psychophysiological measurement of personality measurement of personality eco de geus and david l....

PsychophysiologicalMeasurement of Personality

Eco de Geus and David L. Neumann

INTRODUCTION

Whereas the environment is in constant fluxand causes changes in an individual’s behaviorover time, personality is considered the con-stant factor that causes stability in behavior.This constancy is likely to be hard-wired inour biology, and the brain seems the obviousplace for such hard-wiring. Can we measurethis hard-wiring of behavior? In this chapter wereview attempts to correlate personality traitsto individual differences in central nervoussystem functioning using psychophysiologicalrecording techniques. This relatively smallliterature has been motivated by two differentgoals. The first has emerged from concernsabout the validity of paper-and-pencil person-ality assessment (Eisenberger et al., 2005).Virtually all major personality inventories arebased on potentially flawed subjective linguis-tic self-report. To further advance personalitytesting, it may be necessary to move to objec-tive tests that avoid most of the motivationaland response distortion associated with itemtransparency of self-report instruments.Psychophysiological testing seems a promisingmethod to do so, because voluntary controlover the recorded biological signals is limited

if not absent. A second motive to use psy-chophysiological testing in personalityresearch is to elucidate the biological processesunderlying the major dimensions of personal-ity. Several influential models of personality,such as those by (Eysenck 1967, 1990; Eysenckand Eysenck, 1985) and (Gray 1982; Grayand McNaughton, 2000) have been stronglyinformed by biological theory.

The two central constructs in Eysenck’stheory are neuroticism and extraversion.Neuroticism is related to activation levels inthe limbic system. The limbic system includesthe hippocampus, amygdala, septum, andhypothalamus and regulates emotional statessuch as fear and aggression. Neurotic individ-uals are hypothesized to have higher activationlevels in the limbic system leading to lowerthresholds for emotional responses. Neuroticindividuals are more likely to report feelings ofanxiety, guilt, and tension. Emotionally stableindividuals, on the other, hand, have lower acti-vation and higher thresholds in the limbicsystem, leading to attenuated responses toemotional challenges. The extraversion–introversion dimension is thought to be regulated by the ascending reticular acti-vating system that is involved in regulating

15

9781412946520-Ch15 29/3/08 12:07 PM Page 313

cortical arousal. An individual’s comfort levelat any given time will depend on the interac-tion between their basal cortical arousal andthe type of situation they are in; being under-or over-aroused are both less desirable than amoderate level of arousal. Introverts are proneto be overstimulated by sensory stimuli andconsequently tend to withdraw from social sit-uations, are less active, and less willing to takerisks. In contrast, due to their lower base levelof cortical arousal, extraverts tend to be morelively and sensation seeking and will beattracted to social situations.

In the view of Gray and colleagues, thereare two basic systems that control behavior.The behavioral inhibition system (BIS) is acti-vated by novelty and stimuli associated withpunishment; that is, aversive stimuli or omis-sion of reward. The behavioral approachsystem (BAS) is activated by stimuli associatedwith reinforcement; that is, reward or termi-nation of punishment. Gray and colleagueslocated these systems in the septohippocam-pal system (to which the amygdala was lateradded) and the ventral striatum (including thenucleus accumbens). Engagement of both sys-tems is associated with arousal as reflected inchanges in autonomic nervous system activityand hormonal secretion. Gray and colleaguesproposed that the most salient individual differences reflect the variation in sensitivityto stimuli associated with punishment or rein-forcement and the coupled behavioral tenden-cies of avoidance and approach. Specifically,individual differences in the functioning ofthe reward system in response to appetitivestimuli are implicated in the personality traits ofextraversion and novelty seeking/impulsivity.Individual differences in the punishmentsystem in response to aversive stimuli areimplicated in the personality traits of neuroti-cism and harm avoidance.

Psychophysiological testing

Psychophysiological testing can be used tocorroborate the hypothesized biological correlates of personality, and ultimately chart

the various intermediate steps in the biologicalpathways connecting variation in brain functionto variation in behavior. In the current chapter,we will focus on the electromyographic (EMG)recording of the startle blink reflex, electroen-cephalographic (EEG) recording of corticalelectrical activity either recorded continuouslyor evoked by stimulus events, and changes inbrain blood flow assessed by functional mag-netic resonance imaging (fMRI). These signalscan, first of all, be recorded under pure restingconditions. In such a setting, subjects are typ-ically instructed to relax, focus their attentionon a fixation point and in general made toavoid engaging in any particular state of mind.In keeping with a dispositional model of per-sonality, the idea is that this will allow their‘underlying’ or ‘true’ level of nervous systemactivity to manifest. For example, if extravertedindividuals are characterized by positive and neurotic individuals by negative affect asassessed by affect scales like the PANAS(Watson et al., 1988) these differences in dis-positional affect should be detectable by themeasurement of resting brain blood flow andelectrical activity in brain regions known to beimplicated in affective processing. A numberof studies have indeed reported such correla-tions (Davidson, 1998; Ebmeier et al., 1994;Stenberg et al., 1990, 1993; Youn et al., 2002).

By contrast, Wallace (1966) conceptualizedpersonality attributes as abilities, an approachhe termed the capability model of personality.Others have proposed similar formulations ofpersonality (Mischel and Shoda, 1995). Thecapability model encourages the measurementof individual differences in ‘brain behavior’during controlled laboratory challenges, muchas one might test intelligence or high-jumpingability. Subjects can, for instance, be made to anticipate and actually gain a reward (e.g. 5 cents per correct answer) or be punished (e.g.shock or monetary loss) during a mentallychallenging task. Alternatively, they can beshown series of words with strong negative(e.g. ‘sick’, ‘murder’, ‘hate’, ‘rape’) or positiveconnotations (e.g. ‘happy’, ‘wedding’, ‘love’,‘party’). Two popular sets of affective stimuliare a series of faces compiled by Ekman and

314 THE SAGE HANDBOOK OF PERSONALITY THEORY AND ASSESSMENT

9781412946520-Ch15 29/3/08 12:07 PM Page 314

Friesen (1978) showing primary emotions likesadness, happiness, fear or anger, and a seriesof images compiled by Lang et al. (1998, 2001)in the International Affective Picture System(IAPS). The IAPS stimuli are complex visualscenes that were extensively normalized foremotional valence and arousal. For instance,unpleasant images depicted snarling dogs,spiders, sharks, disgusting objects, violence,severe burns, or corpses, whereas pleasantimages depicted happy babies, appetizingfoods, puppies and kittens, joyful people andloving couples. Response to these emotionalvalenced images are usually contrasted to thoseseen during neutral images of, for instance, abasket, books, or fractal images.

Most of the recent work on EEG, ERP, andEMG startle blink correlates of personalityand almost all of the fMRI studies, follow thecapability model rather than the dispositionalmodel. Instead of assessing differences inresting levels of the EMG startle reflex, EEGor hemodynamic brain activity, these studieshave looked at the magnitude, timing, andtopography of changes in these measuresduring experimentally induced changes inpsychological state.

EMG startle reflex

The startle reflex is a defensive responseelicited by intense and abrupt stimuli. Inhumans, the startle reflex is measured byrecordings of the orbicularis oculi musclearound the eye (Blumenthal et al., 2005).Blinks can be reliably elicited by presentinga brief (about 50 ms) and moderately intensetone, or white noise. Typically, participantsare presented with a lead stimulus that mightbe neutral or emotionally toned and the star-tle eliciting probe stimulus is presented at acertain time, termed a lead interval, followingthe lead stimulus onset. Startle modulation isobserved when the size of the startle reflex isaltered by the lead stimulus. Lead intervalsless than 50 ms can produce very short leadinterval facilitation (Neumann et al., 2004).Lead intervals of 50 to 500 ms produce short

lead interval inhibition, often termed pre-pulse inhibition (PPI), with maximal inhibitionat lead intervals around 100 ms (Neumann et al., 2004).

EEG power and EEG asymmetry

The EEG reflects the electrical activity gen-erated by clusters of neurons in the cortex thatshow synchronized changes in membranepotential due to neural activity in that brainlocation. EEG recordings are made from mul-tiple electrodes affixed to the scalp (the exactnumber varies across studies from the standard18 scalp electrodes up to 128 scalp electrodes).The most striking feature of the brain activityrecorded by an EEG is its oscillatory character.Quantification of EEG data reflects this, in thatthe energy (or power) in various frequencybands is used as a main index. EEG record-ing has many important clinical applicationsbecause there are predictable EEG signaturesassociated with different behavioral states. Arelaxed resting state, for instance, is charac-terized by high power in the alpha (8–12 Hz)frequency band, whereas under-condition ofmental load power in the beta band (12–30 Hz)increases in relative strength. In addition to itssensitivity to within-subject changes in behav-ioral state, EEG power can also be comparedacross subjects, for instance as a function ofpersonality. Because right and left hemi-spheres are suspected to play a differential rolein emotional processing, most attention hasgone to individual differences in the degreeof asymmetry in left and right EEG power(Sutton and Davidson, 1997).

Event-related potentials

The event-related potential (ERP) is a stereo-typed short-term change in EEG activity thatis time-locked to stimulus or cognitive events.Different ERP components can be identifiedaccording to whether the voltage fluctuationis positive or negative and the time (latency)at which the peak voltage fluctuation is seen.

