Early Temperamental and Psychophysiological Precursors of Adult Psychopathic Personality

Journal of Abnormal Psychology
Copyright 2007 by the American Psychological Association
2007, Vol. 116, No. 3, 508 –518
0021-843X/07/$12.00
DOI: 10.1037/0021-843X.116.3.508
Early Temperamental and Psychophysiological Precursors of Adult
Psychopathic Personality
Andrea L. Glenn and Adrian Raine
Peter H. Venables
University of Southern California
University of York
Sarnoff A. Mednick
University of Southern California
Emerging research on psychopathy in children and adolescents raises the question of whether indicators,
such as temperament or psychophysiology, exist very early in life in those with a psychopathic-like
personality in adulthood. This study tests the hypothesis that individuals who are more psychopathic in
adulthood would be less fearful and inhibited and more stimulation seeking/sociable at age 3 and that
they would also show reduced age 3 skin-conductance (SC) responsivity. In a community sample of 335
3-year-olds, behavioral measures of temperament were taken and electrodermal activity was recorded in
response to both orienting and aversive tones. R. D. Hare’s (1985) Self-Report Psychopathy scale
(SRP-II) was administered at follow-up at age 28. Individuals scoring higher on the measure were
significantly less fearful and inhibited, were more sociable, and displayed longer SC half-recovery times
to aversive stimuli compared with controls at age 3. Contrary to predictions, they also showed increased
autonomic arousal and SC orienting. Findings appear to be the first to suggest that a prospective link may
exist between temperament and psychophysiology in very young children and psychopathic personality
in adulthood.
Keywords: psychopathy, childhood, temperament, skin conductance, arousal
Despite decades of research on adult psychopathy, very little is
psychosocial hardship in adolescence and early adulthood, these
known about the early antecedents of this disorder. Recent re-
individuals might be less responsive to laboratory stressors. An-
search has begun to identify psychopathic personality in children
other possibility is that individuals characterized as psychopaths in
and adolescents (Frick, O’Brien, Wooton, & McBurnett, 1994;
adulthood possess qualities of disinhibition, stimulation seeking/
Lynam, 1997), raising the question of whether precursors to psy-
sociability, and lack of fear very early in their lives (and prior to
chopathy may exist very early in life. Adults who are psychopathic
the onset of their antisocial lifestyle) and are thus predisposed to a
have been hypothesized as having unique temperaments and psy-
psychopathic personality at a young age. A longitudinal study that
chophysiological characteristics (Frick, 1998; Hare, 1978). A gap
examines the role of early temperament on prospective adult
in the psychopathy literature is the lack of prospective, longitudi-
psychopathic personality would be necessary to help to clarify
nal studies that examine early childhood factors in individuals who
whether these risk factors precede adult psychopathic personality.
later demonstrate psychopathic personality in adulthood.
In addition to this hypothesized temperamental style, psycho-
The personality of psychopathic adults is characterized by a lack
pathic adults have also been characterized by lower psychophys-
of fear and inhibition and a tendency toward stimulation seeking
iological arousal and hyporesponsivity to stimuli. An extensive
(Hare, 1998). However, there are multiple ways in which these
meta-analysis by Lorber (2004) found psychopathic people to have
characteristics may develop. One possibility is that a psychopathic
lower levels of electrodermal activity at rest. Psychopathy has also
way of life leads to the disinhibition, stimulation seeking/
been associated with reduced skin conductance (SC) orienting,
sociability, and lack of fear observed in adults who are psycho-
reflecting reduced allocation of attentional resources to external
pathic. For example, after experiencing trauma or continuous
stimuli (Dawson, Schell, & Filion, 1990). Both Hare (1968) and
Blackburn (1979) have found reduced responding to orienting
stimuli in psychopathic individuals; however, some studies have
Andrea L. Glenn, Adrian Raine, and Sarnoff A. Mednick, Department of
failed to replicate this finding (Aniskiewicz, 1979; Raine & Ven-
Psychology, University of Southern California; Peter H. Venables, Depart-
ables, 1988). It remains unknown whether reduced arousal and
ment of Psychology, University of York, Heslington, York, England.
orienting in psychopathic adults is a characteristic developed in
This study was supported by Independent Scientist Awards K02
adolescence or adulthood, or whether it may be present and influ-
MH01114 and RO1 AA10206 to Adrian Raine from the National Institutes
ential in the first few years of life.
of Health and by a grant to Peter H. Venables from the U.K. Medical
Longer SC half-recovery time is another psychophysiological
Research Council.
characteristic that has been associated with psychopathy. Mednick
Correspondence concerning this article should be addressed to Andrea
L. Glenn, Department of Psychology, SGM 501, University of Southern
(1977) cited three studies prior to 1977 that suggested longer
California, Los Angeles, CA 90089-1061. E-mail: aglenn@usc.edu
recovery time in antisocial individuals. One review of studies
508

