Journal of Abnormal Child Psychology, Vol. 29, No. 6, December 2001, pp. 499–511 (C ?2001)
Somatic Markers and Response Reversal: Is There
Orbitofrontal Cortex Dysfunction in Boys
With Psychopathic Tendencies?
R. J. R. Blair,1,2,3E. Colledge,1,2and D. G. V. Mitchell1,2
Received December 7, 2000; revision received June 25, 2001; accepted June 25, 2001
This study investigated the performance of boys with psychopathic tendencies and comparison boys,
aged 9 to 17 years, on two tasks believed to be sensitive to amygdala and orbitofrontal cortex func-
tioning. Fifty-one boys were divided into two groups according to the Psychopathy Screening Device
(PSD, P. J. Frick & R. D. Hare, in press) and presented with two tasks. The tasks were the gambling
task (A. Bechara, A. R. Damasio, H. Damasio, & S. W. Anderson, 1994) and the Intradimensional/
Extradimensional (ID/ED) shift task (R. Dias, T. W. Robbins, & A. C. Roberts, 1996). The boys with
psychopathic tendencies showed impaired performance on the gambling task. However, there were
no group differences on the ID/ED task either for response reversal or extradimensional set shifting.
The implications of these results for models of psychopathy are discussed.
KEY WORDS: amygdala; orbitofrontal cortex; psychopathic tendencies; psychopathy.
Psychopathy is a disorder characterized in part by
callousness, a diminished capacity for remorse, superfi-
cial charm, proneness to boredom, and poor behavioral
inals commit a disproportionate amount of crime, habitu-
ally fail to fulfill societal obligations, appear to lack any
sense of loyalty, and are unperturbed when confronted
with the destructive nature of their behavior (Cleckley,
1967; Hare, 1991). It is thought that psychopathy can
be indexed behaviorally in children using the Psychopa-
thy Screening Device (PSD; Frick & Hare, in press) and
in adults by using the Revised Psychopathy Checklist
(PCL-R; Hare, 1991). Both scales index a similar syn-
drome. Factor analysis of both the PSD and PCL-R reveal
lems (I/CP) factor that comprises overt behavioral char-
acteristics such as impulsivity, poor impulse control (e.g.,
1Institute of Cognitive Neuroscience, University College London,
2Department of Psychology, University College London, London,
3Address all correspondence to James Blair, Institute of Cognitive
Square, London, WC1H 3AR, England.
becomes angry when corrected), and delinquent behav-
ior (Frick, O’Brien, Wootton, & McBurnett, 1994; Hare,
1991). As regards the PSD, this factor is highly correlated
with traditional measures of conduct problems, such as
the DSM-IV definition of Conduct Disorder (Frick, 1995).
Secondly, the Callous/Unemotional (C/UN) factor cap-
and superficial charm, which are considered primary in
the clinical description of psychopathy (Cleckley, 1967).
However, despite the similarities between the PSD and
PCL-R it is premature to suggest the two are isomor-
phic. There are some content differences. For example,
some PCL-R items have no PSD counterparts (parasitic
lifestyle and lack of realistic, long-term goals). Likewise
some PSD items have no PCL-R counterparts (concerned
about schoolwork, keeps the same friends, teases other
people). However, it is important to note that the neuro-
cognitive impairments that have been found in adult psy-
chopathic individuals are also being found in children
1998; O’Brien & Frick, 1996).
Currently, three loci of impairment at the anatomi-
cal level have been suggested as causes of psychopathy.