PSYCHOPHYSIOLOGICAL MEASUREMENT OF PERSONALITY 315

9781412946520-Ch15 29/3/08 12:07 PM 15


For instance, the P300 (or P3) reflects a pos-itive voltage change that peaks around 300ms following stimulus onset. These changesin amplitude and timing of the voltage can befurther differentiated as a function of spatiallocation. Thus, it can be determined whetherindividuals with different personality traitsshow different latencies, amplitudes, or pat-terns in the spatial distribution of their ERPs.The ongoing nature of psychologicalprocesses means that the ERP will consist ofseveral potentially overlapping components(Fabiani et al., 2000). The major ERP com-ponents are N1, P2, N2, and P3, followed byslow wave components that return the EEGsignal to baseline. Various other componentsmay also be identified according to specificexperimental conditions, such as the error-related negativity (ERN) and the contingent-negative variation (CNV).

fMRI

In a typical fMRI experiment, a high-resolu-tion structural scan is taken initially thatallows separation of gray and white matterfrom each other and other tissues (cere-brospinal fluid, skull, skin, etc.). Next, thesubject is exposed to repeated stimuli, forexample, pictures from the IAPS. Whennerve cells are activated by these stimuli theyconsume extra oxygen which is carried byhemoglobin in red blood cells from localcapillaries. The local response to this oxygenutilization is decrease in oxygenated bloodfollowed by an increase in blood flow, occur-ring after a delay of approximately 1–5 sec-onds. Corresponding changes in the relativeconcentration of oxyhemoglobin and deoxy-hemoglobin can be detected using an appro-priate magnetic resonance pulse sequencethat gives rise to a so-called blood oxygena-tion level dependent (BOLD) contrast. Theaverage level of BOLD signal intensity canbe compared in two sets of stimuli, forinstance aversive pictures versus neutral pic-tures, to see which parts of the brain were

specifically activated or deactivated by theaversive stimulus compared to the neutralstimulus. Statistical maps that take intoaccount the multiple testing across thousandsof 1–1.5 mm3 blocks of brain tissue –voxels– show the (de)activated areas as colored

blobs. These can be rendered in 3D and plot-ted on top of the original high-resolutionstructural scan for anatomical interpretation.Individual differences in the BOLD signalintensity in certain regions-of-interest (ROI)can be correlated with major personalitytraits like neuroticism and extraversion.

NEUROTICISM

EMG startle reflex

Response size and habituation As a defensive response, the absolute size ofthe startle reflex should be larger and habitu-ate more slowly in individuals with high neu-roticism because of the association of thispersonality trait to hypersensitivity to aver-sive sensory stimuli (Eysenck et al., 1985).However, results have generally failed tosupport this prediction. No association hasbeen found between startle reflex magnitudeand neuroticism in a sample of male patientswith schizophrenia or matched healthy con-trols (Akdag et al., 2003). In addition, thesame study found no association betweenneuroticism and the rate of habituation froman early to a late block of trials. A similarfailure to find any relationship between neu-roticism and startle reflex magnitude has alsobeen reported in other investigations (Hawkand Kowmas, 2003; Kumari et al., 1996).

Prepulse inhibition (PPI)In contrast to the absolute size of the startlereflex, PPI has shown a stronger associationwith neuroticism. Prepulse inhibition is anindex of automatic sensorimotor gating thatserves to protect the processing of the prepulsefrom interruption from the startle stimulus

9781412946520-Ch15 29/3/08 12:07 PM Page 316

(Graham, 1979). High levels of neuroticismare associated with low levels of PPI at leadintervals of 30, 60, and 120 ms and the asso-ciation may be stronger in earlier than in latetrials (Swerdlow et al., 1995; Corr et al., 2002).This reduced PPI may reflect the hypersensi-tivity to aversive sensory stimuli thought to bepresent in neurotic individuals (Eysenck et al.,1985). Alternatively, neurotic individuals mayallocate less attention to the processing of theprepulse and startle-eliciting stimulus, thusreducing the amount of PPI observed (Corr etal., 2002). Based on this explanation, it wouldbe instructive to directly manipulate the par-ticipant’s attention such as by using the dis-crimination and counting task (Filion et al.,1994). If PPI is attenuated in neurotic indi-viduals during attended, but not ignored leadstimuli, it would suggest that the associationbetween neuroticism and PPI is mediated byattention. A final reason for the associationbetween neuroticism and PPI could be that it reflects the activation of emotion circuitsduring the task, which in turn reduces PPI dueto overlapping neural systems between the circuits that govern PPI and emotion (Corr et al., 2002).

Affective modulation According to the motivational priming inter-pretation of affective startle modulation (Langet al., 1997), there will be startle potentiationwhen the lead stimulus induces an affectivestate congruent with the startle stimulus (i.e.unpleasant or aversive lead stimuli) and startleattenuation when the lead stimulus induces anincongruent affective state (i.e. pleasant orappetitive lead stimuli). Hence, affective startlemodulation will be influenced by personalitycharacteristics if these characteristics mediateresponses to aversive or appetitive stimuli. Inparticular, individuals high in neuroticism oron the BIS scale would be expected to showenhanced startle potentiation due to theirheightened sensitivity to aversive stimuli(Eysenck et al., 1985; Gray and Me Naughton.,2000). However, prior research has not always been consistent with these predictions.Startle latency potentiation during unpleasant

compared to neutral pictures has been found inparticipants low in neuroticism, but not inthose high in neuroticism (Corr et al., 1995).High BIS participants have shown increasedattenuation of startle magnitude during pleas-ant pictures compared to neutral pictures,which was not found in low BIS participants(Hawk et al., 2003). However, the expectedstartle potentiation during unpleasant com-pared to neutral pictures in high BIS partici-pants was not found. Absence of a relationshipbetween neuroticism and affective startle mod-ulation has also been reported (Kumari et al.,1996). Various explanations have been put for-ward to explain why the results are inconsis-tent with predictions, including ceiling effectsor reduced attention to unpleasant stimuli inneurotic individuals (Corr et al., 1995), or thatit reflects the nature of stimuli used in terms of their arousal level (Kumari et al., 1996) orcontent (Kumari et al., 1996; Hawk andKowmas., 2003).

The nature of the unpleasant stimuli may bea crucial factor in the association between neu-roticism and affective startle modulation.Startle potentiation is greater when stimuli arefearful than when disgusting (Kaviani et al.,1999). Moreover, the amygdala is implicatedin the fear response and it plays a central rolein startle potentiation (Lang et al., 1997). TheBIS is closely associated with anxiety and fearand the amygdala has been included in this for-malization (Gray et al., 2000). Accordingly, acloser association between neuroticism andaffective startle modulation may be found ifimages of fear, rather than disgust, are used. Ina re-examination of the data reported byKumari et al. (1996), it was found that low-neuroticism participants showed greater startlepotentiation to low fear–high disgust film clipsthan high-neuroticism participants (Wilson etal., 2000). In a study that systematicallymanipulated the nature of unpleasant pictures,high-BIS participants, but not low-BIS partici-pants, exhibited startle potentiation during fearpictures (Caseras et al., 2006). The high-BISparticipants showed significant differences instartle modulation between pleasant and neu-tral, pleasant and blood-disgust, and pleasant


9781412946520-Ch15 29/3/08 12:07 PM Page 317

and fear pictures. The low BIS participantsshowed significant differences in startle modu-lation between pleasant and neutral, pleasantand blood-disgust, and blood-disgust and fearstimuli. The findings support the conclusionthat individual differences in BIS functioning(and by extension, neuroticism) is associatedwith startle modulation during fear elicitingscenes.

EEG asymmetry

One of the most widely studied correlates ofpersonality traits is frontal EEG alpha asym-metry; that is, asymmetry in EEG activity inthe alpha frequency range (8–12 Hz) on frontalelectrodes (Coan and Allen, 2004; Davidson,2004; Davidson et al., 1985; Fox, 1991; Hewiget al., 2006). From their pioneer studiesonward, Davidson and colleagues have linkedfrontal cortical asymmetry to an approachand a withdrawal system which bear strongresemblance to Gray’s BAS and BIS system(Sutton et al., 1997). The approach system isactivated by the perception of goals, elicitsapproach related (pre-goal-attainment) posi-tive affect, and initiates appetitive behaviortowards these goals. The neuroanatomicalbasis of the system is located in the left dor-solateral and medial prefrontal cortex and thebasal ganglia. The withdrawal system is acti-vated by aversive stimulation, elicits negativeemotions, and leads to withdrawal behavior.The neuroanatomical basis of this system isthought to be the right dorsolateral prefrontalcortex, the right temporal polar region, theamygdala, the basal ganglia, and the hypo-thalamus. The amygdala may be a centralstructure connecting the activity of both systems (Ochsner et al., 2002).

Although EEG asymmetry seems a perfectcandidate to act as a psychophysiologicalindicator of neuroticism, the extant literatureshows large inconsistencies in statisticalassociations between frontal EEG asymme-try and neuroticism (and other related traitmeasures of personality), and it is unlikelythat the methodological differences acrosslaboratories can completely account for those

inconsistencies (Allen et al., 2004; Coan and Allen., 2004; Davidson, 2004; Hagemann,2004; Smit et al., 2007). From a conceptualpoint of view, Coan and colleagues (Coan et al., 2006) question the wisdom of the wide-spread use of measuring frontal asymmetry inresting conditions. The use of resting condi-tions derives from a near axiomaticallyaccepted dispositional model of frontal affec-tive style in this field. In this model individu-als are thought to possess a general tendencyto predominantly respond with eitherapproach-related affect (indexed by relativelygreater left frontal activity) or withdrawal-related affect (indexed by relatively greaterright frontal activity) across all or most situa-tions. Instead, the capability model of frontalEEG asymmetry may be more appropriate(Coan et al., 2006). As outlined before, thismodel posits that meaningful individual dif-ferences in frontal EEG asymmetry exist, butthat those individual differences are bestthought of as interactions between the emo-tional demands of specific situations and theemotion-regulatory abilities individuals bringto those situations. During emotional chal-lenges, individual differences in frontal EEGasymmetries were indeed shown to be morepronounced than during a resting condition.Moreover, they were much more reliable; thatis, more resistant to measurement errorinduced by variation in EEG methodology,and had a more reliable relationship with cri-terion measures of ongoing emotional state.These findings suggest that future use ofevoked changes in EEG frontal asymmetry byemotion induction rather than resting EEGmay yield more robust links to personality.