---

GLENN, RAINE, VENABLES, AND MEDNICK
509
conducted since 1977 reported seven studies finding significantly
suggesting that our current understanding of the etiology of adult
longer SC half-recovery times in both psychopathic and antisocial
psychopathy might require some revision.
individuals, and one failing to find a significant association (Raine,
1993). Additionally, shorter SC half-recovery time has been iden-
Method
tified as a possible protective factor, as antisocial adolescents who
desisted from adult crime were found to have shorter SC half-
Participants
recovery times than both life-course persistent criminals and con-
trols (Raine, Venables, & Williams, 1996). Long SC half-recovery
The larger population from which the participants were drawn
time has been interpreted by Venables (1975) as representing a
consisted of 1,795 children from the island of Mauritius (a country
closed stance to environmental stimuli, which would include cues
lying in the Indian Ocean between Africa and India). All children
of future punishment. Despite many studies in the adult literature,
born in 1969 in the two towns of Vacoas and Quatre Bornes were
longer SC half-recovery time does not appear to have been exam-
recruited into the study when aged 3 years between September
ined as a potential developmental precursor of adult psychopathy.
1972 and August 1973. The two towns were chosen to be repre-
In an attempt to gain a developmental perspective, recent re-
sentative of the ethnic distribution of the island. Informed consent
search has begun to examine whether the same mechanisms that
was obtained from the mothers at age 3 and from the participants
underlie adult psychopathy may also be present in children and
at age 28.
adolescents with psychopathic-like characteristics. Children with
From this birth cohort, a sample of 335 participants completed
callous and unemotional traits similar to those seen in primary
a self-report measure of psychopathic personality at the age of 28.
psychopathic individuals have been found to demonstrate fearless-
To assess the unbiased selection of this group, we conducted
ness (Frick, Lilienfeld, Ellis, Loney, & Silverthorn, 1999) and a
comparisons of the means of those tested and those not tested on
lack of behavioral inhibition (Frick et al., 2003). Several studies
measures of temperament (inhibition, t[1791]
0.761, p
.447,
have revealed that juveniles with psychopathic traits demonstrate
d
.023; fearfulness, t[1787]
0.649, p
.517, d
.040;
similar deficits observed in psychopathic adults, including reduced
stimulation seeking/sociability, t[1769]
0.375, p
.708, d
responding to distressing stimuli (Kimonis, Frick, Fazekas, &
.023), psychophysiology (nonspecific skin conductance responses
Loney, 2006), difficulty recognizing sadness and fear (Blair, Col-
[SCRs], t[1793]
1.789, p
.074, d
.113; heart rate,
ledge, Murray, & Mitchell, 2001), slower reaction times to nega-
t[1747]
0.479, p
.501, d
.035; SC level left, t[1754]
tive words (Loney, Frick, Clements, Ellis, & Kerlin, 2003), and
0.891, p
.374, d
.000; SC level right, t[1754]
0.043,
deficits in moral and emotional processing (Blair, 1997). In
p
.966, d
.056; average orienting amplitude, t[1793]
preschool-aged children, Fowles, Kochanska, and Murray (2000)
0.112, p
.911, d
.007; average SC half-recovery time,
showed that those characterized by fearless and uninhibited tem-
t[1361]
0.909, p
.087, d
.138; average amplitude to
peraments had reduced SC responsivity. To our knowledge, only
aversive stimuli, t[1793]
0.837, p
.300, d
.047), socio-
two studies thus far have examined the psychophysiological char-
economic status at age 3 (t[1793]
2.267, p
.023, d
acteristics of specific psychopathic characteristics in juveniles.
.147), social adversity at age 3 (t[1793]
1.901, p
.057, d
Blair (1999) found that 13-year-olds with psychopathic tendencies
.116), and IQ (t [1383]
0.642, p
.521, d
.044). Only
showed less SC responding to distress cues and threatening stim-
socioeconomic status was found to be significantly different
uli. Moreover, Fung et al. (2005) found that 16-year-olds scoring
(higher in the untested group than in the tested group). This
high on the Child Psychopathy Scale (Lynam, 1997) showed
variable was consequently entered as a covariate to determine
reduced anticipatory SC responding to white-noise bursts. These
whether it influenced results. Details of the social adversity and IQ
studies suggest that psychophysiological differences may be linked
measures are found in Raine, Yaralian, Reynolds, Venables, and
to psychopathy in adolescents as well as adults, but no studies
Mednick (2002).
appear to have examined the psychophysiological characteristics
Comparisons of gender and ethnicity revealed that the sample
of children who later develop a psychopathic personality.
consisted of more males (61%) than females (39%), and conse-
The present longitudinal study involves a community sample in
quently gender was included as a moderator in all analyses. Ethnic
which psychophysiology and temperament are first measured at
distribution was as follows: Indian 69%, Creole (African origin)
the relatively early age of 3 years, and psychopathic personality is
26%, and others (Chinese, English, French, and ethnically uniden-
assessed 25 years later at age 28 years. It would be predicted that
tified) 5%. Census data for the island as a whole indicated 66%
individuals who scored higher in psychopathy in adulthood would
Indian, 29% Creole, and 5% other, indicating that the study
be characterized as less fearful and inhibited but higher on stim-
achieved its goal of sampling an ethnically representative popula-
ulation seeking/sociability in early childhood. Furthermore, on the
tion.
basis of long-standing theoretical perspectives, these individuals
would be predicted to have lower autonomic arousal, reduced
Age 28 Psychopathy Scale
orienting responses, reduced responding to aversive stimuli, and
longer SC half-recovery time. If reduced psychophysiological ac-
Hare’s (1985) Self-Report Psychopathy scale (SRP-II), admin-
tivity similar to that found in adolescents and adults can be
istered at age 28, is a 60-item self-report version of the Psychop-
identified in early childhood, then they would advance our under-
athy Checklist—Revised (PCL–R; Hare, 2003). The SRP-II has
standing of the early developmental processes that predispose to
the advantage of a close theoretical association with the PCL–R
adult psychopathy. Conversely, it is possible that the very early
and is designed to assess the same constructs. In a sample of 100
correlates of psychopathy are different to the pattern of findings
prison inmates, Hare (1991) found that the SRP-II and the PCL–R
observed later in life in psychopathic adolescents and adults,
correlate at .54. Furthermore, it has been shown to be a valid

