The Psychopathic Stare:
The Relation Between Fearless Dominance Personality Traits, Pupil Size and
Pupil Dilations
F.S. van Huis
Document: Master’s thesis
Student number: 6049079
UvA Supervisor: Bruno Verschuere
VUmc Supervisor: Koen van Lith
Abstract
Psychopathy has repeatedly been linked to hypo-activity of the stress- or emotional system and fear related brain areas, for example during fear conditioning tasks. Activity of the stress system is often measured by autonomic responses, such as heart rate, skin conductance and the startle reflex. One - easily and unobtrusively measured - correlate of autonomic activity has long been overlooked in psychopathy research: pupil size. The pupil dilates when the system is emotionally aroused. Psychopathy, on the other hand, is thought to be characterized by hypo-arousal of the stress- or emotional system. From a theoretical perspective, it was hypothesized that more Fearless Dominance psychopathic personality traits would be related to smaller baseline pupils. Secondly, it was recently suggested that pupillary reactions can serve as a measure of fear conditioning. The second hypothesis therefore stated that more Fearless Dominance traits would be related to diminished pupil dilations in a fear conditioning task. To test these hypotheses, a group high in Fearless Dominance traits (n=14) was compared to a control group (n=14) in their baseline pupil size and pupil dilations during a classical fear conditioning task. Results suggest that individuals who possess more Fearless Dominance traits display smaller baseline pupil sizes than controls, and show no fear conditioning as measured by pupil dilations. Implications and limitations are discussed.
The Psychopathic Stare
Psychopathy is a developmental disorder with a prevalence of approximately 1% in the general population, and between 15% and 25% in incarcerated populations (Dolan, 2004). It is defined as a severe personality disorder, marked by interpersonal/affective traits - such as emotional detachment, callousness, grandiosity, glibness, and lack of empathy - and antisocial deviance features, including poor behavioral controls, impulsivity, and irresponsibility (Cleckley, 1941, 1976; Hare, 1996). However, the Diagnostic and Statistical Manual of Mental Disorders (DSM-V, American Psychiatric Association) does not mention psychopathy as a distinct disorder from Antisocial Personality Disorder (ASP). While the latest edition of the DSM includes more interpersonal and affective symptoms in their diagnostic criteria for ASP, until recently these criteria consisted mainly of behavioral symptoms, such as repeated lawbreaking, impulsivity, aggressiveness and irresponsibility (APA, 2012). There is continuing debate on whether psychopathy should be seen as analogous to, or as distinct from ASP (Hare, Hart & Harpur, 1991; Blair, et al., 2006; Hare & Neumann, 2010). Those who see psychopathy as a distinct personality disorder, usually agree on the notion that the interpersonal/affective impairments distinguish psychopathy from ASP (Hare, 1996; Cooke & Michie, 2001; Cox et al., 2013). This is reflected in the most commonly used instruments to assess psychopathy, such as Hare’s Psychopathy Checklist (PCL-R, Hare, 2003) and the Psychopathic Personality Inventory (PPI-R, Lilienfeld & Andrews, 1996). For example, the classic PCL-R, a semi-structured interview, follows a distinction between an “Interpersonal Affective” and a “Lifestyle Antisocial” factor. Similarly, the items on the PPI-R, a self-report measure, load on a “Fearless Dominance” (FD, emotional), and an “Impulsive Antisocial” (IA, behavioral) factor. The third scale, Coldheartedness, does not load much on either factor (Rilling et al., 2007).
Examples of emotional impairments seen in psychopathy are reduced emotional empathy, remorselessness, callousness, stress immunity and fearlessness; examples of behavioral symptoms are thrill seeking behavior, irresponsibility and impulsivity (Blair, 2004, 2013; Fowles & Dindo, 2006). Emotional or interpersonal psychopathic traits, such as “Interpersonal Affective” or “Fearless Dominance” traits, may characterize a specific pathway leading to antisocial behavior - consisting of reactive and instrumental aggression, associated with lower Hypothalamus-Pituitary-Adrenal activity and low levels of cortisol (Lopez-Duran et al., 2009). This is what you could call ‘pure’ or ‘primary’ psychopathy. “Impulsive Antisocial” or “Lifestyle Antisocial” traits in the absence of affective symptoms may lead to a more common form of antisocial behavior labeled ‘secondary’ psychopathy or sociopathy, consisting mostly of reactive aggression and over-reactive temperament (Patrick & Brislin, 2007).
Psychophysiology of Psychopathy
Autonomic nervous system activity has been extensively studied in relation to psychopathy, and deviations in autonomic responses have led to several theories about the mechanisms behind psychopathy (Hare, 1978; Portnoy et al., 2014). For example, low arousal or sensation
seeking theories state that the psychopathic individual lacks arousal, and is therefore in a
“chronic state of stimulus hunger” (Hare, 1970). This explains sensation seeking behaviors: psychopathic individuals need more sensory inputs to feel “normal”, and are thus more likely to seek excitement in risky and criminal behaviors (Portnoy et al., 2014). In support of this theory, a low resting heart rate is considered the most reliable biological correlate of psychopathy and of antisocial behavior in children and adolescents (Lorber, 2004; Ortiz & Raine, 2004). Also, psychopathic individuals display lower levels of spontaneous electrodermal activity and lower HPA-axis activity than non-psychopaths (Hansen et al.,
2007; Popma et al., 2006), reflecting chronic cortical and autonomic hypo-arousal (Blackburn, 1979).