ERP

Early ERP components Early ERP components can be influenced bythe sensory characteristics of stimuli, such astheir intensity. Based on the hypothesizedhypersensitivity to sensory stimuli, neuroticismis expected to influence these components.However, research has generally failed to find


9781412946520-Ch15 29/3/08 12:07 PM Page 318

an association between neuroticism and N1,P2, and N2 amplitude or latency (De Pascalis,1993; De Pascalis et al., 1996; Fjell et al.,2005). The error-related negativity (ERN),however, may be influenced by neuroticism.The ERN occurs when participants make errorsin a sensorimotor task (Luu et al., 2000) orwhen outcomes are ‘worse than expected’(Holroyd and Coles, 2002) and is thought tobe generated by the anterior cingulate cortex(Van Veen and Carter, 2002). The negativitypeaks around 150 ms following response onsetand shows a fronto-central scalp distribution(Dehaene et al., 1994). Adult participants whoscored high on measures of negative affectand emotionality have shown larger ERN thanparticipants with low scores during a visualflanker task (Luu et al., 2000). No relationshipbetween neuroticism and ERNs at a fronto-central site during a visual flanker task has alsobeen reported (Santesso et al., 2005), althoughthe lack of an association may reflect the useof 10-year-old children in this study.

P3 The P3 can be elicited in an oddball task inwhich infrequently presented pure tones (odd-balls) are randomly interspersed among fre-quently occurring tones of a different pitch(standards). Stelmack and Houlihan (1995)present a comprehensive review of the earlierresearch on P3 and personality. In one of theirown studies, participants high in neuroticismhad a shorter P3 latency than participants lowin neuroticism (Stelmack et al., 1993). Thedifference may reflect that the former spendless time to evaluate a stimulus (Plooij-vanGorsel, 1981). Alternatively, the increasedworrying and susceptibility to stress, whichare regarded as important elements of neu-roticism that adversely influence cognitiveperformance (Eysenck et al., 1985), may alsoinfluence P3 latency (Stelmack et al., 1993).In addition, neuroticism level produced con-trasting results for P3 latency and RT in thathigh neuroticism was associated with fasterP3 latency but slower RT. This finding wasinterpreted to reflect that individuals with high neuroticism have a hasty and worried

evaluation of a stimulus (short P3 latency)that requires additional processing or check-ing to initiate a response leading to long RT(Stelmack et al., 1993).

No relationship between P3 and neuroti-cism has been found in a word/non-worddetection task (De Pascalis et al., 1996), anauditory oddball task (Polich and Martin,1992; Pritchard, 1989), and auditory startleprobes presented during emotionally tonedslides (Bartussek et al., 1996). It is possiblethat some discrepant results reflect differ-ences in the task requirements or method-ological features of the experiment. Forinstance, using a variety of visually presentedtasks, Stelmack et al. (1993) found an associ-ation between neuroticism and P3 latencywith some tasks, but not others. At least interms of amplitude, the scalp distributionmay be an important variable in that high andlow neuroticism participants show similari-ties in P3 amplitude at some electrode sitesand not others. Low neuroticism participantshave shown the usual scalp distribution witha clear parietal maximum, whereas high neu-roticism participants show a flat scalp distri-bution with nearly the same level of P3amplitude at all electrode sites with both astructural (deciding if a word was longer orshorter than six letters) and affective (ratingemotional valence of pleasant, neutral, andunpleasant pictures) processing task (Bartusseket al., 1996, experiment 1). Finally, the sam-pling of participants may be important in thatgender and age may modulate the effects ofneuroticism on the P3 (Gurrera et al., 2005;Pritchard, 1989).

Several personality researchers have alsoexamined a component of the P3 that followsa third infrequent non-target stimulus that canbe embedded in an oddball task. This P3a, ornovelty-P3, is an earlier, frontally distributedpotential arising in part from the anterior superior temporal gyrus and anterior cingulategyrus and may be regarded as an index of the orienting response (Soltani and Knight,2000). The amplitude of the P3a reflects theamount of attention a participant invests in theirrelevant, unexpected, and distracter stimuli.


9781412946520-Ch15 29/3/08 12:07 PM Page 319

A negative association between neuroticismand P3a amplitude to novel environmentalstimuli embedded in an auditory oddball taskhas been found at frontal left, frontal right,frontal centre, and parietal centre locations(Gurrera et al., 2001). A similar, though sta-tistically non-significant, association for P3aamplitude to irrelevant auditory distracteritems has also been reported (Fjell et al.,2005). In a comparison between target andnovel stimuli, stronger correlations with neu-roticism were generally found for the P3amplitude during novel stimuli (Fjell et al.,2005). The authors suggested that thestronger association may reflect that individ-uals high in neuroticism are more susceptibleto distraction and less able to inhibitresponses to non-target stimuli, whichaccords with Eysenck’s notion of hypervigi-lance in these individuals.

fMRI

Although emotion research has been one ofthe largest beneficiaries of the new brainimaging techniques (Critchley, 2003; Davidsonet al., 2000; Drevets, 2001), there is a surprisingpaucity of studies investigating the correlationbetween neuroticism and fMRI responsivity.That is, there is a large literature on deviantfMRI responses in psychopathology, but mostof these studies used patients with clinicalanxiety disorders or depression. In these stud-ies, it becomes hard to separate the effects ofhaving a psychiatric disorder itself, includingtoxic brain effects of co-morbid hypercorti-solism (Sapolsky et al., 1986), from the effectsof neuroticism per se. Fortunately, there is anincreasing number of studies that addressfMRI correlates of neuroticism in samples of healthy subjects (Canli et al., 2001;Eisenberger et al., 2005; Etkin et al., 2004;Guyer et al., 2006; Most et al., 2006; Paulus et al., 2003; Schwartz et al., 2003).

Most of these fMRI studies have impliedamygdala hyper-reactivity in subjects scoringhigh on neuroticism. Activity in the amygdalarobustly increases in response to unpleasant

stimuli, most prominently to fearful andangry faces, even when these stimuli are rap-idly masked to prevent conscious awareness.Neuroticism was found to be associated withlarger amygdala responses to unpleasant pic-tures from the IAPS (Canli et al., 2001).Related traits of anxiety and harm avoidancewere similarly associated with a larger amyg-dala fMRI response to angry or fearful faces(Etkin et al., 2004; Most et al., 2006). Finally,enhanced amygdala activity was found inresponse to novel, neutral face stimuli inadults who had been classified as inhibited astoddlers compared with adults who were notclassified as such (Schwartz et al., 2003).

Harm avoidance was also associated withlarger anterior cingulate cortex (ACC) reac-tivity while viewing aversive pictures (Mostet al., 2006). Neuroticism was similarly asso-ciated with a significant increase in the acti-vation of the dorsal ACC in response to anoddball task (Eisenberger et al., 2005).Intriguingly, the accuracy of the detection ofinteroceptive signals measured as heart beatperception was much better accounted for by ACC reactivity to the odd ball stimuli (r2 = 0.74) than by self-reported neuroticism(r2 = 0.16) This led the authors to suggest‘that neural reactivities may provide a moredirect measure of personality than self-reportsdo’ (Eisenberger et al., 2005: 196).

Paulus et al. (2003) had subjects perform arisk-taking decision-making task in whichsubjects could opt to try to win a small gainat low risk or larger gains at higher risk.Harm avoidance (r = 0.54) and neuroticism(r = 0.59) significantly predicted the degreeof (mostly right) anterior insula activationduring a punished response. They speculatethat the insular cortex may be critical for thegeneration of anticipatory aversive somaticmarkers that guide risk-taking behavior(Damasio, 1999) and for aversive outcomeprocessing once a decision has been made.Neuroticism appears to be associated withhyper-reactivity of this structure.

There is also some evidence for an effectof neuroticism on reactivity of the frontostri-atal circuitry engaged in reward processing.


9781412946520-Ch15 29/3/08 12:07 PM Page 320

Forty-four children screened for behavioralinhibition (shyness) at 4 months wereretested with fMRI in adolescence (Guyer et al., 2006). This study used the cued reac-tion time task by Knutson and colleagues(described in more detail below) that hadbeen found to consistently engage the cau-date nucleus, putamen, and nucleus accum-bens (Knutson and Bhanji, 2006; Knutson et al., 2001, 2003). In all adolescents, activa-tion of these structures became larger as theamount of monetary reward to be gained orlost increased, but the effect was muchstronger in the group that had been behaviorally inhibited at 4 months.

EXTRAVERSION

EMG startle reflex

Response size and habituation The hypothesized lower cortical arousal inthe ascending reticular activating system inextraverts suggests that extraversion will beassociated with smaller startle responses and faster, more rapid rates of habituation(Blumenthal, 2001). The startle reflex is par-ticularly relevant to test such a hypothesis asthe nucleus reticularis pontis caudalis hasbeen implicated in regulating cortical arousal(Gottesmann et al., 1995) and forms part of theacoustic startle pathway (Davis et al., 1999).In support of Eysenck’s conceptualization,introverts show larger startle magnitude(Blumenthal et al. 1995; Blumenthal, 2001)than extraverts. Introversion also appears to beassociated with a faster response latency to85 dB(A) than to 60 dB(A) acoustic stimuli,whereas extraverts do not show this difference,suggesting that introverts respond more tohigher intensity stimuli (Britt and Blumenthal,1991). However, differences in responsemagnitude between introverts and extravertshave not always been found (Akdag et al.,2003). No relationship between BAS scoresand startle magnitude was found during theintertrial intervals of an affective startle

modulation experiment (Hawk et al., 2003).There is at least one report of an overallgreater response probability (indicating morereactivity) in extraverts than introverts,although this was in the context of an affec-tive startle modulation experiment (Kumariet al., 1996).

Blumenthal (2001) showed that selectiveattention may influence startle magnitudedifferently in introverts and extraverts.Introverts showed reduced startle amplitudewhen they directed attention towards a visualdisplay (and away from an acoustic startle-eliciting stimulus). In contrast, startle ampli-tude tended to increase when extravertsdirect attention towards a visual display. Thedifference may reflect that introverts werebetter able to allocate attention to the visualtask, and were less distracted by the acousticstartle stimulus as less startle reactivity wouldbe expected when attention is directed awayfrom the modality of the eliciting stimulus(Neumann, 2002). Blumenthal (2001) alsoshowed that the rate of startle habituation wasfaster in extraverts to 90 dB startle pulsesthan in introverts. The faster habituation inextraverts has been replicated in a study thatexamined habituation across individual trialsduring a picture slide presentation and duringthe intervals between the slides (LaRowe et al., 2006).