---

510
PRECURSORS OF ADULT PSYCHOPATHIC PERSONALITY
measure of psychopathy in nonforensic, nonclinical populations
which were chosen to reflect the measures of inhibition used by
(K. M. Williams & Paulhus, 2004). In a sample of 289 undergrad-
Kagan et al. (1984). Complete data at this age were available for
uates, K. M. Williams and Paulhus (2004) found the SRP-II to
1,793 participants. Item-total correlations ranged from .19 to .61
correlate .77 with the Psychopathic Personality Inventory (PPI;
(M
0.43). Coefficient alpha for the scale was .72 (Scarpa et al.,
Lilienfeld & Andrews, 1996), another well-validated self-report
1995). Only one rater was used to observe the behaviors of the
measure of psychopathy. Lilienfeld and Andrews (1996) found a
children at age 3, therefore interrater reliability was unable to be
.91 correlation between the PPI and the SRP-II. Zagon and Jackson
assessed. The rater, however, was na?¨ve with respect to any other
(1994) found the SRP-II to correlate positively with narcissism,
variables assessed and to the research hypotheses and, thus, pro-
impulsivity, and dishonesty, and negatively with anxiety and em-
vided an unbiased report of the behaviors.
pathy. K. M. Williams and Paulhus (2004) reported that the SRP-II
Stimulation seeking/sociability and fearfulness–reactivity at age
correlates with narcissism and Machiavellianism and that it cap-
3.
Two relatively independent factors have been shown to un-
tures the interpersonally dark nature of psychopathic persons. They
derlie a collection of eight measures taken from the 3-year-old
also showed that the SRP-II correlates with delinquency even in
assessment battery, that is, stimulation seeking/sociability and
nonforensic samples such as undergraduates. Lilienfeld (1999)
fearfulness–reactivity (see Raine, Reynolds, Venables, Mednick,
also showed that the SRP-II is significantly correlated with the
& Farrington, 1998, for full psychometric details of these two
Minnesota Multiphasic Personality Inventory–2 Psychopathy De-
measures). Briefly, four putative indices of stimulation seeking/
viate subscales. Salekin (2001) also showed that the SRP-II has a
sociability were taken at age 3 years as follows:
high discriminant validity from other personality disorders. On the
1. The child’s exploration away from the mother toward new
basis of this research, the SRP-II appears to be comparable with
toys was assessed in a laboratory room by a research assistant.
other self-report measures of psychopathy.
Exploratory behavior was rated on a 4-point scale as follows: 1
Each item on the SRP-II is scored from 1 (strongly disagree) to
passive, clings to mother, withdrawn; 2
shows interest, exam-
7 (strongly agree). High-scoring (n
56) and low-scoring (n
ines toys but stays close to mother; 3
leaves mother, mild
56) psychopathy groups were created by using a cutoff of one
independent exploration, comes and goes to mother; 4
active
standard deviation above and below the mean on the SRP-II scale.
independent exploration. This behavior was rated on four occa-
The mean and standard deviation of each group is as follows: high
sions during the entire testing session (soon after arrival, before
(M
244.65, SD
10.68), low (M
173.61, SD
8.87). The
psychophysiological testing, between tests, after completion of
mean for all 335 individuals was 207.62 (SD
23.71). To provide
tests). Scores for the four ratings were summed to obtain an overall
a reference for the scores, the mean SRP-II scores for a group of
index of exploration.
community psychopathic persons from the United States as de-
2. Extent of verbalizations to the research assistant during cog-
fined by the PCL–R was 223.20 (SD
27.90; Raine, 2006), which
nitive testing was rated on a 4-point scale, ranging from 1 (very
is lower than the mean for the high-scoring group in this study. The
reluctant to speak) to 4 (many spontaneous comments).
community psychopath grouping has been used in several previ-
3. Friendliness with the research assistant during cognitive test-
ously published studies (Ishikawa, Raine, Lencz, Bihrle, &
ing was rated on a 4-point scale, ranging from 1 (fearful) to 4
Lacasse, 2001; Raine et al., 2004, 2003). In the present study, the
(immediately friendly).
high-scoring group was 69% male and 31% female. The low-
4. Active social play with other children during free play in a
scoring group was 46% male and 54% female. Coefficient alpha
sandbox was rated by a research assistant on a 5-point scale,
for the SRP-II was .85.
including ranges of 1 (solitary), 3 (associates with others), and 5
(cooperative relationship with role reciprocity). These four items
intercorrelated from .25 to .68 (M
0.43). Item-total correlations
Temperament Measures
for this scale ranged from .48 to .59 (M
0.53). Coefficient alpha
Inhibited/disinhibited temperament at age 3.
A measure of
for the scale was .75 (Raine et al., 1998).
general inhibited/disinhibited temperament was developed on this
Four putative indicators of fearfulness–reactivity were assessed
population by Scarpa, Raine, Venables, and Mednick (1995) and
by a research assistant during psychophysiological testing at age 3
uses ratings based on the criteria described by Kagan, Reznick,
years as follow: (a) crying behavior was assessed on a 5-point
Clarke, Snidman, and Garcia-Coll (1984). Briefly, ratings of the
scale, ranging from 1 (no crying) to 5 (uncontrollable crying); (b)
child’s behavior in the laboratory were made by a trained research
the child’s fearful reaction was rated on a 4-point scale, ranging
assistant before and during psychophysiological testing on crying
from 1 (interested) to 4 (very frightened); (c) unresponsiveness to
behavior (1
no crying to 5
cries uncontrollably) and socia-
the experimenter was rated on a 2-point scale, ranging from 1
bility (1
friendly, 2
unresponsive), and by a trained psychol-
(unresponsive) to 2 (friendly); (d) tremor (shaking with fear) was
ogist before and during cognitive testing on crying (1
no crying
rated on a 4-point scale, ranging from 1 (little activity) to 4
to 5
cries uncontrollably), approach–avoidance (1
indepen-
(tremor). These four items intercorrelated from .48 to .72 (M
dent exploration to 5
clings to mother), verbalizations (1
0.56). Item-total correlations ranged from .57 to .76 (M
0.68).
many spontaneous comments to 4
extremely reluctant to speak),
Coefficient alpha for the scale was .84 (Raine et al., 1998). Scores
ease of relationship with tester (1
immediately friendly to 4
for stimulation seeking/sociability and fearfulness–reactivity were
fearful), and social involvement with other children (1
cooper-
calculated by averaging the standard scores (z scores) for the
ative play and exchanges to 5
solitary, away from others). All
respective variables.
variables were scored in the direction of higher scores, reflecting
Confirmatory factor analysis using LISREL 8 (Jo¨reskog &
more inhibited behavior. An inhibition score was calculated by
So¨rbom, 1993) established that stimulation seeking/sociability and
averaging the standard scores (i.e., z scores) for these variables,
fearfulness–reactivity constitute relatively independent tempera-