On the other hand, the fearlessness theory states that psychopathy is characterized not by low arousal in general, but by a lack of fear in reaction to stressful stimuli (Lykken, 1995). The fearlessness theory explains the low resting heart rate-ASP/psychopathy link by proposing that the situation in which resting heart rate is tested is in itself mildly stressful (Portnoy, et al., 2014). Indeed, heart rate measurements are often accompanied by stressful or unpleasant procedures, such as stressor tasks or aversive stimuli. Consistent with the fearlessness theory, ‘pure’ or ‘primary’ psychopathy (as opposed to “acquired sociopathy”) is associated with lower activation in fear related brain areas (Blair, 2003; Cohn et al., 2013). Diminished (autonomic) fear reactivity is one of the most reliable and valid psychophysiological findings in psychopathy research (Lykken, 1957; Birbaumer et al., 2005). Recent research suggests this impairment may be linked particularly or even solely to Fearless Dominance or Interpersonal Affective traits, irrespective of antisocial traits (Dvorak-Bertsch, Curtin, Rubinstein & Newman, 2009; Lopez, 2013; López, Poy, Patrick & Molto, 2013; Veit et al., 2013) and poorer passive avoidance learning than controls (Newman & Kosson, 1986). Also, psychopathic individuals display a weaker electro-dermal response and a diminished startle reflex in anticipation of or following an aversive stimulus (Hare, 1978; Patrick, Bradley & Lang, 1993).
The Pupil
While many studies have focused on the autonomic system in relation to psychopathy, one easily and unobtrusively measured correlate of autonomic activity has been neglected: pupil
size. Pupil dilation and contraction are influenced by the sympathetic and parasympathetic
the dilator muscle, which increases with sympathetic activity. Pupil contraction is mediated by the sphincter muscle, which is in turn influenced by parasympathetic activity (Beatty & Lucero-Wagoner, 2000). The parasympathetic nervous system regulates “rest and digest” type processes, while the sympathetic nervous system controls “fight or flight” type responses (Kalat, 2009). Thus it makes sense that the pupil dilates when the system is emotionally aroused (Partala & Surakka, 2003). During affective picture viewing, pupillary reactions reflect emotional arousal associated with sympathetic activity, thus correlating with heart rate and skin conductance (Bradley, Miccoli, Escrig & Lang, 2008). Also, pupil dilation is mediated almost exclusively by the stimulating “fight flight” neuromodulator norepinephrine (Holmes, 1967; Jepma & Nieuwenhuis, 2011). This strengthens the expectation that psychopathic individuals might display smaller baseline pupil sizes, reflecting chronic hypo arousal and lower levels of norepinephrine. Results differ regarding gender differences in pupil size, with some studies suggesting that women display larger pupils than men (Jones, 1990; Partala & Surakka, 2003), while others state there is no difference (Winn, Whitaker, Elliott & Phillips, 1994). Besides correlating with general ANS activity, the pupil also dilates with fear, much like heart rate and skin conductance increase with fear. Furthermore, a recent study states that the pupil can serve as a measure of fear conditioning (Visser, Scholte, Beemsterboer & Kindt; 2013).
The question emerges if, like other autonomic responses in psychopathy, the pupil too shows signs chronic hypo-arousal. The first research question thus states: do individuals with more psychopathic traits display smaller baseline pupils? Since Fearless Dominance traits (FD) are most strongly correlated with emotional impairments and fear conditioning deficits, this study will focus on the relationship between FD-traits and the pupil. It is hypothesized that individuals high in FD-personality traits will display smaller baseline pupil sizes than individuals low in FD-personality traits. The second research question regards pupillary
reactions: can pupil dilation serve as a measure of fear conditioning, and if so, is it impaired
in psychopathic individuals? The second hypothesis states that individuals higher in FD-personality traits will display less fear conditioning, as reflected in diminished pupillary responses during a fear conditioning task.
This study is the first to investigate pupillary behavior in healthy individuals possessing psychopathic traits. If highly psychopathic individuals indeed display smaller baseline pupils, this would support low arousal theories, and create new possibilities in the study of psychophysiological mechanisms underlying psychopathy. If pupil dilations can serve as a measure of fear conditioning, this supports the idea of deficient fear conditioning in psychopathy and expands existing techniques in fear conditioning studies.
Methods
Participants
Dutch students from different faculties of the Vrije Universiteit in Amsterdam (n=580) filled in the Fearless-Dominance factor of the Dutch PPI-R (Psychopathic Personality Inventory-Revised, Lilienfeld & Andrews, 1996; Lilienfeld & Widows, 2005; Uzieblo, Verschuere & Crombez, 2007) as a psychopathy screener before the start of their class. Students who scored above the 90th percentile or between the 25th and 50th percentile of their sex group - as defined in the PPI-R norms by Uzieblo and colleagues (2007) - were approached to participate in the supplementary study as high FD or control participant respectively (n=40). Participation, for which students received €10, was voluntary. All participants were informed that they could exit the study at any moment. Due to restriction of range, 5 students were excluded after their PPI-R score deviated more than 10 points from their FD-screening score. Seven participants showed unreliable pupil data in more than two trials and were excluded from the analysis. This resulted in a total number of 28 participants with ages ranging from 18 to 23, divided
into two groups: a control group (n=14, 6 males, mean age=19.43), and a high FD group (n=14, mean age=20.07, 6 males).
Apparatus
To assess the extent to which a person possesses psychopathic traits, two questionnaires were used: the PPI-R and the YPI (Youth Psychopathic Traits Inventory, Andershed, Kerr, Stattin, & Levander, 2002). The PPI-R is a self-report questionnaire consisting of 154 questions divisible into two uncorrelated factors: the Impulsivity-Antisocial factor and the Fearless-Dominance factor. The FD-factor consists of 45 statements in three subgroups (Social Influence, Fearlessness and Stress Immunity). Statements were answered on a 4-point scale (1 = false, 2 = somewhat false, 3 = somewhat true, 4 = true). Examples of Fearless Dominance statements are: “When I meet people, I can elicit interest in me with a simple smile”, or “I agree with the motto: ‘When life gets too boring, take risks’”. The PPI is validated for students and has been translated in Dutch (Uzieblo et al., 2007). It has very high internal consistency for the total score and the factor scores (α .81 to .92) and satisfactory to good internal consistency for the separate scales (α .76 up to .90). For the validity scales, internal consistency is low to sufficient. Test-retest reliability is very high, especially for the Fearless Dominance factor (r = .91). Convergent- discriminant- and external validity of the PPI-R factors appear to be good (Uzieblo, Verschuere, Van den Bussche & Crombez, 2010). The Dutch translation of the YPI (Das & De Ruiter, 2003) is a self-report questionnaire that consists of 50 questions; 5 for each of the 10 core psychopathy traits based on Cooke & Michie’s (2001) model of psychopathy. Each question is answered on a 4-point scale. The psychometric properties of the YPI are generally described as satisfying, and it has been validated for assessing personality traits that fit the psychopathic image in young adult males (Colins, Noom & Vanderplasschen, 2012; Das, 2008).