Prepulse inhibition No association between extraversion and PPIhas been found across two independent sam-ples (Corr et al., 2002). However, a compo-nent measure of Gray’s BAS activity, termedBAS-drive, is negatively correlated with PPIat lead intervals of 30, 60, and 120 ms (Corret al., 2002). The association between BASand PPI is further supported by an experi-ment that assessed PPI to an auditory stimulusduring and in between picture presentations(Hawk and Kowmas, 2003). PPI assessedduring a picture presentation was marginallygreater among high BAS participants com-pared to low BAS participants. Although notstatistically reliable, the same difference wasfound when PPI was assessed during the


9781412946520-Ch15 29/3/08 12:07 PM Page 321

intertrial intervals. However, as PPI wasassessed during slides that were emotionallytoned, it is not clear whether the observedeffects were influenced by the affective orattentional effects of the slides. Similar to theassociation between neuroticism and PPI, theassociation between BAS and PPI may reflectthat high BAS participants attend more stronglyto the prepulse stimuli and that this effectmay have increased PPI. Further researchwhich systematically manipulates attentionto the lead stimulus is required to test thisinterpretation.

Affective modulation Predictions of the relationship between affec-tive startle modulation and levels of extraver-sion are not clear. For instance, extraverts mightshow greater startle attenuation during pleas-ant lead stimuli if extraversion is associatedwith higher levels of positive affect. On theother hand, greater attenuation during pleasantand greater potentiation during unpleasantlead stimuli might be expected in introverts dueto their higher level of arousal in the ascend-ing reticular activating system (Kumari et al.,1996) and the observation that lead stimuli thatare higher in arousal elicit more pronouncedaffect startle modulation effects (Lang et al.,1997). An interaction between the valence of pictures and responses latency has beenreported (Corr et al., 1995). The interactionreflected that only extraverted participantsshowed the expected linear pattern of modulation, although the effect seemed to bestrongest for fear potentiation as only the difference between unpleasant and neutralslides was statistically significant. No associ-ation between extraversion and startle modu-lation during emotionally toned film clips hasalso been reported (Kumari et al., 1996).

Gray’s model would suggest enhanced affec-tive startle modulation among high BAS andhigh BIS participants relative to low BAS andlow BIS participants. The main difference forBIS and BAS is that the enhanced modulationfor high BAS participants should reflect greaterattenuation during pleasant stimuli becauseof the association that pleasant stimuli have

with positive reinforcement. In contrast, theenhanced modulation for high BIS participantsshould reflect greater potentiation duringunpleasant stimuli due to the association thatunpleasant stimuli have with punishment.Consistent with these predictions, robustaffective startle modulation has beenobserved for high BAS participants, but not forlow BAS participants (Hawk and Kowmans,2003). Although no groups showed facilitationduring unpleasant pictures compared to neu-tral pictures, only the high BAS participantsshowed greater startle attenuation duringpleasant pictures relative to neutral pictures.

A particularly novel way to examine therelationship between extraversion and psy-chological processes during emotionallytoned situations is to examine startle modula-tion during a social encounter. In such a situ-ation, it might be hypothesized that introvertswill direct their attention inwards more thanextraverts (Blumenthal et al., 1995). To testthis hypothesis, an experimental assistantentered the participant’s room and sat behindthe participant while pretending to takenotes. High extraversion participants did notdiffer in startle amplitude between the socialencounter condition and a control (no socialencounter) condition (Blumenthal et al., 1995).However, in support of the predictions, lowextraversion participants showed smaller star-tle amplitude in the social encounter conditionthan in the control condition (Blumenthal etal., 1995). The examination of startle modu-lation during a social encounter may providea means to specifically target the modulationof startle along the extraversion/introversiondimension, in much the same way that affec-tive startle modulation with fear-provokingstimuli, rather than disgust-provoking stim-uli, may more specifically target neuroticism(Wilson et al., 2000).

ERP

Early ERP components Early research indicated that introverts have agreater N1-P2 amplitude than extraverts when


9781412946520-Ch15 29/3/08 12:07 PM Page 322

elicited by infrequent tones (Stelmack et al.,1977), possibly reflecting that introvertsattended to the tones more than extraverts. DePascalis (De Pascalis, 1993) did not observeany direct association between N1 amplitudeand extraversion, but did find a positive associ-ation between extraversion and auditory andvisual N1 frontal ratios. The associationreflected that extraverts showed predominantlyleft hemisphere engagement. These resultssupport earlier findings in which P2 amplitudediffered between extraverts and introvertswhen measured from the left hemisphere, butnot when measured from the right hemisphere(De Pascalis and Montirosso, 1988).

Differences in attentional engagement alsoappear to underlie differences between intro-verts and extraverts in the N2. Extraverts havea shorter N2 latency than introverts to taskirrelevant tone pipes that are superimposedover meaningful and meaningless speech (DePascalis and Montirosso, 1988). In addition,the N2 amplitude is larger during meaningfulspeech than during meaningless speech forthe extraverts, whereas the opposite differ-ence is found for introverts (De Pascalis and Montirosso, 1988). The difference mayreflect that the extraverts are more engagedor attended to the stimuli when the speech ismeaningful than when it is meaningless,resulting in the greater N2 amplitude for theformer condition. A more positive N2 ampli-tude in extraverts during a gambling task inwhich tones indicated a win or a loss has alsobeen reported (Bartussek et al., 1993),although this difference was not due to a dis-tinct effect on N2, but a consequence of agenerally more positive ERP amplitude ofthe extraverts at the time between 250 and400 ms after stimulus onset. The N2 ampli-tude of the extraverts in this study wasalways more positive when the tone indicateda win than a loss (Bartussek et al., 1993).Introverts, however, showed more positive N2amplitudes to the tones indicating a loss thana win (Bartussek et al., 1993). The data appearconsistent with the prediction that extravertsshow larger reactivity to stimuli associatedwith reward than to stimuli associated with

punishment due to the activation of the BAS,whereas the opposite pattern will be foundfor introverts due to the activation of the BIS(Gray, 1982; Gray et al., 2000).

P3 The P3 amplitude elicited during the oddballparadigm and other similar tasks is smaller inextraverts than in introverts (Daruna et al.,1985; Polich et al., 1992; Pritchard, 1989;Wilson and Languis, 1990) and is likely toreflect the reduced attentional engagement inextraverts. Extraversion may also be relatedto the habituation of P3 amplitude acrosstrials in that extraverts have displayed agreater decrease in P3 amplitude to the infre-quent target stimuli across trial blocks thanintroverts (Ditraglia and Polich, 1991).Longer P3 latency has also been associatedwith higher levels of extraversion in a cate-gory matching and a same–different judg-ment task (Stelmack et al., 1993), a simpleRT task (Doucet and Stelmack, 2000), and areaction time task that varied stimulus andresponse location (Brebner, 1990). In con-trast to these findings, P3 amplitude has beenfound to be greater in extraverts than in intro-verts (Cahill and Polich, 1992; Gurrera et al.,2001, 2005). Others investigations havefound no relationship between extraversionand P3 (Ditraglia et al., 1991; Plooij-vanGorsel, 1981; Pritchard, 1989). One interpre-tation for the inconsistent pattern of results isthat it reflects features of the experimentalprocedure. For instance, if extraverts habitu-ate more rapidly to repetitive stimuli thanintroverts do (Ditraglia et al., 1991; Polich et al., 1992), different results may emergedepending on the number of trials used in theexperiment (Gurrera et al., 2005). The use ofcommunity samples may also be more likelyto yield a relationship than undergraduatestudent samples due to a greater range ofextraversion scores in such samples (Gurreraet al., 2001).

The P3 elicited following the presentationof emotionally toned words may also vary asa function of extraversion for the same rea-sons that extraversion should be related to


9781412946520-Ch15 29/3/08 12:07 PM Page 323

affective startle modulation. Bartussek et al.(1996, experiment 1) found that a complexrelationship emerged when emotionallytoned and neutral words were presentedduring a structural processing task, requiringa decision about whether a word is longer orshorter than six letters, and an affective pro-cessing task, requiring a rating of the affec-tive valence of emotionally toned words. Inextraverts, P3 amplitudes during the struc-tural processing task showed a maximumamplitude at parietal electrode sites and alarger amplitude for pleasant and unpleasantwords than neutral words. The differencebetween the emotionally toned and neutralwords became less pronounced at a centralelectrode and was not present at a frontalelectrode. In contrast, introverts showedhigher P3 amplitudes during the emotionallytoned words than during the neutral words ata parietal site only during the affective pro-cessing task. The results seem to indicate thatextraverts reacted differentially to the emo-tional arousal associated with the words evenin a task that did not require attention to bedirected towards the emotional content. Forthe introverts, P3 amplitude at a frontal elec-trode was also larger to unpleasant and neu-tral words than to pleasant words, a resultconsistent with Gray’s theory of an increasedsensitivity of the BIS, and thus greater reac-tivity to stimuli associated with punishment.In a similar vein, results from the gamblingtask described above found that extravertsshowed larger P3 amplitudes after a loss inthe preceding trial, compared to winnings(Bartussek et al., 1993). The opposite wastrue for the introverts as P3 amplitude waslarger when having won in the preceding trialthan after a loss (Bartussek et al., 1993).Signals presented after a win may have beenperceived as more negative as participantswere likely to expect that the next trial wouldbe a loss because an equal number of trialsresulted in a win and loss. Likewise, signalspresented after a loss were interpreted as pos-itive because participants expected that thenext trial would result in a win (Bartussek et al., 1993). Based on this interpretation of

the meaning of the signals, the greater P3amplitudes to signals presented after a lossfor extraverts and greater P3 amplitudes tosignals presented after a win in introverts isconsistent with Gray’s theory.

fMRI

The link between extraversion and functionalMRI reactivity has been most systematicallystudied by Canli and colleagues (Canli,2004; Canli et al., 2001, 2002, 2004). In afirst study (Canli et al., 2001) they showedthat the neural representation of personalitytraits may be widely distributed throughoutthe brain. In 15 different brain regions extra-version from the NEO-PI showed a signifi-cant correlation to an increase in the fMRIBOLD signal that was selective to pleasantimages from the IAPS contrasted to unpleas-ant images. Two regions clearly stood out:the amygdala and the anterior cingulatecortex (ACC). Differential activation of theseregions in extraverts has since then beenreplicated in different task settings (Amin et al., 2004; Canli et al., 2002). The largerresponse of the amygdala to pleasant IAPSpictures (Canli et al., 2001) as well as happyfacial expressions (Canli et al., 2002) inextraverts is of particular interest becausethis structure had been primarily associatedwith the processing of negative affect.Indeed, as we saw, neuroticism increases theamygdala response to negatively valencedstimuli. Canli et al. (2002) clearly showedthat the extraversion also influences thisstructure, albeit only during the processing ofpositive affect. Interestingly, the activation ofthe amygdala was left-lateralized; that is,located within the hemisphere that has beenassociated with positive emotions and withapproach-related behavior in the EEG asym-metry literature (Davidson et al., 2000).