These are the septo-hippocampal system (Gorenstein &
0091-0627/01/1200-0499$19.50/0 C ?2001 Plenum Publishing Corporation
500Blair, Colledge, and Mitchell
Morris, Frith, Perrett, & Dolan, 1999; Patrick, 1994); and
the orbitofrontal cortex (Damasio, 1994; LaPierre, Braun,
& Hodgins, 1995). The septo-hippocampal position was
prompted by Gray’s Behavioral Inhibition System theory
lesions have been found to induce impairments in pas-
Gorenstein & Newman, 1980; Gray, 1971). However, the
mechanisms thought to be impaired by these lesions are
bition System crucial for emotional learning (Gray, 1971)
but rather a system that is crucial for the representation of
environment; LeDoux, 1995, 1998; O’Keefe, 1991). The
avoidance and aversive conditioning, but such represen-
tations are not necessary for emotional learning to occur
resent a particular environment in order to learn to avoid
of septo-hippocampal neurons. However, if the individual
has to learn which lever to avoid pressing in an environ-
ment, a representation of that environment will not aid
As regards the amygdala position, there are striking
similarities between the performance of psychopathic in-
nitive measures. Thus, psychopathic individuals show re-
duced potentiation of the startle reflex by visual threat
primes (Patrick, Bradley, & Lang, 1993) and impairments
in aversive conditioning (Hare, 1998; Lykken, 1957,
1995). Patients with amygdala lesions also show both
these impairments (Angrilli et al., 1996; Bechara et al.,
1995; LaBar, LeDoux, Spencer, & Phelps, 1995). More-
over, psychopathic individuals, like most patients with
amygdala lesions, show impairments in processing fear-
ful expressions (Adolphs et al., 1999; Blair, Colledge,
Murray, & Mitchell, 2001; Blair & Coles, 2000; Fine &
Blair, 2000; Stevens, Charman & Blair, 2001).
As regards the orbitofrontal cortex position, refer-
tween patients with orbitofrontal cortex lesions and
psychopathic individuals (Blumer & Benson, 1975;
Damasio, 1994). Thus, for example, both groups show
heightened risk for aggression (Grafman, Schwab,
Warden, Pridgen, & Brown, 1996; Hare & Jutai, 1983).
However, care must be taken when considering this ap-
show primarily reactive aggression; their aggression is a
consequence of frustration or perceived threat (Anderson,
Bechara, Damasio, Tranel, & Damasio, 1999; Blair &
Cipolotti, 2000; Grafman et al., 1996). In contrast, the
aggression shown by psychopathic individuals is instru-
reward or increased social respect (Cornell et al., 1996;
Williamson, Hare, & Wong, 1987).
three partially dissociable cognitive systems: the social
response reversal system (Blair & Cipolotti, 2000), the
somatic marker system (Damasio, 1994) and a nonsocial
response reversal system (Rolls, 1997). The social
response reversal system is conceptualized as a system
that is activated by another’s angry expressions. In addi-
tion, it is also activated by representations of situations
that have been previously associated with another indi-
vidual’s angry responses or other negative valence ex-
pressions (e.g., the staring expressions of others that can
precede a sense of embarrassment, and perhaps others’
might be expected. The suggestion is that activation of
this system results in the modulation of current behav-
ioral responding, in particular the modulation of reactive
aggression mediated by the hypothalamus and periaque-
ductal gray (and perhaps sexual behavior mediated by re-
lated systems). In line with this position, neuroimaging
findings have indeed shown the involvement of right or-
bitofrontal cortex in processing angry, but not sad, facial
expressions (Blair et al., 1999). Moreover, patients who
present with reactive aggression following orbitofrontal
cortex lesions have been found to be impaired in process-
ing angry expressions and judging the appropriateness of
behaviors in particular social contexts (Blair & Cipolotti,
2000). However, there are no indications that the social
response reversal system is impaired in psychopathic in-
dividuals. Although adult psychopathic individuals and
children with psychopathic tendencies show impairment
in the processing of fearful facial expressions, they show
no selective difficulties in processing angry expressions
(Blair et al., 2001; Blair & Coles, 2000; Fine & Blair,
2000; Stevens et al., 2001). Moreover, although adult
psychopathic individuals and children with psychopathic
tendencies show impairment in distinguishing moral and
haviors that result in another’s anger (Blair & Cipolotti,
2000). Regarding the somatic marker hypothesis, accord-
ing to Damasio and colleagues, the ventromedial frontal
cortex (orbitofrontal and medial frontal cortex) acts as
a repository, and is involved in the formation of recorded
dispositional linkages between factual knowledge and
Damasio, 1994). When individuals are faced with a
Somatic Markers and Response Reversal501
situation for which some factual aspects have been pre-
viously categorized, the dispositional linkages are acti-
vated. This allows the individual to make appropriate
decisions. The dispositional linkages are activated in ei-
ther of two ways. They can be activated via a “body
loop” in which a “somatic marker” (increased autonomic
arousal) is conveyed to somatosensory cortices. Alterna-
tively, this can occur via an “as-if body loop” in which
the body is bypassed and reaction signals are conveyed
to the somatosensory structures (i.e., no increased auto-
nomic activity can be detected). Under either loop, the
somatosensory structures then adopt an appropriate pat-
tern that constrains option–outcome reasoning. In short,
the somatosensory pattern marks the scenario as either
good or bad, allowing the rapid rejection/endorsement
of specific option–outcome pairs. In other words, the so-
matic marker indicates whether a particular choice would
be advantageous or not. If there is damage to the so-
matic marker system, there will be no somatic marker
to guide behavior. In line with this position, individuals
with “acquired sociopathy” following orbitofrontal cor-
tex lesions are less likely to show autonomic responses
to visually presented social stimuli (scenes of social dis-
aster, mutilation, and nudity; see, for a review, Bechara
et al., 2000). In contrast, they are able to generate condi-
tioned autonomic responses to visual stimuli paired with
an aversive loud sound (Bechara, Damasio, Damasio, &
Lee, 1999). In addition, such patients, unlike controls,
are less likely to shift their behavior away from packs
of cards associated with high risk (Bechara et al., 2000).