Based on the theoretical background of the fearlessness-psychopathy relation, participants also filled in the State Trait Anxiety Inventory-Y (STAI-Y, Spielberger et al., 1983). The STAI-Y assesses different types of anxiety: anxiety in reaction to an event, and anxiety as a personality trait. It was used in the analysis to interpret any findings regarding the main research questions. The Dutch version of the STAI-Y is called the Zelf-Beoordelingsvragenlijst (ZBV, Van der Ploeg 2000). It has high internal consistency and test-retest reliability and sufficient criterion validity and construct validity (Evers, Van Vliet-Mulder & Groot, 2000).
During the fear conditioning task, an Eyelink 1000 eye-tracker (SR Research, Mississauga, Ontario, Canada) followed participants’ eyes, hereby assessing pupil size, dilations, blinks and saccades. The eye tracker was sampled at 1000Hz, using an infrared light, invisible to the human eye. This way of assessing pupil size is comparable to similar research (Bradley et al., 2008; Gilzenrat, Nieuwenhuis, Jepma & Cohen, 2010; Jepma & Nieuwenhuis, 2011). The unconditioned stimulus (US), a mild electric shock, was administered and controlled by a FDA cleared and MDD CE certified current stimulator (model DS7A, Digitimer, U.K.), internally secured against giving off too much energy. This method is frequently used in fear conditioning research and has been classified as minimally intrusive and as exposing participants to only a marginal chance of damage by the Dutch Central Committee on Research regarding Human Subjects (CCMO) and the Medical Ethics Review Committee of the VU University medical centre (VUmc METc).
Procedure
The first PPI-R screening was carried out before a lecture. Students who agreed to participate were met at the psychology faculty of their university for the supplement of the study, which took approximately 50 minutes. Participants were asked to fill out the PPI-R, the YPI and the
STAI-Y, after which the fear conditioning task started. The classical fear conditioning task consists of three phases: a habituation phase, an acquisition phase and an extinction phase (Birbaumer et al., 2005). The mild electric shock, serving as the US, was applied to the right ankle during 20 milliseconds. The shock started at 400 V and was gradually increased in steps of 2 mA until the participant experienced it as “unpleasant but not painful” or until a maximum of 45 mA.
Photographs of two male neutral faces, similar to an earlier study on fear processing in psychopathy, served as the conditioned stimulus (CS, figure 1). They were randomly assigned to function as either the CS+ or the CS-. The photographs were similar in all characteristics, except that one of the faces had a moustache. In the habituation phase, participants saw the two faces alternately (CS+ and CS-) so that both were displayed 4 times, neither followed by a shock (US). In the acquisition phase, both pictures were shown 16 times; the CS+ was followed by the US half of the time, while the CS- was never followed by the US. In the extinction phase, CS+ and CS- were shown 8 times, never followed by the US. The photos were displayed for 4 seconds at all trials, with an inter-trial interval of between 10 and 12 seconds.
Design
To measure baseline pupil size, the mean pupil size was taken from 6 seconds prior to picture onset. This resulted in 48 baselines, consisting of 48 trials (8 habituation trials, 32 acquisition and 8 extinction trials). The effect of blinks and pupil dilation in response to stimulus onset were filtered out. To assess fear conditioning, the maximum pupil dilation was taken from 0.5 seconds to 3.5 post picture onset. This is shown in Figure 1 and Figure 2. For more information regarding the way the pupil data was handled and analyzed, see Appendix A.
The design of this study had one independent variable: fearless dominance traits. Besides creating a high FD group and a control group, FD-traits were also used as a continuous variable for the purpose of a correlation. The correlation functioned as an effect size in this study, expressing the strength of the relation between the two variables. The dependent variables were baseline pupil size and pupil dilations in reaction to the presented stimuli (both in different task-phases and total task-phase). Controlled variables were age, gender and left- or right-handedness.
After smoothing the data in MATLAB 2014a (MathWorks Inc.) the analysis was performed with IBM SPSS Statistics for Windows version 19 (IBM Corp., 2010). To test the first hypothesis, an independent ANOVA design was used to control for gender differences, and a Pearson Correlation was done to measure the relation between Fearless Dominance personality traits and baseline pupil size. The same was done for the second hypothesis, using maximum pupil dilation instead of baseline size.
Figure 1. Duration of trials and inter-trial intervals during fear conditioning task.
Pictures were shown for 4 seconds. In between pictures, the inter-trial interval lasted between 10 and 12 seconds.
Figure 2. Collection of and conditioning data in the fear conditioning task.
For conditioning data, the maximum dilation was calculated from the period between 0.5 seconds and 3.5 seconds post picture onset. For baseline data, the mean pupil size 6 seconds prior to picture onset was used.
Results
Data preparation
To check for any unwanted effect of the electric shock on the baseline pupil size, a paired t-test was done to examine any differences in baseline pupil size between shock and no-shock trials. There was a significant difference between mean pupil size in the acquisition phase with (M=6931.58, SD=1252.94) and without (M=6866.56, SD=1268.05) shock trials,
t(27)=4.35, p<.001. Thus, the rest of the analysis was carried out using the mean acquisition
baseline without shock trials. The Shapiro-Wilk test for normality reported no significance for Fearless Dominance scores, PPI-R, YPI and STAI-Y scores or pupillary data in any of the trials, nor did review of skewness and kurtosis statistics suggest violation of the assumption of normality. The S-W test was significant for age, however, SW(47)=.87, p<.001. A histogram showed a negatively skewed distribution with most participants within the 18 to 20 range and less in the 21 to 23 tail. However, because the age range was quite small and because age did
not correlate with Fearless Dominance or PPI-R total scores, this was not considered problematic. An independent t-test showed no effect of gender on Fearless Dominance scores, PPI-R total, YPI or STAI-Y scores or on any of the baseline trials.