Because the ACC has been associated withattention to emotional stimuli (Whalen et al.,1998) the finding of larger ACC reactivity inresponse to pleasant stimuli would be com-patible with the hypothesized attentional bias


9781412946520-Ch15 29/3/08 12:07 PM Page 324

for reward stimuli in extraverted subjects(Derryberry and Reed, 1994). However, somecaution is in order with the interpretation ofACC reactivity as purely reflecting the activa-tion of an attentional system. Critchley andcolleagues have repeatedly shown that theACC is activated by afferent and efferent activ-ity of the autonomic nervous system (Critchley,2003; Critchley et al., 2000, 2004, 2005).The autonomic nervous system is activated byany form of arousal, either related to fear,anger, or excitement. The larger ACC reactiv-ity seen in extraverts may therefore also reflecta greater arousability. That this greater arous-ability is selective to pleasant stimuli wouldfit the hypothesized higher sensitivity toreward in extraverts.

In keeping, extraversion has been linked to activity in the mesolimbic dopaminergicreward processing system running from theventral tegmental area to the lateral hypothala-mus, the ventral striatum (most prominentlythe nucleus accumbens) and parts of themedial prefrontal cortex (Knutson and Bhanji,et al., 2006). To test reactivity of this systemwith fMRI recordings, Knutson et al. (2001,2003, 2006) developed a monetary incentive-processing task in which participants see a cueindicating whether they can gain or avoidlosing money and how much is at stake. Aftera short delay subjects must rapidly respond toa presented target. Feedback on their success(i.e. were they fast enough?) is immediate andthey are presented with the amount lost orgained and their cumulative total. FunctionalMRI recording during this task showed thatpotential losses only activated the thalamusand caudate nucleus, but that the nucleusaccumbens was additionally activated by theanticipation of potential gain. Large individualdifferences in the degree of nucleus accumbensactivation were found that corresponded to thesubjectively experienced cue-elicited happi-ness in this study and to excitement in anotherstudy (Bjork et al., 2004). Activation of thenucleus accumbens was selective to gain anticipation rather than gain outcome, since the increase in BOLD fMRI signal intensity ceased after feedback of success

(Knutson et al., 2001). In contrast, gain out-come, but not gain anticipation, did engagethe medial prefrontal cortex (Knutson et al.,2003). Pooling across their studies extraver-sion was found to correlate with activation inthe nucleus accumbens, medial caudate, andMPFC for gain versus non-gain anticipation,but not to loss versus non-loss anticipation(Knutson et al., 2006). The larger reactivityof these areas in extraversion was corrobo-rated by another study (Cohen et al., 2005),although in this study reward evaluationrather than reward anticipation seemed to bemore important.

The effects of extraversion on brain func-tioning are not limited to affective processingbut are also evident during cognitive process-ing (Kumari et al., 2004; Stenberg et al., 1990).One study examined the influence of extra-version in fMRI activity during an ‘n-back’task involving memory loads (0-, 1-, 2-, and3-back) and a rest condition in healthy men(Kumari et al., 2004). As predicted by Eysenck,maintaining adequate cognitive performancerequired a larger increase in cortical arousalin the extraverts as evidenced by greaterchanges in fMRI signal intensity from rest tothe 3-back condition in the ACC and the dor-solateral prefrontal cortex. In further keepingwith the neurobiological underpinnings ofEysenck’s model, higher extraversion scoreswere negatively associated with resting fMRIsignals in the bilateral thalamus, cuneus andleft hemisphere language areas, suggestinglower baseline arousal in these areas, possi-bly due to extraverts engaging in less internalself-talk than introverts.

DISCUSSION

The above review identified a number of theoretically meaningful and reproducible cor-relations between psychophysiological meas-ures and personality traits measured by someof the major questionnaires used in the field(Carver and White, 1994; Cloninger et al.,1991; Costa and McCrae, 1992; Eysenck and Eysenck, 1994). Neuroticism and related


9781412946520-Ch15 29/3/08 12:07 PM Page 325

traits have been related to larger startle modu-lation during fear eliciting scenes (Caseras et al., 2006; Wilson et al., 2000). These findingsare in accord with Eysenck’s notion of hyper-vigilance to threat/danger in these individuals.Such hypervigilance is also compatible withthe gist of studies comparing fMRI responsesin low and high neurotic subjects. These stud-ies tended to converge on areas involved in theprocessing of stimuli signaling conflict threator disgust; that is, the amygdala, the insula,and the ACC (Canli et al., 2002; Eisenbergeret al., 2005; Paulus et al., 2003). Research onextraversion has shown that introverts are gen-erally more reactive with startle amplitudethan extraverts, with or without instructions toattend (Blumenthal, 2001). Furthermore, trial-by-trial habituation of the startle reflex is morerapid in extraverts. The N2 amplitude ofextraverts was found to be more positive whenthe signal indicates a win than when the signalindicates a loss whereas the opposite pattern isfound for introverts. This is consistent with theprediction that extraverts compared to intro-verts will show larger reactivity to stimuliassociated with reward than to stimuli associ-ated with punishment. In keeping, functionalMRI studies have found extraversion to corre-late with the extent of activation in the nucleusaccumbens, medial caudate, and MPFC ingain versus non-gain anticipation trials(Knutson et al., 2006) and with the extent ofactivation in the nucleus accumbens andMPFC in response to actual reward (Cohen et al., 2005). Pleasant stimuli (e.g. happy faces)only engage the amygdala and the anteriorcingulate cortex (ACC) in extraverts (Amin et al., 2004; Canli et al., 2002).

In summary, despite small sample sizesand the absence of special subject selectionin most studies, significant correlations existbetween psychophysiological reactivity andpaper-and-pencil-based assessment of person-ality. These correlations are generally modestbut reach into the range of large effects sizes(r > 0.70) where up to half of the variance in evoked EMG, EEG, or fMRI activity and self-reported personality derives from ashared underlying factor. Note, however, that

we can also turn that argument around: a largepart of the variance in paper-and-pencil-basedpersonality is not shared with psychophysio-logical reactivity. This raises the question ofwhich of these types of measures best capturesthe theoretical construct of personality. Beforewe can compare their relative merit, psycho-physiological indices must first be subjectedto the same rigorous psychometric demandsas paper-and-pencil tests. Put otherwise, theymust be shown to have good test–retest relia-bility, validly index the brain processes theyclaim to index (construct validity) and to pre-dict behavior across a wide range of situation(predictive validity). This is a huge challenge,both in terms of person-power and finance,because the complexities in the data acquisi-tion and data analysis in psychophysiologicalresearch often reduce the sample size to verysmall numbers, particularly when comparedwith questionnaire based research.

The test–retest reliability for some psycho-physiological measures has been shown to bemoderate to high. For instance, good to excel-lent reliability has been reported for the P3(Smit et al. in press), fMRI activation duringmemory encoding (Aron et al., 2006; Wagneret al., 2005), and the habituation and prepulseinhibition of the startle eye-blink reflex (Abelet al., 1998; Flaten, 2002). However, it has beendisappointingly low for other measures, includ-ing affective startle modulation (Anokhin et al.,2007), EEG asymmetry (Smit et al., 2007), andfMRI activation during a working memorytask (Manoach et al., 2001). This may be dueto the inherent complexity of signal generation,flaws in recording techniques and strategies, or a failure to standardize testing conditions(e.g. stimuli, instructions, participant-experi-menter interaction), all of which can reducereliability.

Based on a large body of experimental stud-ies in animals and neurological patients, theconstruct validity of brain activity as the basisfor behavioral tendencies is very large (Kandelet al., 2000). However, there is no perfect one-to-one mapping of psychophysiologicalindices on brain activity. Although the tempo-ral resolution of EEG recordings is excellent,


9781412946520-Ch15 29/3/08 12:07 PM Page 326

it is hard to estimate the exact sources in thebrain generating the observed patterns ofelectrical activity. This so-called inverseproblem arises because an infinite number ofpossible charge distributions in the headcould lead to the same pattern on the scalp.Although clever modeling techniques havebeen successfully used to tackle this prob-lem, signals that arise from more than a fewdipoles remain hard to localize with preci-sion. Localization of the source of brainactivity is much better with fMRI than withEEG, but the relation between the recordedsignal (blood oxygenation) and underlyingneural activity remains a matter of debate.Although it is commonly assumed that it represents excitatory neural activity, there is disagreement about whether it cannot also represent inhibitory neural activity(Waldvogel et al., 2000). Second, an increasein activation during one condition is equivalentto a decrease during the other condition. Whatis interpreted as an increase in activation topleasant pictures, for instance, could insteadhave represented a decrease in activation tounpleasant pictures. Finally, correlation doesnot equal causation. A brain region that isactivated during a task may not play a criticalrole in the task’s performance. The regionmay merely be ‘listening in’ to the activity inother brain areas that constitute the truesources of individual differences.