ment of the somatic marker system might underlie devel-
& Damasio, 1990). However, it should be noted that psy-
responses to visual threatening images although they are
less responsive to the sad and fearful expressions of oth-
Smith, 1997; Patrick et al., 1993). Yet, psychopathic indi-
viduals are less likely to generate conditioned autonomic
responses to conditioned stimuli (e.g., Lykken, 1957). In
addition, in the first investigation of the performance of
psychopathic adults on Bechara’s four pack card play-
ing task, psychopathic individuals performed similarly to
comparison individuals (Schmitt, Brinkley, & Newman,
1999). However, it should be noted that the task instruc-
tions used by Schmitt et al. (1999) differed from those
of Bechara et al. (2000). Indeed, in a second study, us-
ing the revised instructions provided by Bechara, adult
psychopathic individuals did perform more poorly than
comparison individuals (Mitchell, Colledge, Leonard, &
Blair, Submitted, 2001).
As regards the response reversal system, there are
considerable animal data suggesting that the orbitofron-
tal cortex is involved in altering previously acquired
stimulus–reward associations when they are no longer
and others have demonstrated that patients with OFC le-
of the stimuli (i.e., when the reward values of the stimuli
lesions (Rahman, Sahakian, Hodges, Rogers, & Robbins,
ies have found indications of a response reversal impair-
ment in adult psychopathic individuals (LaPierre et al.,
1995; Mitchell et al., Submitted, 2001). LaPierre et al.
(1995) found that psychopathic individuals performed
more poorly on a Go/No-Go task where they had initially
ulus and then had to reverse this response. The psycho-
pathic individuals were significantly more likely than the
ulus. Mitchell et al. (Submitted, 2001) explored perfor-
mance of adult psychopathic individuals and controls on
the Intradimensional/Extradimensional Discrimination
(ID/ED) task. In the ID/ED task (described more fully
below), the participant is taught to respond to one of two
stimuli and then, having reached criterion, must reverse
this response so that they respond to the other stimulus
(Dias et al., 1996). The psychopathic individuals were
significantly impaired in response reversal on this task.
gate orbitofrontal cortex functioning in children with psy-
chopathic tendencies. Two tasks will be used (see Method
of Bechara et al. (1994). This task is interesting because
patients with both amygdala and orbitofrontal cortex, but
not dorsolateral prefrontal cortex, lesions show impair-
Tranel, & Anderson, 1998). Thus, both the amygdala and
orbitofrontal cortex positions would predict that children
with psychopathic tendencies should show impairment
on this task. The second task is the Intradimensional/
Extradimensional Discrimination (ID/ED) task. This task
is interesting because it assesses two dissociable abilities.
First, it assesses the ability to perform response rever-
sals; the participant is initially rewarded for one behav-
ioral choice and then must reverse his/her behavior when
this stimulus is no longer associated with reward. Second,
it assesses the ability of the participant to perform extra-
dimensional shifts; to shift their response set from one
stimulus property to another (for example, shifting atten-
502 Blair, Colledge, and Mitchell
with impairments in extradimensional shift learning, but
frontal cortex produce impairments in the reversal learn-
ing, but not in the extradimensional shift learning (Dias
et al., 1996). Although functional imaging work has in-
dicated that the reversal learning component of this task
is associated with amygdala activation (Rogers, personal
communication), temporal lobe lesions, including lesions
of the hippocampus, do not result in impairment on any
& Robbins, 1991). Recent work also indicates that these
different forms of shift learning depend on distinct and
dissociable neurochemical substrates (Dias et al., 1996;
Rahman et al., 1999; Rogers et al., 1999). Although the
amygdala position would predict no impairment on this
task, it would be expected that if there was generalised
dencies should show impairment on this task.