---

GLENN, RAINE, VENABLES, AND MEDNICK
511
mental factors (see Raine et al., 1998, for full details). The two
Nagelkerke statistic used for variance estimation. Temperament
factors intercorrelated at a level of .04 and, thus, were largely
and autonomic variables were entered by using a stepwise forward
orthogonal. Virtually identical findings were obtained for boys and
procedure (Wald chi-square) with an entry probability of .05 and
girls and for Indian and Creole individuals (Raine et al., 1998).
a removal probability of .10.
The relationship between measures of temperament and psycho-
Autonomic Measures
physiology for the full sample of 1,795 participants at age 3 has
been published previously (Scarpa, Raine, Venables, & Mednick,
Resting levels at age 3.
The electrocardiogram (ECG) was
1997). Although some measures correlate to a limited extent, they
recorded by using Beckman silver/silver chloride electrodes
also may make independent contributions to psychopathy and thus
(Beckman Instruments Inc., Richmond, CA), and Cambridge elec-
are evaluated and discussed independently.
trode gel (Cambridge Instrument Co. Ltd., Cambridge, England).
A Standard Lead I recording configuration was used, with ECG
Results
amplified by using a Grass type 79 polygraph and a 7P5 pream-
plifier. Resting heart rate was recorded during a 1-min rest period
Temperament
preceding the orienting paradigm (described in detail by Venables,
Means and standard deviations of temperament and psycho-
1978). Interbeat intervals were measured for the first artifact-free
physiology measures are listed in Table 1. Those scoring higher on
10 beats in the rest period, and heart rate in beats per minute was
the psychopathy measure relative to those scoring lower on the
calculated from the average of these 10 interbeat intervals.
measure at age 28 were less inhibited at age 3, t(109)
2.424,
Initial SC levels and number of nonspecific SC responses were
p
.017, d
.46 (see Figure 1). They were also less fearful at age
measured for 1 min prior to the onset of the first orienting tone (see
3, t(109)
2.103, p
.038, d
.40. The higher scorers were
below). SC was recorded from bipolar leads on the medial pha-
nonsignificantly higher on stimulation seeking/sociability,
langes of the first and second fingers of the left hand, using a
t(106)
1.633, p
.106, d
.32. To explore this latter trend
constant voltage system (Venables & Christie, 1973). Beckman
further, we conducted analyses on each of the four components of
miniature Ag/AgCl-type (4 mm in diameter) electrodes were filled
stimulation seeking/sociability. The group scoring higher in psy-
with 0.5% KCl in 2% agar–agar as the electrolyte. The number of
chopathy was significantly higher on three of the four components
nonspecific SC responses greater than 0.05
S occurring during
of stimulation seeking/sociability, with significantly more verbal-
the 1-min rest was taken as an indicator of electrodermal arousal.
izations, t(109)
2.157, p
.033, d
.28; social involvement,
Responsivity to orienting and aversive stimuli.
Full details of
t(109)
2.052, p
.043, d
.41; and friendliness toward the
SC measures and stimuli are given in Venables (1978). Partici-
experimenter, t(109)
2.044, p
.043, d
.39, but did not
pants were presented with orienting stimuli that consisted of six
significantly differ on the amount of exploration away from the
neutral pure tones of 75-dB intensity 1 s in duration followed by
mother, t(109)
0.875, p
.384, d
.17. As the three
six aversive stimuli of 90-dB intensity 4.5 s in duration. The
significant components seemed to be most closely related to the
stimuli were presented to the participant binaurally through head-
sociability aspect of the scale, these components were combined
phones while the child was positioned on the mother’s lap. The
into a single index; the higher scoring psychopathy group was
amplitude and SC half-recovery time of each response was re-
found to be significantly more sociable at age 3, t(109)
2.22,
corded. Average SC half-recovery time to aversive stimuli was
p
.029, d
.44. It should be noted that this new sociability
computed based on the averaged SC half-recovery times to the
index has not been tested as an independent factor.
stimuli on which the participant gave an SC response. SC half-
recovery time to orienting stimuli could not be calculated because
Autonomic
of the more substantial number of participants failing to give
orienting responses from which SC half recovery could be calcu-
Arousal.
During the rest period, the higher scoring group gave
lated (Venables, 1978). Interstimulus intervals ranged from 30 s to
more nonspecific SCRs than the lower scorers, t(109)
2.167,
45 s, and all stimuli were 1 s in duration. Responses greater than
p
.032, d
.41. The higher scoring psychopathy group did not
0.05
S occurring within a 1- to 3-s poststimulus window were
differ from the lower scoring group on mean resting heart rate,
scored.
t(104)
0.676, p
.501, or SC levels in the left hand, t(103)
0.300, p
.765, or right hand, t(101)
0.237, p
.813.
Statistical Analyses
Orienting stimuli.
A repeated measures multivariate analysis
of variance revealed a significant Group
Stimulus interaction,
In all cases, t-test comparisons were two-tailed. Effect sizes
F(1, 86)
2.569, p
.045,
2
.029 (see Figure 2). Those
reported are Cohen’s d (Cohen, 1988). Effect sizes of 0.20 are
scoring higher in psychopathy showed higher amplitudes on the
deemed as small, 0.50 as medium, and 0.80 as large (Cohen,
first orienting trial, t (109)
1.992, p
.05, d
.38, but
1
1988). A 2 (higher/lower psychopathy group)
6 (stimulus)
comparisons were nonsignificant on subsequent trials ( p
.207).
repeated measures multivariate analysis of variance was used to
Aversive stimuli.
The higher scoring group showed longer
test the Group
Stimulus interaction for orienting and aversive
average SC half-recovery time to the 90-db stimuli than the lower
stimuli. Partial eta squared is the effect size used in the analysis of
scoring group, t(104)
2.278, p
.025, d
.53. A 2 (higher/
variance tests and describes the proportion of variance accounted
lower psychopathy group)
6 (aversive stimulus) multivariate
for. The ability of measures to independently predict group mem-
repeated measures analysis of amplitudes revealed no main effect
bership was assessed by using logistic regression and the Wald
of group, F(1, 86)
0.879, p
.475; and no significant interac-
chi-square statistic with a classification cutoff of 0.5, and with the
tion between group and stimulus, F(1, 86)
2.303, p
.133.

---

512
PRECURSORS OF ADULT PSYCHOPATHIC PERSONALITY
Table 1
Comparisons (t tests) Between Higher and Lower Scoring Psychopathy Groups for Temperament and Psychophysiology Measures
High psychopathy
Low psychopathy
Variable
M
SD
n
M
SD
n
t
df
p
d
Temperament
Fearful
0.089
0.765
55
0.267
1.00
56
2.103
109
.038
.40
Inhibition
0.106
0.643
55
0.197
0.67
56
2.424
109
.017
.46
Stimulation seeking/sociable
0.077
0.584
52
0.116
0.64
56
1.633
106
.106
.32
Verbal
2.910
1.830
55
2.300
1.03
56
2.157
109
.033
.28
Social
2.760
2.000
55
2.140
1.05
56
2.052
109
.043
.22
Friendly
3.160
1.870
55
2.540
1.32
56
2.044
109
.043
.39
Explore
1.780
0.712
55
1.930
1.02
56
0.875
109
.384
.17
Arousal
Nonspecific SCR
2.800
2.290
55
1.890
2.12
56
2.167
109
.032
.41
Heart rate
124
16.00
55
126
18.10
51
0.676
104
.501
.12
SC level (left)
2.460
1.840
55
2.350
1.85
50
0.300
103
.765
.06
SC level (right)
2.620
1.800
55
2.540
1.55
48
0.237
101
.813
.05
Orienting
Amp—Trial 1
0.137
0.216
56
0.067
0.15
56
1.992
110
.049
.38
Average amplitude
0.084
0.128
56
0.056
0.08
56
1.335
110
.185
.26
Aversive
Half-recovery
4.170
2.510
45
2.960
2.07
34
2.278
77
.025
.53
Average amplitude
0.196
0.213
47
0.133
0.18
51
1.556
96
.123
.48
Note.
All data in boldface type are statistically significant at p
.05. SCR
skin conductance response; SC
skin conductance; Amp
amplitude.
Correlations
no significant interaction effect of gender with psychopathy group-
ing for temperament ( p
.298) or psychophysiology ( p
.197).
Correlations for SRP-II scores and temperament and psycho-
There was also no interaction effect of ethnicity for temperament
physiology measures of the entire sample of 335 participants can
( p
.126) or psychophysiology ( p
.335). Gender, ethnicity,
be found in Table 2. Significant positive correlations were found
socioeconomic status, social adversity, and IQ at age 3 were also
for measures of amplitude to first orienting response and half-
entered as covariates, but results remained significant (Table 3).
recovery time to aversive stimuli. Significant negative correlations
were found for measures of inhibition and fearfulness.
Independent Predictors
Potential Moderating Effects
The results show that the high-scoring psychopathy group was
The comparisons of temperament and autonomic variables were
characterized by low fearfulness, high sociability, increased non-
repeated by using gender and ethnicity as moderators. There was
specific SC responses, higher amplitude to the first orienting
Figure 1.
Mean z scores for measures of fearfulness, stimulation seeking/sociability, and inhibition at age 3 for
the higher scoring (n
53) and lower scoring (n
53) psychopathy groups at age 28.