After excluding participants with unreliable FD scores, an independent t-test was conducted to check the assignment of participants to the control or high FD group. There was a significant difference between the control group (M=99.64, SD=10.21) and the high FD group (M=141.5, SD=13.17) in Fearless Dominance scores, t(26)=-9.40, p<.001. Groups differed significantly on PPI-R total scores, t(26)=-6.46, p<.001, on YPI scores, t(26)=-2.72,
p=.011, and STAI-Y scores, t(26)=4.03, p<.001. The high FD group displayed higher FD-,
PPI-R- and YPI-scores, and displayed lower STAI-Y scores than the control group. Chosen shock voltage did not differ between groups. Scores on the Fearless Dominance segment of the PPI-R, PPI-R total scores, scores on the YPI and STAI-Y are displayed in Table 1, as are the means and standard deviations of the voltage at which participants set their shock level.
Table 1. Scores on the Fearless Dominance scale, Psychopathic Personality Inventory
(PPI-R), Youth Psychopathic Traits Inventory (YPI), State Trait Anxiety Inventory (STAI), and set shock voltage in different groups.
*p<.05, **p<.001 Control (n=14) High FD (n=14) M (SD) M (SD) P Fearless Dominance 99.64 (10.21) 141.5 (13.17) <.001** PPI-R 269.14 (16.36) 316.57 (22.08) <.001** YPI 38.36 (12.31) 53.86 (17.4) .011* STAI-Y 36.79 (12.31) 26.79 (4.25) <.001** Shock 279.29 (108.87) 310 (120.64) .486
Hypothesis 1: Baseline pupil size and Fearless Dominance traits
Levene’s test for equality of variances was not significant for any phase of the fear conditioning task. After checking assumptions of normality, a One-way ANOVA was conducted to investigate differences in baseline pupil size between control and high FD groups. There was a significant effect of group on total baseline pupil size throughout the task, F(1, 24) = 4.90, p = .037. In the different phases of the task, there was a significant effect of group on baseline pupil size in Late Acquisition, F(1, 26) = 5.95, p = .022. The effect of group on baseline pupil size was also significant in Extinction, F(1, 26) = 5.50, p = .027. Means and standard deviations of baseline pupil sizes per group are displayed in Table 2.
A t-test was conducted to assess which group showed larger pupil sizes. This showed a significant difference in total baseline pupil size between the two groups, t(24)=2.22, p=.037. Participants in the high FD group displayed significant smaller baseline pupil sizes over the averaged whole of the fear conditioning task. There was no significant difference between the two groups in the habituation phase or in the early acquisition phase. The two groups differed significantly in late acquisition, t(26)=2.44, p=.022, and in extinction, t(26)=2.35, p=.027. In all these phases, the high FD group displayed smaller baseline pupil sizes than the control group. A bivariate correlation showed a significant relation between FD traits and total baseline pupil size, r(25)=-.40, p=.046. This negative correlation suggests that higher FD scores are related to smaller baseline pupil sizes.
Table 2. Baseline pupil sizes, divided by group and task phase.
*p<.05
Post hoc
To examine possible gender differences, the analysis was repeated for males and females separately. For females, there was no difference in pupil size between the high FD and control group in any of the task phases. For males, however, there was a significant difference in baseline pupil size between the high FD and the control group in the total task average,
t(9)=3.90, p=.004. The male groups differed significantly in baseline pupil size in habituation, t(10)=2.54, p=.029, early acquisition, t(10)=3.15, p=.01, late acquisition, t(10)=2.94, p=.015,
and extinction, t(10)=2.86, p=.017. Males in the high FD group displayed smaller baseline pupil size than males in the control group in all phases. This is visualized in Appendix B.
A Pearson correlation displayed no significant relation between FD traits and baseline pupil size for females in any of the task phases. For males, on the contrary, the relation between FD scores and total baseline pupil size was significant, r(11)=-.76, p=.007. The relation between FD scores and baseline pupil size was also significant in habituation, r(12)=-.62, p=.03, early acquisition, r(12)=-.68, p=.016, late acquisition, r(12)=-.620, p=.032, and
Control (n=14) High FD (n=14) M (SD) M (SD) P Habituation 7777.59 (1082.08) 7176.90 (1211.27) .178 Early Acquisition 7366.35 (856.41) 6501.60 (1428.66) .063 Late Acquisition 7377.28 (1033.86) 6228.35* (1426.82) .022* Extinction 7512.33 (1027.28) 6447.62* (1352.10) .027* Total 7597.03 (872.75) 6605.04* (1358.99) .037*
extinction, r(12)=-.58, p=.05. For males, higher FD scores were associated with smaller baseline pupil size in all task phases. Baseline pupil size did not correlate with scores on the PPI-R, the YPI or STAI.
Hypothesis 2: Pupil dilations and Fearless Dominance traits
Pupil dilation responses (PDR’s) in CS+ and CS- trials were calculated for each trial and each task phase by subtracting the baseline from the maximum dilation. If pupil dilation is a valid measure of fear conditioning, one would expect to see response differentiation: a significant difference between pupil dilation responses in habituation and acquisition (equal dilation responses to CS+ and CS- trials in habituation, and larger pupil dilation responses to CS+ than to CS- trials in acquisition). There was no effect of age or gender on pupil dilation responses.
To assess if the pupil showed fear conditioning, pupil dilation responses were compared between CS+ and CS- trials. A paired t-test showed that - for the whole group - there was no significant difference between PDR’s in CS+ and CS- trials in habituation, suggesting no response differentiation. The group as a whole did however display significant response differentiation in acquisition, t(27)=-2.30, p=.029. This difference in pupil dilation responses between CS+ and CS- trials was significant only for early acquisition, t(27)=-2.69,
p=.012. In these phases, the pupil dilation response was larger in CS+ trials than in CS- trials.