Thus, in summary, the amygdala position predicts
impairment on the four-pack card playing task but not the
ID/ED task. In contrast, the orbitofrontal cortex position
predicts impairment on the four pack card playing task
and a reversal learning impairment on the ID/ED task.
This study tests these predictions.
The participants were recruited from three schools
for boys with emotional and behavioral difficulties.
Table I. Participant Characteristics
The gambling taskThe ID/ED task
PT (N = 20)
group (N = 23)
PT (N = 19)
group (N = 23)
[26 to 37]
[6 to 11]
[13 to 20]
[9–8 to 16–0 yrs]
[68 to 108]
[0 to 18]
[0 to 6]
[0 to 9]
[10–5 to 13–9 yrs]
[76 to 116]
[26 to 37]
[5 to 11]
[13.5 to 20]
[9–9 to 17–1 yrs]
[68 to 96]
[0 to 18]
[0 to 6]
[0 to 10]
[10–4 to 14–0 yrs]
[77 to 116]
Note. PSD: Psychopathy Screening Device; C/U: The callous and unemotional component of
psychopathy (e.g., a lack of guilt); I/CP: the impulsive and conduct problems component of
psychopathy (e.g., engaging in crime); IQ: Intelligence quotient; N: Number of participants.
Standard deviations are in parentheses and ranges in square brackets.
∗∗p < 0.001.
All of these children had received statements under
the Education Act of 1993 as being too problematic for
mainstream education. In addition, there was a further
pool of children attending a comprehensive school in an
area with high unemployment who all also have rec-
eived statements under the Education Act of 1993. Con-
sent was obtained from all the parents of the children
who were approached regarding this work. The child-
ren were also free to withdraw from the study at any
thy Screening Device. Almost all of the participants were
from homes of low socioeconomic status. In line with
previous work (Blair, 1997, 1999; Fisher & Blair, 1998),
participants with a PSD score above 25 formed the psy-
below 20 formed the comparison group. Fifty-one boys
meeting the inclusion criteria participated in the study. It
was not possible to test the same sample on both tasks be-
cause of children leaving school during the course of the
sion. Thus, 32 boys received both tasks, 11 boys received
only the four-pack card playing task, whereas 10 boys re-
ceived only the ID/ED task; see Table I for full participant
details divided by task.
The British Picture Vocabulary Scale (Dunn,
Wheklan, & Pintillie, 1982) was administered to provide
an estimate of intelligence. The groups did not differ sig-
nificantly in either age or estimated verbal IQ. The sam-
ple was made up of 50 Caucasian and 1 Afro-Caribbean
Somatic Markers and Response Reversal503
Psychopathy Screening Device
The PSD consists of 20 behavioral items presented
in a questionnaire format and scored by two teachers, or a
tence of two highly interrelated behavioral factors (Frick
responds to affective and interpersonal traits whereas the
Impulsivity/Conduct problems (I/CP) factor is defined by
behaviors characteristic of conduct disorder (Frick, 1995;
Frick et al., 1994). The C/UN factor items on the PSD in-
clude such characteristics as “acts charming in ways that
seem insincere,” “emotions seem shallow,” “does not feel
bad or guilty,” and “is not concerned about the feelings
of others,” whereas the I/CP factor includes such items
as “gets bored easily,” “becomes angry when corrected,”
“acts without thinking,” and “engages in risky or danger-
ous activities” (Frick & Hare, in press).
Participants were rated by two teachers, or in the
case of boarding students by a teacher and a residential
social worker. Pearson correlation of the ratings of the
two raters were 0.699 for total PSD score, 0.582 for the
are significant at the 0.01 level (two-tailed). Participants
received the average score for each item of the two raters.