---

GLENN, RAINE, VENABLES, AND MEDNICK
513
Figure 2.
Age 3 mean amplitudes to the six orienting stimuli for the higher scoring (n
53) and lower scoring
(n
53) psychopathy groups at age 28.
response, and longer half-recovery time to aversive stimuli. Al-
Fearfulness and the amplitude of the first orienting response in-
though inhibition was also significantly lower, it was not included
creased the percentage of variance explained but were only mar-
because of the considerable overlap with fearfulness. To assess
ginally significant. Half-recovery time to aversive stimuli was a
which of these measures independently predicted to group mem-
significant independent predictor of group membership. Taken
bership, we entered each on the second step of a logistic regression
together, the predictors accounted for 21% of the variance in group
after entry of all other variables on the first step (see Table 4).
membership.
Results indicated that nonspecific SC responses and sociability did
not predict group membership over and above all other variables.
Discussion
This study aimed to explore the possible existence of very early
Table 2
temperamental and psychophysiological precursors of adult psy-
Correlations Between Self-Report Psychopathy Scale (SRP-II)
chopathic personality. Individuals with higher psychopathy scores
Scores and Measures of Temperament and Psychophysiology
in adulthood were significantly less fearful and inhibited at age 3
than those with lower psychopathy scores as adults, and they
SRP-II
scored higher on the sociability component of the stimulation-
seeking measure. Similar to findings in psychopathic adults, the
Variable
n
r
p
higher scoring psychopathy group at age 3 had significantly longer
Temperament
SC half-recovery times to aversive stimuli. Contrary to predic-
Inhibition
333
.123
.025
tions, however, the higher scoring group demonstrated signifi-
Fearful
333
.125
.023
cantly increased arousal, as measured by more nonspecific SCRs,
Stimulation seeking/sociable
333
.042
.453
Verbal
333
.106
.054
and increased amplitude (a two-fold increase) to orienting stimuli.
Social
333
.088
.108
Effect sizes were small to moderate in all cases, with values of .40
Friendly
333
.102
.063
for fearfulness, .46 for disinhibition, .44 for sociability, .41 for
Explore
333
.036
.515
nonspecific SCRs, .38 for SC orienting, and .53 for SC half-
Arousal
recovery time. To our knowledge, this is the first report that
Nonspecific SCR
333
.064
.241
Heart rate
318
.059
.292
demonstrates a prospective relationship between characteristics of
SC level (left)
320
.007
.895
preschool children and psychopathic-like personality in adulthood.
SC level (right)
312
.032
.569
These prospective longitudinal findings suggest that at least some
Orienting
biological and temperamental predispositions to psychopathic per-
Average amplitude
334
.084
.125
Amplitude to 1st response
334
.128
.019
sonality may be in place at a young age.
Aversive
The more psychopathic adults were characterized as being no-
Average SC half-recovery
334
.150
.021
ticeably less fearful and inhibited at age 3. Kochanska (1993)
Average amplitude
332
.047
.387
suggested that children’s fearfulness contributes to the develop-
ment of moral emotions such as guilt, shame, and empathy; chil-
Note.
All data in boldface type are statistically significant at p
.05.
SCR
skin conductance response; SC
skin conductance.
dren who are more fearful tend to feel remorse after wrongdoing,

---

514
PRECURSORS OF ADULT PSYCHOPATHIC PERSONALITY
Table 3
Group Differences on Temperament and Psychophysiological Variables After Controlling for Demographic and Cognitive Factors
Inhibition
Fear
nSCRs
Amp. 1st OR
Half-recovery
Covariate
F
p
F
p
F
p
F
p
F
p
Ethnicity
4.092
.046
3.951
.042
4.775
.031
3.967
.049
6.040
.016
Gender
5.752
.018
4.851
.030
4.727
.032
3.561
.042
3.481
.046
SES
6.431
.013
4.526
.036
4.892
.029
4.071
.046
5.516
.021
Adversity
6.646
.011
4.697
.032
4.985
.028
4.080
.046
7.036
.003
IQ total
12.04
.001
5.910
.017
9.007
.004
4.016
.047
3.640
.047
Verbal IQ
11.178
.001
5.483
.022
8.991
.004
3.770
.046
3.700
.044
Spatial IQ
6.896
.010
4.781
.032
10.018
.002
3.652
.049
3.945
.048
Note.
nSCRs
nonspecific skin conductance responses; Amp. 1st OR
amplitude to the first orienting response; SES
socioeconomic status.
are concerned with consequences related to their actions, and are
hypotheses, the authors can only speculate about possible inter-
generally deterred from future wrongdoings by feelings of discom-
pretations, and replication in future research is recommended. One
fort compared with nonfearful children. This study provides initial
potential explanation for these findings is that increased arousal
prospective, longitudinal evidence that young children with low
and orienting may be a factor that distinguishes individuals in this
levels of fearfulness and inhibition are at risk for the subsequent
community sample who have relatively higher levels of psycho-
development of a psychopathic personality in adulthood.
pathic traits from caught and convicted psychopathic individuals.
Stimulation seeking/sociability, which is often thought to be
This sample may be similar to “successful” psychopathic persons
linked to the antisocial behavior features of psychopathy, was
(Ishikawa et al., 2001), who have been found to demonstrate
nonsignificantly higher at age 3 in the higher scoring psychopathy
heightened autonomic stress reactivity versus “unsuccessful” psy-
group at age 28 ( p
.11, d
.32). Statistically significant results
chopathic persons who show reduced reactivity in comparison to
were found for three of the four components of the measure:
controls (Ishikawa et al., 2001). High levels of arousal and orient-
verbalizations, friendliness, and active social play. These three
ing are thought to be indicative of more proficient attentional
components seem to be most related to the sociability aspect of the
processing. Individuals at high risk for an antisocial outcome may
measure, suggesting that being particularly sociable at an early age
be protected from adult antisocial behavior by increased arousal
may be one component of developing a psychopathic-like person-
and orienting or are better able to avoid detection (Raine, Ven-
ality later in life. Sociability in childhood may translate into traits
ables, & Williams, 1995). It should be noted that no data are
such as glibness, superficial charm, conning, and manipulation
available to classify individuals in the higher scoring group as truly
observed in psychopathic adults. What remains unresolved is what
successful psychopaths; the suggestion that these individuals may
other process morphs such positive early sociability into the devi-
be similar to prior work on successful psychopathic persons as
ant psychopathic features of glibness and conning.
defined by Ishikawa et al. (2001) is based on the fact that they were
Consistent with prior studies of individuals with psychopathy
living in the community and were not incarcerated, at the time of
(Lorber, 2004; Raine, 1993), no differences were detected in heart
testing at age 28.
rate or SCLs between the higher and lower scoring psychopathy
Increased SC orienting may also reflect better functioning of the
groups. Contrary to predictions however, the higher scoring group
prefrontal cortex, in contrast to deficits in prefrontal functioning
at age 28 demonstrated significantly higher arousal, as measured
sometimes seen in incarcerated psychopathic persons (LaPierre,
by nonspecific SCRs and higher orienting amplitudes at age 3, two
Braun, & Hodgins, 1995; Mitchell, Colledge, Leonard, & Blair,
measures that tend to be closely associated (Crider, 1993). These
2002). Prior structural and functional brain imaging research has
findings are discrepant with the literature on antisocial and psy-
shown that SC orienting is related to both prefrontal structure
chopathic behavior in adults. As these findings contradict the
(Raine, Reynolds, & Sheard, 1991) and frontal functioning (Haz-
lett, Dawson, Buchsbaum, & Nuechterlein, 1993; L. M. Williams
et al., 2000). Good prefrontal functioning may contribute to some
Table 4
of the more adaptive features of psychopathy, such as glibness,
Test of Whether Temperament and Psychophysiological
superficial charm, lying or conning, and the ability to manipulate
Variables Predict Group Membership
others (Hare, 2003). Yang et al. (2005) found that successful
% of increase
psychopathic persons do not show the reduction in prefrontal gray
Predictors
Wald 2
p
in variance
B weightsa
volume that unsuccessful psychopathic individuals show. Further-
more, Ishikawa et al. (2001) showed that successful psychopaths
Half-recovery time
5.819
.036
8.7
0.261
have significantly better executive functioning than both unsuc-
Amplitude to 1st response
3.428
.064
5.0
2.493
Fearful
3.020
.082
4.4
0.577
cessful psychopathic persons and controls. Increased orienting in
Nonspecific SCRs
0.215
.643
0.3
0.057
the group that scored higher in psychopathy could potentially
Sociability
0.000
.988
0.0
0.004
mean that these individuals lack the impairments in prefrontal
functioning frequently observed in unsuccessful incarcerated psy-
Note.
SCRs
skin conductance responses.
a Beta weights for all variables entered on Step 2 of the logistic regression.
chopathic persons and, thus, may be more skilled at deceiving and

