificantly better executive attention and that (b)meditators show a significantly higher degree ofattentional processing efficiency than controls.

The present study can be considered as ananswer to Posner’s observation that “training ofattention either explicitly or implicitly is some-times a part of the school curriculum, butadditional studies are needed to determineexactly how and when attention training can bestbe accomplished and its long-lasting importance”(Posner & Rothbart, 2007, p. 13).

Despite of our study being cross-sectional, ourresults seem to point to the possibility to trainour attentional system through an extensivemental training, known as mindfulness medita-tion, in order to obtain a systematic and sustainedbetterment in attentional processing as comparedto normal functioning.

Original manuscript received 11 February 2009Accepted revision received 17 July 2009

First published online 29 September 2009


VAN DEN HURK ET AL.


REFERENCES

Atin, J. A. (1997). Stress reduction through mindfulnessmeditation. Effects on psychological symptomato-logy, sense of control, and spiritual experiences.Psychotherapy and Psychosomatics, 66(2), 97–106.

Baer, R. A. (2003). Mindfulness training as a clinicalintervention: A conceptual and empirical review.Clinical Psychology: Science and Practice, 10, 125–143.

Bishop,S.R., Lau,M., Shapiro, S.,Carlson,L.,Anderson,N. D., Carmody, J., et al. (2004). Mindfulness: Aproposed operational definition. Clinical Psychology:Science and Practice, 11, 230–241.

Callejas, A., Lupianez, J., Funes, M. J., & Tudela, P.(2005). Modulations among the alerting, orientingand executive control networks. Experimental BrainResearch, 167, 27–37.

Callejas, A., Lupianez, J., & Tudela, P. (2004). Thethree attentional networks: On their independenceand interactions. Brain and Cognition, 54, 225–227.

Chan, D., & Woollacott, M. (2007). Effects of level ofmeditation experience on attentional focus: Is theefficiency of executive or orientation networksimproved? Journal of Alternative and ComplementaryMedicine, 13(6), 651–657.

Fan, J., McCandliss, B. D., Fossella, J., Flombaum, J. I.,& Posner, M. I. (2005). The activation of attentionalnetworks. NeuroImage, 26, 471–479.

Fan, J., McCandliss, B. D., Sommer, T., Raz, A., &Posner, M. I. (2002). Testing the efficiency andindependence of attentional networks. Journal ofCognitive Neuroscience, 14, 340–347.

Fan, J., & Posner, M. I. (2004). Human attentionalnetworks. Psychiatrische Praxis, 31, 210–214.

Fernandez-Duque, D., Baird, J. A., & Posner, M. I.(2000). Executive attention and meta-cognitiveregulation. Consciousness and Cognition, 9, 288–307.

Funes, M. J., & Lupianez, J. (2003). La teorıa atencionalde Posner: Una tarea para medir las funcionesatencionales de Orientacion, Alerta y ControlCognitivo y la interaccion entre ellas [Theattentional theory of Posner: A task to measure thefunctioning of the orienting, alerting and executiveattention and the interaction between them].Psicothema, 15(2), 260–266.

Giommi, F. (2006). Mindfulness and its challenge tocognitive-behavioral practice. In M. G. T. Kwee,K. J. Gergen, & F. Koshikawa (Eds.), Horizons inBuddhist psychology. Practice, research & theory(pp. 209–224). Chagrin Falls, OH: Taos InstitutePublications.

Grossman, P., Niemann, L., Schmidt, S., &Walach, H.(2004). Mindfulness-based stress reduction andhealth benefits. A meta-analysis. Journal ofPsychosomatic Research, 57(1), 35–43.

Hovels-Gurich, H. H., Konrad, K., Skorzenskia, D.,Herpertz-Dahlmann, B., Messmer, B. J., &Seghaye, M. C. (2007). Attentional dysfunction inchildren after corrective cardiac surgery in infancy.Annals of Thoracic Surgery, 83, 1425–1430.

James, W. (1890). The principles of psychology. New York:Holt.

Jha, A. P., Krompinger, J., & Baime, M. J. (2007).Mindfulness training modifies subsystems of atten-tion. Cognitive, Affective & Behavioral Neuroscience,7(2), 109–119.

Kabat-Zinn, J. (1990). Full catastrophe living.New York: Bantam Doubleday Dell.

Kabat-Zinn, J. (1994). Wherever you go, there you are:Mindfulness meditation in everyday life. New York:Hyperion.