This was taken as a sign that the fear conditioning task was effective. In extinction however, pupil dilation response to CS+ trials was significantly smaller than the response to CS- trials,
t(27)=-3.69, p<.001 When looked at the groups separately, both the control and the high FD
group showed no response differentiation in the habituation phase. For the high FD group the difference between PDR’s to CS+ and CS- trials was significant only in extinction, t(13)=-2.64, p=.02. The PDR’s to CS+ were smaller than the PDR’s to CS-. The control group displayed significantly different PDR’s in CS+ and CS- trials in total acquisition, t(13)=4.62,
p<.001, early acquisition, t(13)=2.91, p=.012, late acquisition, t(13)=2.49, p=.027, and
extinction, t(13)=-2.48, p=.028. In the control group, participants showed greater pupil dilations in response to CS+ than CS- trials in acquisition. In extinction, this process was reversed. Means and standard deviations of PDR’s are displayed in Table 3.
Table 3. Pupil dilation responses (PDR) in different task phases, divided by group and trial.
*p<.05, **p<.001. P value displays significant within-group differences in pupil dilation responses to CS+ and CS-. CS+ (n=14) CS- (n=14) M (SD) M (SD) P PDR Habituation Control High FD 420.38 (1335.78) 902.05 (2058.1) 426.95 (1456.35) 796.02 (2106.98) .960 .475 PDR Total Acq. Control High FD 1000.95 (1227.67) 1647.21 (1839.35) 632.08 (1242.12) 1683.4 (1800.21) <.001** .710 PDR Early Acq. Control High FD 754.53 (1320.33) 1381.47 (1783.32) 521.06 (1253.69) 1310.15 (1810.91) .012* .368 PDR Late Acq. Control High FD 722.90 (1172.38) 1510.77 (1778.54) 369.52 (1246.28) 1577.75 (1787.76) .027* .558 PDR Extinction Control High FD 279.64 (1226.28) 934.43 (1554.42) 615.54 (1314.33) 1294.16 (1683.33) .028* .020*
To examine group differences in conditioning throughout the task phases, a One-way ANOVA was conducted. This showed there was a significant difference in the degree of response differentiation between the control and high FD group in acquisition, F(1, 26)=10.65, p=.003. This difference turned out to be significant only in late acquisition, F(1, 26)=5.42, p=.028. Subsequently, an independent t-test showed no difference in response differentiation between the control and high FD group in habituation. The t-test showed a significant difference in total acquisition, t(26)=3.26, p=.003, and late acquisition, t(26)=2.33,
p=.028. The control group showed greater response differentiation than the high FD group in
the (late) acquisition phase of the fear conditioning task. Response differentiation in different task phases is displayed in Table 4.
Table 4. Response differentiation (RD) in different task phases, divided by group
*p<.05
Discussion
The results showed that, as was hypothesized, individuals with more Fearless Dominance personality traits had smaller baseline pupils. Moreover, results supported the assumption that pupil dilation can be used as a measure of fear conditioning, and showed that individuals with
Control (n=14) High FD (n=14) M (SD) M (SD) P RD Habituation -6.57 (482.10) 106.04 (538.99) .565 RD Total Acq. 368.88* (298.60) -36.19 (355.60) .003* RD Early Acq. 233.47 (300.35) 71.32 (286.05) .156 RD Late Acq. 353.38* (531.70) -66.98 (417.24) .028* RD Extinction -355.89 (507.07) -359.73 (510.28) .902
more Fearless Dominance traits demonstrated no fear conditioning in their pupillary responses. Because pupil size is regulated by the autonomic nervous system, these results support earlier findings of reduced autonomic activity in psychopathic individuals. These findings are in accordance with the low arousal and low fear theories, suggesting psychopathy is characterized by reduced activity of the stress system.
A first important limitation in the interpretation of the results regards the baseline pupil data. Results showed a difference in baseline pupil size between groups only in later phases of the fear conditioning task, and not in habituation. That means the difference is only visible from the moment the shocks are given. One possible explanation for this finding is that it reflects a difference between groups in their response to slight arousal. The word ‘baseline’ implies that the pupil is at rest when measured. However, in this study it might reflect a difference in how the two groups handle low levels of arousal: being in a laboratory with strangers, having an eye tracker close to your face, receiving shocks in an unfamiliar task. Portnoy and colleagues (2014) used the same explanation when interpreting the low resting heart rate in psychopathic individuals: the test environment in itself is a source of stress. Another possible limitation comes from the fact that no correction was applied for the substantial amount of repeated testing. This increases the probability of a type one error - a false positive - considerably (Cabin & Mitchel, 2000). Given the high variation in pupil size and reactions and the low number of participants, any existing effect of psychopathic personality traits on the pupil may have not been visible with a Bonferroni correction. However, with the effect size - as measured by the correlation between psychopathic traits and pupil size - being moderate to strong, it appears the findings are not just a result of repeated testing. It is important to repeat the study with a higher number of participants and a correction for repeated testing.
Related to this problem, analysing the baseline results for male and female participants separately decreased the number in each group even further. Besides the expectation that females would show relatively larger pupils throughout the task (Partala & Surakka, 2003), there was no solid a priori reason to conduct this analysis separately for males and females. Therefore, these results have to be interpreted with caution. The results were remarkable however; suggesting that the relationship between Fearless Dominance traits and pupil size is present only in male participants (see Appendix B). For females, there was no relation between pupil size and psychopathic traits. This is an important finding, since growing body of literature questions if the construct of psychopathy, which was originally formed on a male image, is generalizable to females (Verona & Vitale, 2006). Factors such as gender-role socialization and biological sex differences might result in a different expression of psychopathic traits in men and women (Nicholls & Petrilla, 2005). It would be informative to conduct a study with a larger amount of participants to see if this gender difference is genuine. The very strong correlation between men’s Fearless Dominance traits and pupil size suggests this might be a genuine gender difference. If that is the case, pupillometry might prove to be valuable in research into the genetic and underlying psychophysiological mechanisms of psychopathy and gender differences in this disorder.