---

GLENN, RAINE, VENABLES, AND MEDNICK
515
manipulating others to avoid negative consequences and detection.
processes that shape psychopathic personality. In conjunction with
Further, the same pattern of orienting observed in this study
other constructs, however, measures of temperament and psycho-
(increased amplitude on the first but not subsequent orienting
physiology may ultimately help to elucidate the etiological basis to
stimuli) has also been found in high sensation seekers (Feij,
psychopathic personality. Half-recovery time was found to predict
Orlebeke, Gazendam, & Van Ziulen, 1985; Neary & Zuckerman,
psychopathy group membership independent of all other predic-
1976; Robinson & Zahn, 1983), who have been found to be
tors, suggesting that it is not confounded by other temperament and
significantly more attentive to novel stimuli (Zuckerman, 1994).
psychophysiological variables. A relationship between long half-
This heightened awareness of environmental cues may be benefi-
recovery time and psychopathy has been found in all but one prior
cial, especially in situations involving impending punishment.
study and is surprisingly robust (Raine, 1993). There was also
However, additional research is needed to explore the possible
more limited evidence that fearfulness and amplitude of the first
causes and implications of increased arousal and orienting.
orienting response also independently contribute to the prediction
The finding of longer SC half-recovery time in children who
of group membership, suggesting that they too may be factors in
score higher in psychopathy as adults is consistent with the hy-
the development of psychopathy and are worthy of further inves-
pothesized association between long-SC half-recovery time and
tigation in longitudinal studies.
antisocial behavior (Venables, 1975). In animal studies, long-SC
An alternative interpretation of the overall results of this study,
half-recovery time has been associated with lesions in the amyg-
which should also be considered, is that the expected precursors of
dala (Pribram & McGuiness, 1975), a region thought to be com-
adult psychopathy do not clearly exist as early as age 3. The
promised in psychopathic individuals (Blair, 2004; Patrick, 1994).
strongest predictor of adult psychopathic traits is SC half-recovery
The amygdala is important in processing cues of threat or harm
time: Although this is a characteristic of adult psychopathic indi-
(LeDoux, 1995; Morris et al., 1996) and in fear conditioning
viduals (Hare, 1978), some have expressed reservations about the
(Davis, 2000; Knight, Nguyen, & Bandettini, 2005; Maren, 2001).
interpretation of this correlate (Fowles, 1993). The significant
Long half-recovery time has also recently been associated with
findings on orienting and arousal contradict some of the most
low levels of harm avoidance (Mardaga, Laloyaux, & Hansenne,
consistent findings in adult psychopathy, whereas findings on SC
2006), which is conceptualized as reflecting fearfulness of physical
levels were nonsignificant. In addition, temperament findings can
danger (Tellegen, 1982) and has been shown to be negatively
be taken to indicate that the lower scoring group was inhibited and
correlated with psychopathy (Benning, Patrick, Hicks, Blonigen,
fearful, rather than the higher scoring group being fearless and
& Krueger, 2003; Levenson, Kiehl, & Fitzpatrick, 1995). In addi-
disinhibited (see Figure 1). Budding psychopathic individuals in-
tion, individuals with low levels of harm avoidance demonstrated
stead may simply be gregarious as children, inconsistent with the
relatively little amygdala activity in response to irrelevant emo-
characterization of psychopathic adults as unfriendly. Alterna-
tional distractors (Most, Chun, Johnson, & Kiehl, 2006). The
tively, we believe from clinical experience (Cleckley, 1941) that
finding of long-SC half-recovery time at age 3 in adults scoring
psychopathic persons frequently present as superficially friendly
higher in psychopathy could therefore be an indicator of amygdala
and gregarious, not as unfriendly. The contradictory findings of
dysfunction at an early age, resulting in a failure to learn to avoid
high arousal and orienting would be consistent with prior findings
harm or punishment. Future prospective longitudinal imaging stud-
on successful psychopathic persons (see above), or alternatively
ies could test this initial hypothesis.
could be specific to the self-report outcome measure of psychop-
An alternative position that should be considered is that SC
athy, which has limitations; findings more theoretically consistent
half-recovery time is an artifact. Fowles (1993) has argued that
with the adult literature could have emerged by using an interview-
long-SC half-recovery time simply reflects reduced prior electro-
based measure of psychopathy. Despite these alternative perspec-
dermal activity, although Dawson, Schell, and Filion (2000) and
tives, results are the first of their kind and may truly represent
Raine, Venables, and Williams (1996) have argued that this issue
important processes in the development of psychopathy; future
remains unsettled. Venables and Fletcher (1981) argued that it is
prospective longitudinal research is required to replicate and ex-
still worthwhile to measure SC half-recovery time as an indepen-
tend these preliminary findings before firm conclusions can be
dent variable as there is insufficient evidence to consider it redun-
drawn.
dant. If long-recovery time to aversive stimuli was a function of
Several limitations of this study should be recognized. Findings
reduced prior electrodermal activity, then one would predict re-
do not inform us of the early antecedents of psychopathy in caught
duced amplitudes to aversive stimuli in the higher scorers. No such
and convicted offenders; conclusions can only be applied at this
effects were observed, and indeed the higher scoring psychopathy
point in time to community samples with psychopathic personality.
group demonstrated nonsignificantly higher amplitudes to the
Although the self-report measure of psychopathy used correlates
aversive stimuli. The correlation between the amplitude to aversive
quite highly with the “gold standard” (PCL–R) measure of insti-
stimuli and half-recovery time was .16 ( p
.013). As such,
tutionalized psychopathy (Hare, 2003), those scoring higher on the
although the counterexplanation of Fowles (1993) may ultimately
psychopathy measure are not necessarily psychopathic but instead
be correct and would help to resolve discrepant results in the
constitute individuals with a psychopathic personality. Results
current study, it cannot easily be invoked in the context of these
should be extended in future studies with a more objective measure
specific findings.
of psychopathy. However, Hare’s SRP-II appears to be a reason-
The logistic regression indicated that 21% of the variance in
ably good correlate of psychopathy in noninstitutionalized popu-
psychopathy grouping could be explained by temperament and
lations (Forth, Brown, Hart, & Hare, 1996; Hare, 1991; Paulhus &
autonomic variables. Conversely, 79% of the variance remains
Williams, 2002; Zagon & Jackson, 1994). A more stringent study
unaccounted. This clearly illustrates that temperament and auto-
with appropriate corrections for Type I error would also render
nomic functioning represent only two of the likely multiple early
several of the significant effects nonsignificant but at the same