Kabat-Zinn, J. (2003). Mindfulness based interventionsin context: Past, present, and future. ClinicalPsychology: Science and Practice, 10, 144–156.

Kabat-Zinn, J., Lipworth, L., & Burney, R. (1985). Theclinical use of mindfulness meditation for the self-regulation of chronic pain. Journal of BehavioralMedicine, 8, 163–190.

Kabat-Zinn, J., Massion, A. O., Kristeller, J., Peterson,L. G., Fletcher, K. E., Pbert, L., et al. (1992).Effectiveness of a meditation-based stress reductionprogram in the treatment of anxiety disorders.American Journal of Psychiatry, 149, 936–943.

Kristeller, J. L., & Hallett, C. B. (1999). An exploratorystudy of a meditation-based intervention for bingeeating disorder. Journal of Health Psychology, 4,357–363.

Leskin, L. P., & White, P. M. (2007). Attentionalnetworks reveal executive function deficits in post-traumatic stress disorder. Neuropsychology, 21(3),275–284.

Neuhaus, A. H., Koehler, S., Opgen-Rhein, C.,Urbanek, C., Hahn, E., & Dettling, M. (2007).Selective anterior cingulate cortex deficit duringconflict solution in schizophrenia: An event-relatedpotential study. Journal of Psychiatric Research, 41,635–644.

Posner, M. I. (1978). Chronometric explorations of mind.Hillsdale, NJ: Lawrence Erlbaum Associates.

Posner, M. I., & Badgaiyan, R. D. (1998). Attentionand neural networks. In R. W. Parks & D. S.Levine (Eds.), Fundamentals of neural network




modeling: Neuropsychology and cognitive neuroscience(pp. 61–76). Cambridge, MA: MIT Press.

Posner, M. I., & Gilbert, C. D. (1999). Attention andprimary visual cortex. Proceedings of the NationalAcademy of Sciences, USA, 96, 2585–2587.

Posner, M. I., & Petersen, S. E. (1990). The attentionsystem of the human brain. Annual Review ofNeuroscience, 13, 25–42.

Posner, M. I., & Rothbart, M. K. (1998). Attention,self-regulation and consciousness. PhilosophicalTransactions of the Royal Society of London. Series B,Biological Sciences, 353, 1915–1927.

Posner, M. I., & Rothbart, M. K. (2007). Research onattention networks as a model for the integrationof psychological science. Annual Review ofPsychology, 58, 1–23.

Shapiro, S. L., Carlson, L. E., Astin, J. A., & Freedman,B. (2006). Mechanisms of mindfulness. Journal ofClinical Psychology, 62(3), 373–386.

Slagter, H. A., Lutz, A., Greischar, L. L., Francis, A. D.,Nieuwenhuis, S., Davis, J. M., et al. (2007). Mentaltraining affects distribution of limited brainresources. PLoS Biology, 5(6): e138. doi:10.1371/journal.pbio.0050138

Speca, M., Carlson, L. E., Goodey, E., & Angen, M.(2000). A randomized, wait-list controlled clinical

trial: The effect of a mindfulness meditation-basedstress reduction program on mood and symptomsof stress in cancer outpatients. PsychosomaticMedicine, 62, 613–622.

Tang, Y. Y., Ma, Y., Wang, J., Fan, Y., Feng, S., Lu, Q.et al. (2007). Short-termmeditation training improvesattention and self-regulation. Proceedings of theNational Academy of Sciences,USA, 104, 17152–17156.

Teasdale, J. D., Segal, Z. V., & Williams, J. M. G.(1995). How does cognitive therapy prevent depress-ive relapse and why should attentional control(mindfulness) training help? Behaviour Research andTherapy, 33, 25–39.

Teasdale, J. D., Segal, Z. V., Williams, J. M., Ridgeway,V. A., Soulsby, J. M., & Lau, M. A. (2000).Prevention of relapse/recurrence in major depressionby mindfulness-based cognitive therapy. Journal ofConsulting and Clinical Psychology, 68, 615–623.

Wang, K., Fan, J., Dong, Y., Wang, C. Q., Lee, T. M.,& Posner, M. I. (2005). Selective impairment ofattentional networks of orienting and executivecontrol in schizophrenia. Schizophrenia Research, 78,235–241.

Williams, J. M. G., Watts, F. N., MacLeod, C., &Mathews, A. (1997). Cognitive psychology and theemotional disorders (2nd ed.). New York: Wiley.


VAN DEN HURK ET AL.


meghan searl, ph.d., abpp-cn - home -...

Documents