To date surprisingly little is known aboutthe predictive validity of psychophysiologi-cal reactivity. There are a number of studiesshowing predictive validity of psychophysio-logical reactivity to the development of psy-chopathology; for example, in the areas ofschizophrenia (Keshavan et al., 2005) andalcoholism (Hill and Shen, 2002). However,this literature is much less well developedthan that for paper-and-pencil measures ofpersonality that are known to significantlypredict anxiety disorders, depression, drugabuse, and antisocial behavior (Cloninger et al., 2006; Masse and Tremblay, 1997). Also,no studies known to us have addressed thepredictive validity of psychophysiology forlifestyle parameters (smoking, exercise, sexual

practices) or behavioral outcomes like careerchoices or job success.

In short, reliability of psychophysiologicalmeasures is currently less convincing thanthose for paper-and-pencil measures and valid-ity has been far more rigorously tested for thelatter. It is of note, however, that many of thestudies reviewed here have silently adoptedthe stance that a single psychophysiologicalmeasure should index personality. This makesfor an unfair comparison to paper-and-penciltests that use multiple weighed items to arriveat a summary score. We believe a large increasein reliability and validity of a psychophysio-logical test battery could be achieved by looking at patterns of psychophysiologicalresponding on multiple measures and to multiple types of stimuli rather than a singlemeasure on a single class of stimuli. Ideally,multiple psychophysiological responses(ECG, SCL ERP, and fMRI) would need to berecorded consecutively, or even more ideallysimultaneously, in the same subject in responseto various classes of stimuli (e.g. emotionalpictures, reward, and punishment). Theresponse of each measure to each particularclass of stimuli can be considered a singlepsychophysiological ‘item’. These items wouldbe subjected to factor analyses just as theitems of personality inventories to obtain afactor structure and ultimately sum scores.Indeed ‘ordinary’ items for a personality inven-tory may be merged with psychophysiologicalitems to obtain a hybrid ‘subjective/objective’personality measure. This has the potential toanchor the new personality measure to thelarge existing theoretical framework basedon paper-and-pencil while adding biologicalfoundation. Therefore, rather than pitch psychophysiological recordings as more scientific alternatives to paper-and-pencilassessment, we propose to use these meas-urements as a way to support theory-buildingon the neurobiological basis of personality.

Implications for neuroscience

Although much work remains, the psychophys-iological studies on personality research


9781412946520-Ch15 29/3/08 12:07 PM Page 327

reviewed in this chapter already convey a veryimportant message to the field of neuroscience.Mainstream neuroscience is still very muchfocused on universal affective and cognitivebrain processes at the expense of individualdifferences (Kosslyn et al., 2002; Plomin andKosslyn, 2001). By not taking individual dif-ferences into account, or considering them amere nuisance variable, many neurosciencestudies may have failed to detect a link betweena brain structure and the putative affective andcognitive processes in which it is involved. Ithad, for instance, been assumed from groupfMRI recordings that the amygdala is notinvolved in the processing of stimuli with apositive emotional value. By bringing person-ality into the equation, Canli et al. (2002) haveproven this wrong. The amygdala does stronglyinfluence such processing, but its total activa-tion is a function of the level of extraversion.In short, by not taking personality differencesinto account, neuroscience runs the risk of pre-senting us with ‘universal processes’ in affectand cognition that reflect the average patternof brain activity across many individuals, butdo not really occur in any single individual.

REFERENCES

Abel, K., Waikar, M., Pedro, B., Hemsley, D. andGeyer, M. (1998) ‘Repeated testing of prepulse inhibition and habituation of thestartle reflex: A study in healthy human controls’, Journal of Psychopharmacology,12: 330–37.

Akdag, S.J., Nestor, P.G., O’Donnell, B.F.,Niznikiewicz, M.A., Shenton, M.E. andMcCarley, R.W. (2003) ‘The startle reflex inschizophrenia: Habituation and personalitycorrelates’, Schizophrenia Research, 64(2-3):165–73.

Allen, J.J., Coan, J.A. and Nazarian, M. (2004)‘Issues and assumptions on the road fromraw signals to metrics of frontal EEG asym-metry in emotion’, Biological Psychology,67(1-2): 183–218.

Amin, Z., Constable, R.T. and Canli, T. (2004)‘Attentional bias for valenced stimuli as afunction of personality in the dot-probe

task’, Journal of Research in Personality,38(1): 15–23.

Anokhin, A.P., Golosheykin, S. and Heath, A.C.(2007) ‘Genetic and environmental influenceson emotion-modulated startle reflex: A twinstudy’, Psychophysiology, 44(1): 106–12.

Aron, A.R., Gluck, M.A. and Poldrack, R.A.(2006) ‘Long-term test–retest reliability offunctional MRI in a classification learningtask’, Neuroimage, 29(3): 1000–6.

Bartussek, D., Becker, G., Diedrich, O.,Naumann, E. and Maier, S. (1996)‘Extraversion, neuroticism, and event-relatedbrain potentials in response to emotionalstimuli’, Personality and IndividualDifferences, 20(3): 301–12.

Bartussek, D., Diedrich, O., Naumann, E. andCollet, W. (1993) ‘Introversion extroversionand event-related potential (Erp) – A test ofGray, J.A. theory’, Personality and IndividualDifferences, 14(4): 565–74.

Bjork, J.M., Knutson, B., Fong, G.W.,Caggiano, D.M., Bennett, S.M. andHommer, D.W. (2004) ‘Incentive-elicitedbrain activation in adolescents: Similaritiesand differences from young adults’, Journalof Neuroscience, 24(8): 1793–802.

Blumenthal, T.D. (2001) ‘Extraversion, attention,and startle response reactivity’, Personalityand Individual Differences, 31(4): 495–503.

Blumenthal, T.D., Chapman, J.G. and Muse,K.B. (1995) ‘Effects of social anxiety, atten-tion, and extraversion on the acoustic startleeyeblink response’, Personality and IndividualDifferences, 19(6): 797–807.

Blumenthal, T.D., Cuthbert, B.N., Filion, D.L.,Hackley, S., Lipp, O.V. and van Boxtel, A.(2005) ‘Committee report: Guidelines forhuman startle eyeblink electromyographicstudies’, Psychophysiology, 42(1): 1–15.

Brebner, J. (1990) ‘Psychological and neurophys-iological factors in stimulus-response compat-ibility’, in R.W. Proctor and T.G. Reeves (eds),Stimulus–response Compatibility. Amsterdam:Elsevier. pp.241–60.

Britt, T.W. and Blumenthal, T.D. (1991)‘Motoneuronal insensitivity in extraverts asrevealed by the startle response paradigm’,Personality and Individual Differences, 12(5):387–93.

Cahill and Polich, J. (1992) ‘P300, probability,and introverted/extraverted personalitytypes’, Biological Psychology, 33(1): 23–35.


9781412946520-Ch15 29/3/08 12:07 PM Page 328

Canli, T. (2004) ‘Functional brain mapping ofextraversion and neuroticism: Learning fromindividual differences in emotion processing’,Journal of Personality, 72(6): 1105–32.

Canli, T., Amin, Z., Haas, B., Omura, K. andConstable, R.T. (2004) ‘A double dissociationbetween mood states and personality traitsin the anterior cingulate’, BehavioralNeuroscience, 118(5): 897–904.

Canli, T., Sivers, H., Whitfield, S.L., Gotlib, I.H.and Gabrieli, J.D. (2002) ‘Amygdala responseto happy faces as a function of extraversion’,Science, 296(5576): 2191.

Canli, T., Zhao, Z., Desmond, J.E., Kang, E.,Gross, J. and Gabrieli, J.D. (2001) ‘An fMRIstudy of personality influences on brain reac-tivity to emotional stimuli’, BehavioralNeuroscience, 115(1): 33–42.

Carver, C.S. and White, T.L. (1994) ‘Behavioral-inhibition, behavioral activation, and affec-tive responses to impending reward andpunishment – the Bis Bas Scales’, Journal ofPersonality and Social Psychology, 67(2):319–33.

Caseras, F.X., Fullana, M.A., Riba, J., Barbanoj,M.J., Aluja, A. and Torrubia, R. (2006)‘Influence of individual differences in theBehavioural Inhibition System and stimuluscontent (fear versus blood-disgust) on affec-tive startle reflex modulation’, BiologicalPsychology, 72(3): 251–6.

Cloninger, C.R., Przybeck, T.R. and Svrakic,D.M. (1991) ‘The Tridimensional PersonalityQuestionnaire: US normative data’,Psychological Reports, 69(3pt1): 1047–57.

Cloninger, C.R., Svrakic, D.M. and Przybeck,T.R. (2006) ‘Can personality assessment pre-dict future depression? A twelve-monthfollow-up of 631 subjects’, Journal ofAffective Disorders, 92(1): 35–44.

Coan, J.A. and Allen, J.J. (2004) ‘Frontal EEGasymmetry as a moderator and mediator ofemotion’, Biological Psychology, 67(1–2):7–49.

Coan, J.A., Allen, J.J. and McKnight, P.E. (2006)‘A capability model of individual differencesin frontal EEG asymmetry’, BiologicalPsychology, 72(2): 198–207.

Cohen, M.X., Young, J., Baek, J.M., Kessler, C.and Ranganath, C. (2005) ‘Individual differ-ences in extraversion and dopamine geneticspredict neural reward responses’, CognitiveBrain Research, 25(3): 851–61.

Corr, P.J., Tynan, A. and Kumari, V. (2002)‘Personality correlates of prepulse inhibitionof the startle reflex at three lead intervals’,Journal of Psychophysiology, 16(2): 82–91.

Corr, P.J., Wilson, G.D., Fotiadou, M., Kumari,V., Gray, N.S., Checkley, S.A. et al. (1995)‘Personality and affective modulation of thestartle reflex. Personality and IndividualDifferences, 19(4): 543–53.

Costa, P.T. and McCrae (1992) ProfessionalManual of the revIsed NEO PersonalityInventory (NEO-PI-R) and NEO Five FactorInventory (NEO-FFI). Odessa, FL: PsychologicalAssessment Resources.

Critchley, H. (2003) ‘Emotion and its disorders’,British Medical Bulletin, 65: 35–47.

Critchley, H.D., Elliott, R., Mathias, C.J. andDolan, R.J. (2000) ‘Neural activity relating togeneration and representation of galvanicskin conductance responses: A functionalmagnetic resonance imaging study’, Journalof Neuroscience, 20(8): 3033–40.