Another outcome worth investigating further is the finding that scores on PPI-R-total are not related to pupil size; only Fearless Dominance traits show this relation. This is consistent with the notion that an interpersonal/affective component distinguishes psychopathy from Antisocial Personality Disorder (ASP): while psychopathy is characterized by diminished autonomic responses, ASP is not. This fits the differentiation between the “primary psychopath”, displaying both behavioral and emotional symptoms, and the “secondary psychopath”, displaying mainly behavioral symptoms.
Another surprising finding was the weakened response to the CS+ in extinction, compared to the CS-. This “reversed response differentiation” might suggest some sort of “unlearning” process. This is not a well-documented effect, which might mean it’s a false positive. However, it could also mean the pupil is a sensitive measure, and shows delicate differences that are not visible in skin conductance and heart rate. Also regarding the clinical utility of pupil dilation as a measure of fear conditioning, it would be interesting to compare pupil dilation to other well-studied psychophysiological measurements of conditioned fear - such as skin conductance, heart rate or the startle reflex. This way, any benefits of pupillometry might be discovered regarding reliability, intrusiveness, costs or validity. For example, while a subject may be aware of increased heart rate, increased sweat of a startle reflex, pupillary dilations are not consciously observed. This may be an advantage of the use of pupillometry in fear conditioning research, as the pupil thus provides an honest and very hard to influence correlate of conditioned fear.
Lastly, the results are interesting regarding the neurobiological study of psychopathy, since pupil dilation is mediated almost exclusively by Norepinephrine. Blair (2003) suggested that pathologies in the norepinephrine system could lead to amygdala dysfunction, which is in turn associated with psychopathy. Patients experiencing anxiety symptoms and panic attacks display a greater sensitivity to noradrenergic function. Their symptoms are the opposite of the symptoms seen in psychopathy, such as greater attention to distress cues, faster heart rate, higher levels of cortisol and HPA-axis activity, and ANS and amygdala hyperresponsivity (Charney, Heninger & Breier, 1984; Friedman & Thayer, 1998; Rauch, Shin & Wright, 2003; Abelson, Khan, Liberzon & Young, 2007). Pupillometry could be an efficient and unobtrusive way of studying norepinephrine levels in psychopathic individuals. This could in turn lead to new insights in the neuropsychological mechanisms underlying psychopathy. For example, since norepinephrine plays a large role in attentional processes, pupillometry could be very
valuable in studying attentional deficits in psychopathy, as postulated by Newman and colleagues (2010).
All in all, pupillometry is presented as a promising new development in psychopathy research. The findings of the current study encourage future research into existing theories about psychopathy, such as the low arousal and low fear hypotheses, and new possibilities in studying the neurobiological and psychophysiological mechanisms underlying this disorder, relating to gender differences, attentional deficits and subtypes of psychopathy.
Appendix A. Pupil analysis explained.
Pupillary data from the eye tracker was pre-processed by MATLAB 2014a (MathWorks Inc), merging all the data points documented by the eye-tracker into 1 baseline size per trial for the baseline data, and 12 data points per trial for the dilation data. According to existing literature, the pupil reaches its maximum constriction in response to a light stimulus in between 0.6 and 1.6 seconds after stimulus onset and returns to baseline approximately a second later (Bradley et al., 2008; Steinhauer, Condray & Kasparek, 2000). Mean baseline pupil size was calculated from 6 seconds prior to picture onset, thus leaving the first 4 to 10 seconds after the previous picture offset out of the baseline measurement. This way, any content and/or luminance effects of the previous picture on the pupil size would have seized as much as possible, leading to a cleaner baseline picture. The baseline measurement ended before the next picture onset, as not to influence baseline with the pupillary response to that picture. The effect of the unconditioned stimulus (US) on the pupil size was taken into account as well, seeing that emotionally negative stimuli dilate the pupil for at least 2 seconds after stimulus onset (Partala, Jokiniemi & Surakka, 2000; Partala & Surakka, 2003). Even though any pupillary response to the US might have seized before the start of the baseline (4 seconds after US offset), analysis were conducted with and without CS+ trials.
For the analysis regarding the degree of conditioning, pupil dilation in response to the stimulus (PDR) was taken as the maximum dilation during 0.5 to 3.5 seconds post picture onset (Bradley et al, 2008; Nieuwenhuis et al., 2011). Since the pupillary light reflex is a great deal stronger than the pupil dilation following an emotional stimulus, it is theoretically possible that the constriction of the pupil following the light reflex is so strong that the maximum dilation in response to the conditioning task does not rise above the ‘baseline’ from the first 500 milliseconds after picture offset. In that case, the largest pupil size measured is a point even before the light reflex kicks in; which means one is not assessing conditioning but
in fact looking at baseline pupil size. To avoid getting this first 500 millisecond ‘baseline’ as the maximum dilation, these 500 milliseconds were not taken into the scope of the conditioning data. The end of the scope was chosen to be 3.5 seconds after picture onset, as to make sure no effect of a possible US could confound the data.
Unfortunately, it was technically impossible to calculate a single point as the maximum dilation, because the computer could not handle such a load. Therefore, a smoothing procedure yielded ‘bites’ of 250 milliseconds from 0.5 up to 3.5 seconds after picture onset. This resulted in 12 pieces of 250 milliseconds per trial: a total of 576 ‘bites’ and 48 maximum pupil dilations. To interpret these pieces of pupil data as a degree of conditioning, first the pupillary dilation responses were calculated as the difference between baseline pupil size and maximum dilation. The greater the pupillary dilation response, the greater the difference is between baseline and maximum dilation in that trial. Following this,
response differentiation could be calculated; this is the difference between PDR’s to CS+
trials and PDR’s to CS- trials. Response differentiation can be understood as the extent to which participants’ pupils respond differently to trials where they are presented with the CS+ versus CS- trials and are thus a general degree of conditioning for each individual participant: the higher one’s response differentiation, the stronger they display conditioning.
Appendix B. Group (control vs. high FD) differences in baseline pupil size, for both genders
(fig. 1) and split by gender for males (fig. 2) and females (fig. 3) separately.