The Gambling Task (Bechara et al., 1994, 1999)
ized format with a schedule of reinforcement as described
in Bechara et al. (1999). The participant saw four decks
of cards on a computer screen. At the top of each of these
decks were the labels A, B, C, and D. Above these decks
of money won or lost by the participant. A gain was in-
dicated by a proportionate increase in the length of the
green bar, and a loss was indicated by a proportionate
decrease in the bar length. The participant clicked on a
card from any of the four decks by using a mouse. The
computer tracked the sequence of the cards selected from
by clicking on a deck, the computer generated a distinct
sound (similar to a casino slot machine). The face of the
card appeared on top of the deck (the color was either red
or black), and a message was displayed on the screen in-
dicating the amount of money the subject had won or lost.
Once the money was added or subtracted, the face of the
card disappeared, and the participant could select another
card. The intertrial interval between making two consec-
utive card selections was set at 1 s. Each participant had
100 trials (card selections). After these trials, the program
shut off automatically. The participant was not informed
in advance how many trials there would be.
On the screen, the backs of the cards appeared iden-
tical, like real decks of cards. Whenever the participant
picked from Packs A or B the computer displayed a $100
reward and whenever the participant picked from Packs C
or D the computer displayed a $50 reward. The sequence
of gains and losses for each card selection were based on
the original version of this task (Bechara et al., 1994).
When the schedule indicated that a card choice would
be punished, the computer displayed a message: “...You
have won X dollars, but you also have lost Y dollars...”
the square), and the net loss was reflected automatically
on the green bar on the screen. In brief, every 10 cards
from Deck A over the course of trials gained $1000, but
$150 to $350, that brought the total loss to $1250. Every
10 cards from Deck B gained $1000, but there was also
one big punishment for $1250. On the other hand, every
10 cards from Deck C or D only gained a total amount
of $500, but the losses were also smaller, that is, $250
(ranging from $25 to $75 in Deck C and one $250 loss in
Deck D), bringing a net gain of $250. In summary, decks
A and B were equivalent in terms of overall net loss over
trials. In a similar way, Decks C and D were equivalent
in terms of overall net gains. However, Decks A and C
had a higher frequency, but a lower magnitude, of punish-
ment. In contrast, Decks B and D had a lower frequency,
but a higher magnitude, of punishment. Thus, the partic-
ipant should learn to avoid Decks A and B because they
cost more in the long run and, instead, learn to respond to
Decks C and D because they resulted in an overall gain
in the long run. Full participant instructions are given in
Bechara et al. (1999).
To score the performance of the subject on the gam-
bling task, we added the number of cards picked from
Decks A and B in each block of 10 cards.
The ID/ED Task (Dias et al., 1996)
The ID/ED task is a learning task, where the partici-
pant has to select one of two stimuli presented to them on
a computer screen. The test stimuli can involve up to two
dimensions; object shape and line shape. The shapes and
tion that might involve the participant selecting between
the stimuli Shape 1 + Line 2 and Shape 2 + Line 1, for
example. The correct stimulus for a discrimination was
always specified by one exemplar from one dimension
(e.g., Shape 1 whether it is paired with Line 1 or 2).
504 Blair, Colledge, and Mitchell
On any one trial, the two test stimuli appeared ran-
domly in two of four rectangles positioned toward the
sides of the screen, and the participant was required to
click with the mouse on the box containing the correct
stimulus. If the participant chose correctly, the word
the screen. If the participant chose incorrectly, the word
INCORRECT, written in red ink, appeared. The subject
was considered to have learnt a given discrimination to
criterion after choosing the correct stimulus eight times in
the next stage is a run of 8 correct choices within 50 trials.
If this criterion is not achieved at a given stage, the test is
discontinued. The nine stages are:
to respond to Shape 1.
(2) Simple reversal, using the same stimuli but with
the contingencies reversed. The participant must
reverse his/her responding to Shape 1 and re-
spond to Shape 2.
(3) Compound discrimination-separate. A pair of
white line patterns is introduced (Line 1 and
Line 2). However, the contingencies remain un-
changed. The participant should maintain res-
ponding to Shape 2 irrespective of whether
Shape 2 is paired with Line 1 or Line 2. The
pink shapes and white line are kept separate to
encourage the subject to perceive them as dis-
tinct. Pairing is pseudorandom: the same pair-
ings (e.g., Shape 1–Line 1 and Shape 2–Line 2)
appear in runs of no more than three trials.
(4) Compound discrimination superimposed. The
white lines are superimposed on the pink forms
learning cannot be attributed to locational learn-
ticipant should maintain responding to Shape 2.