Critchley, H.D., Tang, J., Glaser, D., Butterworth,B. and Dolan, R.J. (2005) ‘Anterior cingulateactivity during error and autonomicresponse’, Neuroimage, 27(4): 885–95.

Critchley, H.D., Wiens, S., Rotshtein, P.,Ohman, A. and Dolan, R.J. (2004) ‘Neuralsystems supporting interoceptive aware-ness’, Nature Neuroscience, 7(2): 189–95.

Damasio, A.R. (1999) The Feeling of WhatHappens: Body and Emotion in the Makingof Consciousness. New York: Harcourt Brace.

Daruna, J.H., Karrer, R. and Rosen, A.J. (1985)‘Introversion, attention, and the late positivecomponent of event-related potentials’,Biological Psychology, 20(4): 249–59.

Davidson, R.J. (1998) ‘Affective style and affectivedisorders: Perspectives from affective neuro-science’, Cognition and Emotion, 12(3):307–30.

Davidson, R.J. (2004) ‘What does the prefrontalcortex “do” in affect: Perspectives on frontalEEG asymmetry research’, BiologicalPsychology, 67(1–2): 219–33.

Davidson, R.J., Jackson, D.C. and Kalin, N.H.(2000) ‘Emotion, plasticity, context, and reg-ulation: Perspectives from affective neuro-science’, Psychological Bulletin, 126(6):890–909.

Davidson, R.J., Schaffer, C.E. and Saron, C.(1985) ‘Effects of lateralized presentations offaces on self-reports of emotion and EEG


9781412946520-Ch15 29/3/08 12:07 PM Page 329

asymmetry in depressed and non-depressed subjects’, Psychophysiology,22(3): 353–64.

Davis, M., Walker, D.L. and Lee, Y. (1999)‘Neurophysiology and neuropharmacologyof startle and its affective modulation’, in M.E.Dawson, A.M. Schell and A.H. Bohmelt (eds),Startle Modification: Implications for Neuroscience, Cognitive Science, and ClinicalScience. Cambridge, Cambridge UniversityPress. pp. 54–9.

De Pascalis V., and Montirosso R., (1988)‘Extraversion (neuroticism and individual dif-ferences in event-related potentials’,Personality and Individual Differences. 9:353–60.

De Pascalis, V. (1993) ‘Hemispheric asymmetry,personality and temperament’, Personalityand Individual Differences, 14(6): 825–34.

De Pascalis, V., Fiore, A.D. and Sparita, A. (1996)‘Personality, event-related potential (ERP)and heart rate (HR): An investigation of Gray’stheory’, Personality and Individual Differences,20(6): 733–46.

Dehaene, S., Posner, M.I. and Tucker, D.M. (1994)‘Localization of a neural system for error detec-tion and compensation’, Psychological Science,5(5): 303–5.

Derryberry, D. and Reed, M.A. (1994)‘Temperament and the self-organization of personality’, Development and Psycho-pathology, 6(4): 653–76.

Ditraglia, G.M. and Polich, J. (1991) ‘P300 andintroverted/extraverted personality types’,Psychophysiology, 28(2): 177–84.

Doucet and Stelmack, R.M. (2000) ‘An event-related potential analysis of extraversion andindividual differences in cognitive processingspeed and response execution’, Journal ofPersonality and Social Psychology, 78(5):956–64.

Drevets, W.C. (2001) ‘Neuroimaging and neu-ropathological studies of depression:Implications for the cognitive-emotional fea-tures of mood disorders’, Current Opinion inNeurobiology, 11(2): 240–49.

Ebmeier, K.P., Deary, I.J., Ocarroll, R.E., Prentice,N., Moffoot, A.P.R. and Goodwin, G.M.(1994) ‘Personality associations with theuptake of the cerebral blood-flow marker(99m)Tc-exametazime estimated with single-photon emission tomography’, Personalityand Individual Differences, 17(5): 587–95.

Eisenberger, N.I., Lieberman, M.D. andSatpute, A.B. (2005) ‘Personality from a con-trolled processing perspective: An fMRI studyof neuroticism, extraversion, and self-con-sciousness’, Cognitive and AffectiveBehavioral Neuroscience, 5(2): 169–81.

Ekman, P. and Friesen, W. (1978) The FacialAction Coding System. Palo Alto, CA:Consulting Psychologists Press.

Etkin, A., Klemenhagen, K.C., Dudman, J.T.,Rogan, M.T., Hen, R., Kandel, E.R. et al.(2004) ‘Individual differences in trait anxietypredict the response of the basolateralamygdala to unconsciously processed fearfulfaces’, Neuron, 44(6): 1043–55.

Eysenck, H.J. (1967) The Biological Basis ofPersonality. Springfield, IL: Thomas.

Eysenck, H.J. (1990) ‘Biological dimensions ofpersonality’, in L.A.Pervin (ed.), Handbook ofPersonality: Theory and Research. New York:Guilford Press. pp. 100–2.

Eysenck, H.J. and Eysenck, M.W. (1985)Personality and Individual Differences. NewYork: Plenum Press.

Eysenck, H.J. and Eysenck, S.B.G. (1994) EysenckPersonality Questionnaire – Revised. San Diego:Educational and Industrial Testing Service.

Fabiani, M., Gratton, G. and Coles, M.G.(2000) in J.T. Cacioppo, L.G. Tassinary and G.Berntson (eds), Handbook of Physiology (2ndedition). Cambridge: Cambridge UniversityPress. pp. 53–84.

Filion D.L., Dawson M.E., Schell A.M., et al.(1994). ‘Probing the Orienting Response withStartle Modification and Secondary Reaction-Time’, Psychophysiology. 31(1): 68–78.

Fjell, A.M., Walhovd, K.B., Meling, S. andJohansen, M.B. (2005) ‘Basic informationprocessing of neurotics and stable: An exper-imental ERP approach to personality and dis-tractibility’, Scandinavian Journal ofPsychology, 46(6): 493–502.

Flaten, M.A. (2002) ‘Test–retest reliability of thesomatosensory blink reflex and its inhibition’,International Journal of Psychophysiology,45(3): 261–5.

Fox, N.A. (1991) ‘If it’s not left, it’s right.Electroencephalograph asymmetry and thedevelopment of emotion’, AmericanPsychologist, 46(8): 863–72.

Gottesmann, C., Gandolfo, G. and Zernicki, B.(1995) ‘Sleep-waking cycle in chronic ratpreparations with brain stem transected at


9781412946520-Ch15 29/3/08 12:07 PM Page 330

the caudopontine level’, Brain ResearchBulletin, 36(6): 573–80.

Graham, F.K. (1979) ‘Distinguishing among orienting, defence, and startle reflexes’, in H.D. Kimmel, E.H. van Olst and J.F.Orlebeke (eds), The Orienting Reflex inHumans. Hillsdale, NJ: Erlbaum. pp. 137–67.

Gray, J.A. (1982) The Neuropsychology ofAnxiety: An Enquiry into the Functions of theSepto-hippocampal System. Oxford: OxfordUniversity Press.

Gray, J.A. and McNaughton, N. (2000) TheNeuropsychology of Anxiety: An Enquiry intothe Functions of the Septo-hippocampalSystem (2nd edition). Oxford: OxfordUniversity Press.

Gurrera, R.J., O’Donnell, B.F., Nestor, P.G.,Gainski, J. and McCarley, R.W. (2001) ‘TheP3 auditory event-related brain potentialindexes major personality traits’, BiologicalPsychiatry, 49(11): 922–9.

Gurrera, R.J., Salisbury, D.F., O’Donnell, B.F.,Nestor, P.G. and McCarley, R.W. (2005)‘Auditory P3 indexes personality traits and cog-nitive function in healthy men and women’,Psychiatry Research, 133(2–3): 215–28.

Guyer, A.E., Nelson, E.E., Perez-Edgar, K.,Hardin, M.G., Roberson-Nay, R., Monk, C.S.et al. (2006) ‘Striatal functional alteration inadolescents characterized by early childhoodbehavioral inhibition’, Journal of Neuro-science, 26(24): 6399–405.

Hagemann, D. (2004) ‘Individual differences inanterior EEG asymmetry: Methodologicalproblems and solutions’, BiologicalPsychology, 67(1–2): 157–82.

Hawk, L.W. Jr. and Kowmas, A.D. (2003)‘Affective modulation and prepulse inhibi-tion of startle among undergraduates highand low in behavioral inhibition andapproach’, Psychophysiology, 40(1): 131–8.

Hewig, J., Hagemann, D., Seifert, J., Naumann,E. and Bartussek, D. (2006) ‘The relation ofcortical activity and BIS/BAS on the traitlevel’, Biological Psychology, 71(1): 42–53.

Hill, S.Y. and Shen, S. (2002) ‘Neurodevelop-mental patterns of visual P3b in associationwith familial risk for alcohol dependence andchildhood diagnosis’, Biological Psychiatry,51(8): 621–31.

Holroyd, C.B. and Coles, M.G. (2002) ‘Theneural basis of human error processing:Reinforcement learning, dopamine, and the

error–related negativity’, PsychologicalReview, 109(4): 679–709.

Kandel, E.R., Schwartz, H.S. and Jessel, T.M.(2000) Principles of Neural Science (4th edi-tion). New York: McGraw-Hill.

Kaviani, H., Gray, J.A., Checkley, S.A., Kumari,V. and Wilson, G.D. (1999) ‘Modulation ofthe acoustic startle reflex by emotionally-toned film clips’, International Journal ofPsychophysiology, 32(1): 47–54.

Keshavan, M.S., Diwadkar, V.A., Montrose,D.M., Rajarethinam, R. and Sweeney, J.A.(2005) ‘Premorbid indicators and risk forschizophrenia: A selective review and update’,Schizophrenia Research, 79(1): 45–57.

Knutson, B. and Bhanji, J. (2006) ‘Neural sub-strates for emotional traits? The case ofextraversion’, in T.Canli (ed.), Biology ofPersonality and Individual Differences. NewYork: The Guilford Press. pp. 116–32.