Figure. 1. Baseline pupil size during 48 trials of the fear conditioning task. The blue line
Figure 2. Males show a significant difference in baseline pupil size in all phases.
References
Abelson, J. L., Khan, S., Liberzon, I., & Young, E. A. (2007). HPA axis activity in patients with panic disorder: Review and synthesis of four studies. Depression and Anxiety,
24, 66–76.
Andershed, H., Kerr, M., Stattin, H., & Levander, S. (2002). Psychopathic traits in
non-referred youths: A new assessment tool. Psychopaths: Current International
Perspectives, 131-158.
Beatty, J., & Lucero-Wagoner, B. (2000). The pupillary system. Handbook of
Psychophysiology, 2, 142-162.
Birbaumer, N., Veit, R., Lotze, M., Erb, M., Hermann, C., Grodd, W., & Flor, H. (2005). Deficient fear conditioning in psychopathy: A functional magnetic resonance imaging study. Archives of General Psychiatry, 62(7), 799-805.
Blackburn, R. (1979). Cortical and autonomic arousal in primary and secondary psychopaths. Psychophysiology, 16(2), 143-150.
Blair, R. J. R. (2003). Neurobiological basis of psychopathy. The British Journal of
Psychiatry, 182(1), 5-7.
Blair, R. J. R. (2004). The emergence of psychopathy: Implications for the
Blair, R. J. R., Peschardt, K. S., Budhani, S., Mitchell, D. G. V., & Pine, D. S. (2006). The development of psychopathy. Journal of Child Psychology and Psychiatry, 47(3‐4), 262-276.
Bradley, M. M., Miccoli, L., Escrig, M. A., & Lang, P. J. (2008). The pupil as a measure of emotional arousal and autonomic activation. Psychophysiology, 45(4), 602-607.
Cabin, R. J., & Mitchell, R. J. (2000). To Bonferroni or not to Bonferroni: When and how are the questions. Bulletin of the Ecological Society of America, 246-248.
Charney, D. S., Heninger, G. R., & Breier, A. (1984). Noradrenergic function in panic anxiety: effects of yohimbine in healthy subjects and patients with agoraphobia and panic disorder. Archives of General Psychiatry, 41(8), 751-763.
Cleckley, H. (1941). The mask of sanity; an attempt to reinterpret the so-called psychopathic
personality.Oxford, England: Mosby.
Cleckley, H. (1976). The mask of insanity. 5th ed. St Louis, MO: Mosby.
Cohn, M. D., Popma, A., van den Brink, W., Pape, L. E., Kindt, M., van Domburgh, L., ... & Veltman, D. J. (2013). Fear conditioning, persistence of disruptive behavior and psychopathic traits: An fMRI study. Translational Psychiatry, 3(10), 319.
inventory-short version: A further test of the internal consistency and criterion validity. Journal of Psychopathology and Behavioral Assessment, 34(4), 476-486.
Cooke, D. J., & Michie, C. (2001). Refining the construct of psychopathy: Towards a hierarchical model. Psychological Assessment, 13(2), 171.
Cox, J., Edens, J. F., Magyar, M. S., Lilienfeld, S. O., Douglas, K. S., & Poythress, N. G. (2013). Using the Psychopathic Personality Inventory to identify subtypes of antisocial personality disorder. Journal of Criminal Justice, 41(2), 125-134.
Das, J. (2008). Psychopathic traits in Dutch adolescent offender and community samples: Relationships with gender, age, institutional misbehavior, substance use and interpersonal style. Maastricht: Universiteit Maastricht.
Das, J., & de Ruiter, C. (2003). Youth Psychopathic traits Inventory: Geautoriseerde Nederlandse vertaling.
Dolan, M. (2004). Psychopathic personality in young people. Advances in Psychiatric
Treatment, 10(6), 466-473.
Dvorak‐Bertsch, J. D., Curtin, J. J., Rubinstein, T. J., & Newman, J. P. (2009). Psychopathic traits moderate the interaction between cognitive and affective
processing. Psychophysiology, 46(5), 913-921.
test research in the Netherlands (COTAN), part I and II. Assen: Van Gorcum.
Fowles, D. C., & Dindo, L. (2006). A dual-deficit model of psychopathy. Handbook of
Psychopathy, 14-34.
Friedman, B. H., & Thayer, J. F. (1998). Autonomic balance revisited: Panic anxiety and heart rate variability. Journal of Psychosomatic Research, 44(1), 133-151.
Gilzenrat, M. S., Nieuwenhuis, S., Jepma, M., & Cohen, J. D. (2010). Pupil size tracks changes in control state predicted by the adaptive gain theory of locus coeruleus function. Cognitive, Affective, & Behavioral Neuroscience,10(2), 252-269.
Hansen, A. L., Johnsen, B. H., Thornton, D., Waage, L., & Thayer, J. F. (2007). Facets of psychopathy, heart rate variability and cognitive function. Journal of Personality
Disorders, 21(5), 568-582.
Hare, R. D. (1970). Psychopathy: Theory and research. Oxford, England: John Wiley.
Hare, R. D. (1978). Psychopathy and electrodermal responses to nonsignal stimulation. Biological Psychology, 6(4), 237-246.
Hare, R. D. (1996). Psychopathy and antisocial personality disorder: A case of diagnostic confusion. Psychiatric Times, 13(2), 39-40.
antisocial personality disorder. Journal of Abnormal Psychology, 100(3), 391.
Hare, R. D., & Neumann, C. S. (2010). Psychopathy: Assessment and forensic
implications. Responsibility and Psychopathy: Interfacing Law, Psychiatry and
Philosophy, 93-123.
Holmes, D. S. (1967). Pupillary response, conditioning, and personality. Journal of
Personality and Social Psychology, 5(1), 98.
Jepma, M., & Nieuwenhuis, S. (2011). Pupil size predicts changes in the exploration- exploitation trade-off: Evidence for the adaptive gain theory. Journal of Cognitive
Neuroscience, 23(7), 1587-1596.