---

516
PRECURSORS OF ADULT PSYCHOPATHIC PERSONALITY
time would run the risk of Type II errors and misleading null
Davis, M. (2000). The role of the amygdala in conditioned and uncondi-
conclusions, which are particularly serious in initial, preliminary
tioned fear and anxiety. In J. P. Aggleton (Ed.), The amygdala: A
studies. Finally, no other psychobiological or behavioral measures
functional analysis (pp. 213–287). New York: Oxford University.
were taken at age 28, so we could not determine whether partici-
Dawson, M. E., Schell, A., & Filion, D. (1990). The electrodermal system.
pants scoring high on the SRP-II display other factors often ob-
In J. T. Cacioppo & L. G. Tassinary (Eds.), Principles of psychophysi-
ology (pp. 295–324). Cambridge, England: Cambridge University Press.
served in psychopathic behavior.
Dawson, M. E., Schell, A., & Filion, D. (2000). The electrodermal system.
The prospective longitudinal design is thought to be a strength
In J. T. Cacioppo, L. G. Tassinary, & G. G. Berntson (Eds.), Handbook
of the study in that it helps to begin the process of establishing the
of psychophysiology (pp. 200 –223). Cambridge, England: Cambridge
early developmental precursors of psychopathy, as opposed to the
University Press.
more common examination of the cross-sectional correlates of this
Feij, J. A., Orlebeke, J. F., Gazendam, A., & Van Ziulen, R. W. (1985).
condition. Furthermore, this study eliminates the problems asso-
Sensation seeking: Measurement and psychophysiological correlates. In
ciated with retrospective data and provides relatively unique psy-
J. Strelau, F. H. Farley, & A. Gale (Eds.), The biological bases of
chophysiological knowledge of early autonomic functioning that
personality and behavior: Theories, measurement techniques, and de-
can never be determined retrospectively.
velopment (Vol. 1, pp. 195–210). New York: Hemisphere/Harper &
In conclusion, this study suggests that some indicators of adult
Row.
psychopathic personality may originate and be observable very
Forth, A. E., Brown, S. L., Hart, S. D., & Hare, R. D. (1996). The
early in life. Individuals who were higher in psychopathy at age 28
assessment of psychopathy in male and female noncriminals: Reliability
and validity. Personality & Individual Differences, 20, 531–543.
were characterized by a less fearful and inhibited temperament at
Fowles, D. C. (1993). Electrodermal activity and antisocial behavior:
age 3, and they were also more sociable. Psychophysiologically,
Empirical findings and theoretical issues. In J. C. Roy, W. Boucsein,
they showed longer SC half-recovery times, similar to findings in
D. C. Fowles, & J. H. Gruzelier (Eds.), Progress in electrodermal
psychopathic adults. However, contrary to previous SC findings in
research (pp. 223–237). New York: Plenum Press.
psychopathic adults, this group of community individuals showed
Fowles, D. C., Kochanska, G., & Murray, K. (2000). Electrodermal activity
increased autonomic arousal and electrodermal orienting. This
and temperament in preschool children. Psychophysiology, 37, 777–787.
raises the possibility that (a) high arousal and orienting may reflect
Frick, P. J. (1998). Callous-unemotional traits and conduct problems:
increased attentional processing, which may serve to protect them
Applying the two-factor model of psychopathy to children. In D. J.
from being caught and convicted, despite their psychopathic per-
Cooke, A. E. Forth, & R. D. Hare (Eds.), Psychopathy: Theory, research
sonality; and (b) increased arousal and orienting may reflect good
and implications for society (pp. 161–187). Dordrecht, the Netherlands:
prefrontal functioning, which allows for the more adaptive features
Kluwer Academic Publishing.
Frick, P. J., Cornell, A. H., Bodin, S. D., Dane, H. E., Barry, C. T., &
of psychopathy such as the ability to be smooth, engaging, and
Loney, B. R. (2003). Callous-unemotional traits and developmental
manipulative. Findings of this study must be treated as initial and
pathways to severe conduct problems. Developmental Psychology, 39,
provisional, but nevertheless they could have implications for
372–378.
furthering our understanding of the development of psychopathy,
Frick, P. J., Lilienfeld, S. O., Ellis, M. L., Loney, B. R., & Silverthorn, P.
a first step toward intervention and prevention.
(1999). The association between anxiety and psychopathy dimensions in
children. Journal of Abnormal Child Psychology, 27, 383–392.
References
Frick, P. J., O’Brien, B. S., Wooton, J. M., & McBurnett, K. (1994).
Aniskiewicz, A. S. (1979). Autonomic components of vicarious condition-
Psychopathy and conduct problems in children. Journal of Abnormal
ing and psychopathy. Journal of Clinical Psychology, 35, 60 – 67.
Psychology, 103, 700 –707.
Benning, S. D., Patrick, C. J., Hicks, B. M., Blonigen, D. M., & Krueger,
Fung, M. T., Raine, A., Loeber, R., Lynam, D. R., Steinhauer, S. R.,
R. F. (2003). Factor structure of the Psychopathic Personality Inventory:
Venables, P. H., et al. (2005). Reduced electrodermal activity in
Validity and implications for clinical assessment. Psychological Assess-
psychopathy-prone adolescents. Journal of Abnormal Psychology, 114,
ment, 15, 340 –350.
187–196.
Blackburn, R. (1979). Cortical and autonomic response arousal in primary
Hare, R. D. (1968). Psychopathy, autonomic functioning, and the orienting
and secondary psychopaths. Psychophysiology, 16, 143–150.
response. Journal of Abnormal Psychology Monograph Supplement,
Blair, R. J. R. (1997). Moral reasoning in the child with psychopathic
73(3), 1–24.
tendencies. Personality & Individual Differences, 22, 731–739.
Hare, R. D. (1978). Electrodermal and cardiovascular correlates of psy-
Blair, R. J. R. (1999). Responsiveness to distress cues in children with
chopathy. In R. D. Hare & D. Schalling (Eds.), Psychopathic behavior:
psychopathic tendencies. Personality & Individual Differences, 27, 135–
Approaches to research (pp. 107–144). New York: Wiley.
145.
Hare, R. D. (1985). Comparison of procedures for the assessment of
Blair, R. J. (2004). The roles of the orbital frontal cortex in the modulation
psychopathy. Journal of Consulting and Clinical Psychology, 53, 7–16.
of antisocial behavior. Brain and Cognition, 55, 198 –208.
Hare, R. D. (1991). Manual for the Hare Psychopathy Checklist—Revised.
Blair, R. J. R., Colledge, E., Murray, L., & Mitchell, D. G. V. (2001). A
Toronto, Ontario, Canada: Multi-Health Systems.
selective impairment in the processing of sad and fearful facial expres-
Hare, R. D. (1998). Psychopathy, affect and behavior. In D. J. Cooke, A. E.
sions in children with psychopathic tendencies. Journal of Abnormal
Forth, & R. D. Hare (Eds.), Psychopathy: Theory, research and impli-
Child Psychology, 29, 491– 498.
cations for society (pp. 105–137). Dordrecht, the Netherlands: Kluwer
Cleckley, H. (1941). The mask of sanity. St. Louis, MO: Mosby.
Academic Publishing.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences
Hare, R. D. (2003). Hare Psychopathy Checklist—Revised (PCL–R; 2nd
(2nd ed.). Hillsdale, NJ: Erlbaum.
ed.). Toronto, Ontario, Canada: Multi-Health Systems.
Crider, A. (1993). Electrodermal response lability–stability: Individual
Hazlett, E., Dawson, M. E., Buchsbaum, M. S., & Nuechterlein, K. (1993).
difference correlates. In J. C. Roy, W. Boucsein, D. C. Fowles, & J. H.
Reduced regional brain glucose metabolism assessed by PET in electro-
Gruzelier (Eds.), Progress in electrodermal research (pp. 173–186).
dermal nonresponder schizophrenics: A pilot study. Journal of Abnor-
New York: Plenum Press.
mal Psychology, 102, 39 – 46.