(5) Compound Reversal. The same stimuli are used
but the contingencies are reversed. The partici-
pant must reverse his/her responding to Shape 2
and respond to Shape 1. The lines remain with-
out predictive power for deciding on a correct/
(6) Intradimensional shift. New shapes and lines are
introduced (Shape 3 and Shape 4 and Line 3 and
Line 4). The participant must learn to respond to
Shape 3 irrespective of whether it is paired with
Line 3 or Line 4.
reversed. The participant must reverse his/her
response to Shape 3 in favor of responding to
and Line 6). However, in this phase of the task
the participant must shift attentional set from the
shapes to the lines. Thus, the participant must
it is paired with Shape 5 or Shape 6.
(9) Extradimensional reversal. The contingencies
response to Line 5 in favor of responding to
The number of errors the participant makes for each
stage are calculated by the computer.
Each participant was tested individually in a quiet
interview room allocated by the school. The tasks (ID/ED
and the Gambling Task) were run on two separate testing
sessions. Each task was described without informing the
participant of the investigation’s specific objectives and
expectations. BPVS was used to test participants’ IQ dur-
ing one of the testing sessions.
The Gambling Task
Because of the broad range in age and IQ in our
participants, an initial exploratory analysis investigated if
between IQ and task performance in this population for
any of the blocks (greatest r value for any block = 0.17;
ns). However, there was a significant association between
age and task performance (r for last block = −0.39; p =
0.01); the older children were more likely to avoid the
bad packs by the end of the paradigm. Thus, despite the
on the data, with age treated as a covariate.
Based on previous reports on adults (Bechara et al.,
1994, 1999) and our preliminary data on adult psycho-
pathic individuals (Mitchell et al., Submitted, 2001), we
predicted that comparison children would sample ran-
domly at first but then develop a preference for the less
risky decks. In contrast, we predicted that children with
psychopathic tendencies would be less likely to switch to
Somatic Markers and Response Reversal505
the safer C and D packs. The two groups were initially
compared on global performance by block using a mixed
model 2 (comparison children vs. children with psycho-
pathic tendencies) × 10 (blocks of 10 trials) ANCOVA.
Mauchley’s test of sphericity was significant (Mauchley’s
W = 0.07, df = 44, p < 0.01) thus the more conservative
Greenhouse–Geisser test was used for the analyses. This
revealed no significant main effect of block, F(5, 215) =
1.51, ns, but there was a significant effect of group,
F(1,40)=6.43, p <0.05;thechildrenwithpsychopathic
tendencies were more likely than the comparison group
to play from the disadvantageous A and B packs. Cru-
F(5, 215) = 2.32, p < 0.05. As the task continued, the
children with psychopathic tendencies failed to learn to
avoid the A and B packs whereas the comparison children
learnt to avoid these packs; see Fig. 1. There was a sig-
nificant main effect of the covariate age, F(1, 40) = 5.08,
p < 0.05.However,therewasnosignificantBlock×Age
A one-way ANCOVA was used to explore the pre-
diction that the difference in responding was specifically
related to the inability of children with psychopathic ten-
dencies to shift away from the bad packs. We compared
the difference in the number of times the children sam-
children with psychopathic tendencies were significantly
less likely to avoid the unfavourable packs than the com-
Fig. 1. The selections of boys with psychopathic tendencies and comparison boys from the risky packs across the 10
blocks of 10 trials. The boys with psychopathic tendencies were significantly more likely to pick from the risky packs
than were the comparison boys. Points represent the mean number of selections from the risky decks per 10 selections;
vertical lines depict standard errors of the means.
parisonchildren,F(1,42)=7.59, p < 0.01.Thecovariate
age was again significant, F(1, 42) = 10.74, p < 0.01.
Because of the broad range in age and IQ in our
participants, an initial exploratory analysis investigated
(Phases 1 and 6); reversal learning (Phases 2, 5, 7, and
9); and the extradimensional shift stage (Phase 8). This
revealed no significant associations between either age or
IQ and performance on any of these three components.
Thus, ANOVAs were used for data analysis.
Rahman et al., 1999; Rogers et al., 1999) and our pre-
liminary data on adult psychopathic individuals (Mitchell
et al., Submitted, 2001), we predicted that there would be
using a mixed-model 2 (comparison children vs. children
with psychopathic tendencies) × 3 (component: learning,
reversal learning, and extradimensional shift) ANOVA.