Jones, R. (1990). Do women and myopes have larger pupils? Investigative Ophthalmology &
Visual Science, 31(7), 1413-1415.
Kalat, J. W. (2009). Biological Psychology. Boston, MA: Cengage Learning.
Lilienfeld, S. O., & Andrews, B. P. (1996). Development and preliminary validation of a self- report measure of psychopathic personality traits in noncriminal population. Journal of
Personality Assessment, 66(3), 488-524.
Lilienfeld, S. O., & Widows, M. R. (2005). Psychopathic Personality Inventory-Revised: Professional Manual. Lutz, Florida: Psychological Assessment Resources, Inc.
López, R., Poy, R., Patrick, C. J., & Moltó, J. (2013). Deficient fear conditioning and self‐ reported psychopathy: The role of fearless dominance. Psychophysiology, 50(2), 210- 218.
Lopez-Duran, N. L., Olson, S. L., Hajal, N. J., Felt, B. T., & Vazquez, D. M. (2009).
Hypothalamic pituitary adrenal axis functioning in reactive and proactive aggression in children. Journal of Abnormal Child Psychology, 37(2), 169-182.
Lorber, M. F. (2004). Psychophysiology of aggression, psychopathy, and conduct problems: A meta-analysis. Psychological Bulletin, 130(4), 531.
Lykken, D. T. (1957). A study of anxiety in the sociopathic personality. The Journal of
Abnormal and Social Psychology, 55(1), 6.
Lykken, D. T. (1995). The antisocial personalities. Hillsdale, NJ:Erlbaum.
Newman, J. P., Curtin, J. J., Bertsch, J. D., & Baskin-Sommers, A. R. (2010). Attention moderates the fearlessness of psychopathic offenders. Biological Psychiatry, 67(1), 66-70.
Newman, J. P., & Kosson, D.S. (1986). Passive avoidance learning in psychopathic and nonpsychopathic offenders. Journal of Abnormal Psychology. 96, 257–263.
Nicholls, T. L., & Patrilla, J. (2005). Gender and psychopathy: An overview of important issues and introduction to the special issue. Behavioral Sciences and the Law, 26(6),
729-741.
Öhman, A. (2002). Automaticity and the amygdala: Nonconscious responses to emotional faces. Current Directions in Psychological Science, 11(2), 62-66.
Ortiz, J., & Raine, A. (2004). Heart rate level and antisocial behavior in children and adolescents: A meta-analysis. Journal of the American Academy of Child &
Adolescent Psychiatry, 43(2), 154-162.
Partala, T., Jokiniemi, M., & Surakka, V. (2000). Pupillary responses to emotionally provocative stimuli. Eye tracking Research & Applications Symposium, 123-129.
Partala, T., & Surakka, V. (2003). Pupil size variation as an indication of affective processing. International Journal of Human-Computer Studies, 59(1), 185-198.
Patrick, C. J., Bradley, M. M., & Lang, P. J. (1993). Emotion in the criminal psychopath: Startle reflex modulation. Journal of Abnormal Psychology, 102(1), 82.
Patrick, C. J., & Brislin, S. J. (2007). Antisocial Personality Disorder/Psychopathy. The
Encyclopedia of Clinical Psychology, 1-10.
Ploeg, van der, H.M. (2000). Handleiding bij de Zelf-Beoordelings Vragenlijst. Amsterdam: Harcourt Assessment B.V.
Doreleijers, T. A. (2006). Hypothalamus pituitary adrenal axis and autonomic activity during stress in delinquent male adolescents and controls. Psychoneuroendocrinology,
31(8), 948-957.
Portnoy, J., Raine, A., Chen, F. R., Pardini, D., Loeber, R., & Jennings, J. R. (2014). Heart rate and antisocial behavior: The mediating role of impulsive sensation
seeking. Criminology, 52(2), 292-311.
Rauch, S. L., Shin, L. M., & Wright, C. I. (2003). Neuroimaging studies of amygdala function in anxiety disorders. Annals of the New York Academy of Sciences, 985(1), 389-410.
Rilling, J. K., Glenn, A. L., Jairam, M. R., Pagnoni, G., Goldsmith, D. R., Elfenbein, H. A., & Lilienfeld, S. O. (2007). Neural correlates of social cooperation and non-cooperation as a function of psychopathy. Biological Psychiatry, 61(11), 1260-1271.
Spielberger, C. D., Gorsuch, R. C., Lushene, R. E., Vagg, P. R., & Jacobs, G. A. (1983).
Manual for the State-Trait Anxiety Inventory. Palo Alto: Consulting Psychologists
Press.
Steinhauer, S. R., Condray, R., & Kasparek, A. (2000). Cognitive modulation of midbrain function: Task-induced reduction of the pupillary light reflex. International Journal of
Psychophysiology, 39(1), 21-30.
Uzieblo, K., Verschueure, B., Crombez, G. (2007). The Psychopathic Personality Inventory: Construct validity of the two-factor structure. Personality and Individual Differences
43, 657-667.
Uzieblo, K., Verschuere, B., Van den Bussche, E., & Crombez, G. (2010). The validity of the Psychopathic Personality Inventory—Revised in a community sample.
Assessment, 17(3), 334-346.
Veit, R., Konicar, L., Klinzing, J. G., Barth, B., Yilmaz, Ö., & Birbaumer, N. (2013).
Deficient fear conditioning in psychopathy as a function of interpersonal and affective disturbances. Frontiers in Human Neuroscience, 7.
Verona, E., & Vitale, J. (2006). Psychopathy in Women: Assessment, Manifestations, and Etiology. Handbook of Psychopathy (415-436). New York, NY: Guilford Press.
Visser, R. M., Scholte, H. S., Beemsterboer, T., & Kindt, M. (2013). Neural pattern similarity predicts long-term fear memory. Nature Neuroscience, 16(4), 388-390.
Winn, B., Whitaker, D., Elliott, D. B., & Phillips, N. J. (1994). Factors affecting light-adapted pupil size in normal human subjects. Investigative Ophthalmology & Visual