Knutson, B., Fong, G.W., Adams, C.M., Varner,J.L. and Hommer, D. (2001) ‘Dissociation ofreward anticipation and outcome with event-related fMRI’, Neuroreport, 12(17): 3683–7.

Knutson, B., Fong, G.W., Bennett, S.M.,Adams, C.M. and Hommer, D. (2003) ‘A region of mesial prefrontal cortex tracks monetarily rewarding outcomes:Characterization with rapid event-relatedfMRI’, Neuroimage, 18(2): 263–72.

Kosslyn, S.M., Cacioppo, J.T., Davidson, R.J.,Hugdahl, K., Lovallo, W.R., Spiegel, D.et al.(2002) ‘Bridging psychology and biology. Theanalysis of individuals in groups’, AmericanPsychologist, 57(5): 341–51.

Kumari, V., Corr, P.J., Wilson, G.D., Kaviani, H.,Thornton, J.C., Checkley, S.A. et al. (1996)‘Personality and modulation of the startle reflexby emotionally-toned filmclips’, Personality andIndividual Differences, 21(6): 1029–41.

Kumari, V., Ffytche, D.H., Williams, S.C. andGray, J.A. (2004) ‘Personality predicts brainresponses to cognitive demands’, Journal ofNeuroscience, 24(47): 10636–41.

Lang, P.J., Bradley, M.M. and Cuthbert, B.N.(1997) ‘Motivated attention: Affect, activa-tion, and action’, in P.J. Lang, R.F. Simons andM.T. Balaban (eds), Attention and Orienting.Cliffside, NJ: Lawrence Erlbaum. pp. 97–136.

Lang, P.J., Bradley, M.M. and Cuthbert, B.N.(1998) ‘Emotion, motivation, and anxiety:brain mechanisms and psychophysiology’,Biological Psychiatry, 44(12): 1248–63.


9781412946520-Ch15 29/3/08 12:07 PM Page 331

Lang, P.J., Bradley, M.M. and Cuthbert, B.N.(2001) International Affective PictureSystem. Gainesville: Center for Reserach inPsychophysiology, University of Florida.

LaRowe, S.D., Patrick, C.J., Curtin, J.J. andKline, J.P. (2006) ‘Personality correlates ofstartle habituation’, Biological Psychology,72(3): 257–64.

Luu, P., Collins, P. and Tucker, D.M. (2000) ‘Mood,personality, and self-monitoring: Negativeaffect and emotionality in relation to frontallobe mechanism of error monitoring’, Journalof Experimental Psychology: General: 129(1):43–60.

Manoach, D.S., Halpern, E.F., Kramer, T.S.,Chang, Y., Goff, D.C., Rauch, S.L. et al. (2001) ‘Test–retest reliability of a functionalMRI working memory paradigm in normal and schizophrenic subjects’,American Journal of Psychiatry, 158(6):955–8.

Masse, L.C. and Tremblay, R.E. (1997) ‘Behaviorof boys in kindergarten and the onset ofsubstance use during adolescence’, Archivesof General Psychiatry, 54(1): 62–8.

Mischel, W. and Shoda, Y. (1995) ‘A cognitive-affective system-theory of personality –reconceptualizing situations, dispositions,dynamics, and invariance in personality structure’, Psychological Review, 102(2):246–68.

Most, S.B., Chun, M.M., Johnson, M.R. andKiehl, K.A. (2006) ‘Attentional modulationof the amygdala varies with personality’,Neuroimage, 31(2): 934–44.

Neumann, D.L. (2002) ‘Effect of varying levelsof mental workload on startle eyeblink mod-ulation’, Ergonomics, 45(8): 583–602.

Neumann, D.L., Lipp, O.V. and Mchugh, M.J.(2004) ‘The effect of stimulus modality andtask difficulty on attentional modulation ofblink startle’, Psychophysiology, 41(3):407–16.

Ochsner, K.N., Bunge, S.A., Gross, J.J. andGabrieli, J.D. (2002) ‘Rethinking feelings: AnFMRI study of the cognitive regulation ofemotion’, Journal of Cognitive Neuroscience,14(8): 1215–29.

Paulus, M.P., Rogalsky, C., Simmons, A., Feinstein,J.S. and Stein, M.B. (2003) ‘Increased activationin the right insula during risk-taking decisionmaking is related to harm avoidance and neu-roticism’, Neuroimage, 19(4): 1439–48.

Plomin, R. and Kosslyn, S.M. (2001) ‘Genes,brain and cognition’, Nature Neuroscience,4(12): 1153–5.

Plooij-van Gorsel, E. (1981) ‘EEG and cardiaccorrelates of neuroticism: A psychophysio-logical comparison of neurotics and normalcontrols in relation to personality’, BiologicalPsychology, 13: 141–56.

Polich, J. and Martin, S. (1992) ‘P300, cognitivecapability and personality: A correlationalstudy of university undergraduates’, Personalityand Individual Differences, 13(5): 533–43.

Pritchard, W.S. (1989) ‘P300 and EPQ/STPI per-sonality traits’, Personality and IndividualDifferences, 10(1): 15–24.

Santesso, D.L., Segalowitz, S.J. and Schmidt, L.A.(2005) ‘ERP correlates of error monitoring in10-year olds are related to socialization’,Biological Psychology, 70(2): 79–87.

Sapolsky, R.M., Krey, L.C. and McEwen, B.S.(1986) ‘The neuroendocrinology of stress andaging: The glucocorticoid cascade hypothe-sis’, Endocrinology Reviews, 7(3): 284–301.

Schwartz, C.E., Wright, C.I., Shin, L.M., Kagan,J., Whalen, P.J., McMullin, K.G. et al. (2003)‘Differential amygdalar response to novelversus newly familiar neutral faces: a func-tional MRI probe developed for studying inhib-ited temperament’, Biological Psychiatry,53(10): 854–62.

Smit, D.J., Posthuma, D., Boomsma, D.I. andDe Geus, E.J. (2007) ‘Genetic contribution tothe P3 in young and middle-aged adults’,Twin Research and Human Genetics 10(2):335–47.

Smit, D.J., Posthuma, D., Boomsma, D.I. and DeGeus, E.J. (2007) ‘The relation between frontalEEG asymmetry and the risk for anxiety anddepression’, Biological Psychology, 74(1):26–33.

Soltani, M. and Knight, R.T. (2000) ‘Neural originsof the P300’, Critical Reviews in Neurobiology,14(3–4): 199–224.

Stelmack, R.M., Achorn, E. and Michaud, A.(1977) ‘Extraversion and individual differ-ences in auditory and evoked responses’,Psychophysiology, 14(4): 368–74.

Stelmack, R.M. and Houlihan, M. (1995)‘Event-related potentials, personality andintelligence: Concepts, issues and evidence’,in D.H. Saklofske and M. Zeidner (eds),International handbook of Personality andintelligence, New York: Plenum Press.


9781412946520-Ch15 29/3/08 12:07 PM Page 332

Stelmack, R.M., Houlihan, M. and McGarry-Roberts, P.A. (1993) ‘Personality, reactiontime, and event-related potentials’, Journalof Personality and Social Psychology, 65(2):399–409.

Stenberg, G., Risberg, J., Warkentin, S. andRosen, I. (1990) ‘Regional patterns of corticalblood-flow distinguish extroverts from intro-verts’, Personality and Individual Differences,11(7): 663–73.

Stenberg, G., Wendt, P.E. and Risberg, J. (1993)‘Regional cerebral blood-flow and extrover-sion’, Personality and Individual Differences,15(5): 547–54.

Sutton, S.K. and Davidson, R.J. (1997) ‘Prefrontalbrain asymmetry: A biological substrate of thebehavioral approach and inhibition systems’,Psychological Science, 8(3): 204–10.

Swerdlow, N.R., Filion, D., Geyer, M.A. and Braff,D.L. (1995) ‘“Normal” personality correlatesof sensorimotor, cognitive, and visuospatialgating’, Biological Psychiatry, 37(5): 286–99.

van Veen, V. and Carter, C.S. (2002) ‘Thetiming of action-monitoring processes in theanterior cingulate cortex’, Journal ofCognitive Neuroscience, 14(4): 593–602.

Wagner, K., Frings, L., Quiske, A., Unterrainer,J., Schwarzwald, R., Spreer, J. et al. (2005)‘The reliability of fMRI activations in themedial temporal lobes in a verbal episodicmemory task’, Neuroimage, 28(1): 122–31.

Waldvogel, D., van Gelderen, P., Muellbacher,W., Ziemann, U., Immisch, I. and Hallett, M.

(2000) ‘The relative metabolic demand of inhi-bition and excitation’, Nature, 406(6799):995–8.

Wallace, J. (1966) ‘An abilities conception ofpersonality – some implications for personal-ity measurement’, American Psychologist,21(2): 132–8.

Watson, D., Clark, L.A. and Tellegen, A. (1988)‘Development and validation of brief meas-ures of positive and negative affect – thePanas scales’, Journal of Personality andSocial Psychology, 54(6): 1063–70.

Whalen, P.J., Bush, G., McNally, R.J., Wilhelm,S., McInerney, S.C., Jenike, M.A. et al. (1998)‘The emotional counting Stroop paradigm: Afunctional magnetic resonance imagingprobe of the anterior cingulate affective divi-sion’, Biological Psychiatry, 44(12): 1219–28.

Wilson, G.D., Kumari, V., Gray, J.A. and Corr,P.J. (2000) ‘The role of neuroticism in startlereactions to fearful and disgusting stimuli’,Personality and Individual Differences, 29(6):1077–82.

Wilson, G.D. and Languis, M.L. (1990) ‘A topo-graphic study of differences in the P300between introverts and extraverts’, BrainTopography, 2(4): 269–74.

Youn, T., Lyoo, I.K., Kim, J.K., Park, H.J., Ha,K.S., Lee, D.S. et al. (2002) ‘Relationshipbetween personality trait and regional cere-bral glucose metabolism assessed withpositron emission tomography’, BiologicalPsychology, 60(2–3): 109–20.


9781412946520-Ch15 29/3/08 12:07 PM Page 333

psychophysiological measurement of personality measurement of personality eco de geus and david l....

Documents