---

GLENN, RAINE, VENABLES, AND MEDNICK
517
Ishikawa, S. S., Raine, A., Lencz, T., Bihrle, S., & Lacasse, L. (2001).
Patrick, C. J. (1994). Emotion and psychopathy: Startling new insights.
Autonomic stress reactivity and executive functions in successful and
Psychophysiology, 31, 319 –330.
unsuccessful criminal psychopaths from the community. Journal of
Paulhus, D. L., & Williams, K. M. (2002). The dark triad of personality:
Abnormal Psychology, 110, 423– 432.
Narcissism, Machiavellianism, and psychopathy. Journal of Research in
Jo¨reskog, K. G., & So¨rbom, D. (1993). LISREL 8 (Version 8.03) [Com-
Personality, 36, 556 –563.
puter software]. Chicago: Scientific Software International.
Pribram, K. H., & McGuiness, D. (1975). Arousal, activation, and effort in
Kagan, J., Reznick, J. S., Clarke, C., Snidman, N., & Garcia-Coll, C.
the control of attention. Psychological Review, 82, 116 –149.
(1984). Behavioral inhibition to the unfamiliar. Child Development, 55,
Raine, A. (1993). The psychopathology of crime: Criminal behavior as a
2212–2225.
clinical disorder. San Diego, CA: Academic Press.
Kimonis, E. R., Frick, P. J., Fazekas, H., & Loney, B. R. (2006). Psychop-
Raine, A. (2006). [SRP-II scores in community psychopaths compared to
athy, aggression, and the processing of emotional stimuli in non-referred
controls]. Unpublished raw data.
boys and girls. Behavioral Sciences & the Law, 24, 21–37.
Raine, A., Ishikawa, S. S., Arce, E., Lencz, T., Knuth, K. H., Bihrle, S., et
Knight, D. C., Nguyen, H. T., & Bandettini, P. A. (2005). The role of the
al. (2004). Hippocampal structural asymmetry in unsuccessful psycho-
human amygdala in the production of conditioned fear responses. Neu-
paths. Biological Psychiatry, 55, 185–191.
roimage, 26, 1193–1200.
Raine, A., Lencz, T., Taylor, K., Hellige, J. B., Bihrle, S., Lacasse, L., et
Kochanska, G. (1993). Toward a synthesis of parental socialization and
al. (2003). Corpus callosum abnormalities in psychopathic antisocial
child temperament in early development of conscience. Child Develop-
individuals. Archives of General Psychiatry, 60, 1134 –1142.
ment, 64, 325–347.
Raine, A., Reynolds, C., Venables, P. H., Mednick, S. A., & Farrington,
LaPierre, D., Braun, C. M. J., & Hodgins, S. (1995). Ventral frontal deficits
D. P. (1998). Fearlessness, stimulation-seeking, and large body size at
in psychopathy: Neuropsychological test findings. Neuropsychologica,
age 3 years as early predispositions to childhood aggression at age 11
33, 139 –151.
yea