Mauchly’s test of sphericity was significant (Mauchley’s
W = 0.643,df = 2, p < 0.01)thusthemoreconservative
Greenhouse–Geisser test was used for the analyses. This
revealed a highly significant main effect of component,
506 Blair, Colledge, and Mitchell
Fig. 2. The numbers of errors committed by boys with psychopathic tendencies and comparison boys on the three
main components of the ID/ED task. The Extradimensional shift component was the most difficult component of the
task. However, there were no significant group differences. Points represent the mean number of selections from the
risky decks per 10 selections; vertical lines depict standard errors of the means.
F(1.5, 59) = 18.27; p < 0.01; the extradimensional shift
component was indeed the most difficult component of
the task; see Fig. 2. However, there was no significant
groupeffect, F(1,40) = 0.673;ns,orsignificantGroup×
Component interaction F(1.5,59) = 0.25; ns.
As far as we are aware, this is the first study to
specifically investigate orbitofrontal cortex functioning in
children with psychopathic tendencies. In line with pre-
dictions, children with psychopathic tendencies were less
likely than comparison children in the gambling task to
avoid the risky packs. However, in contrast to predictions,
the children with psychopathic tendencies did not show
significantly more reversal errors on the ID/ED task.
tioning in adult psychopathic individuals have indicated
2001; LaPierre et al., 1995). In contrast, the present study
found no indications of response reversal impairment in
children with psychopathic tendencies. As regards perfor-
mance on the gambling test, the data are more equivo-
cal (Mitchell et al., Submitted, 2001; Blair & Cipolotti,
2000; Schmitt et al., 1999). Schmitt et al. (1999), and
Blair & Cipolotti (2000) in a small group sample, found
that adult psychopathic individuals were not less likely
than the comparison individuals to avoid the risky packs.
However, both of these studies used task instructions that
differed from those of Bechara (Bechara et al., 1999).
Specifically, there were no instructions categorically in-
forming participants that some decks involve more loss
than others and that participants could win more money
overall if they avoid the costly decks. In the only study
where such instructions have been used, the adult psycho-
pathic individuals were significantly less likely than the
comparison individuals to avoid the risky packs (Mitchell
et al., Submitted, 2001). The current study, also using the
Bechara et al. (1999) instructions, replicates this result in
children with psychopathic tendencies.
It is interesting to compare the present results of the
gambling and ID/ED tasks to those obtained with a re-
lated paradigm; the one-pack card playing task (Newman,
Patterson, & Kosson, 1987). In this task, the participant
has to decide whether to play a card. Initially, the partici-
pant’s choices to play are reinforcing but as the number
of trials increases, the probability of reward decreases.
The participant should terminate responding before he
receives greater levels of punishment than reward. Both
children with psychopathic tendencies and adult psycho-
pathic individuals fail to modulate their responding as the
probability of punishment increases and perform more
Newman et al., 1987; O’Brien & Frick, 1996). This task
could be conceptualized as a response reversal task under
Somatic Markers and Response Reversal507
However, the present results suggest that it should not be.
The current data indicate that children with psychopathic
tendencies show no apparent general impairment in re-
The current data have specific implications for
current neurocognitive models of psychopathy. First it
appears that psychopathy, particularly when seen in chil-
dren, is not associated with generalized orbitofrontal cor-
tex dysfunction. Children with psychopathic tendencies
show no impairment in social response reversal; they re-
spond appropriately to angry expressions and can detect
inappropriate social behavior that is likely to cause anger
indicate that they also show no impairment in nonsocial
response reversal; they are sensitive to changes, at least if
they are sudden, in reinforcement schedules and reverse
their behavior accordingly.
be due to dysfunction in the somatic marker system, a
system that is mediated by a circuit that includes the or-
bitofrontal cortex (Damasio, 1994; Damasio et al., 1990).
The current data are in line with this position. However,
the somatic marker hypothesis faces at least one major
challenge when attempting to account for psychopathy. A
core prediction of the hypothesis is that individuals with
somatic marker impairment will not show autonomic re-
Damasio et al., 1990). However, children with psycho-
pathic tendencies and adult psychopaths do show arousal
to such stimuli, as long as these stimuli are not sad or
Thus, although the current data are compatible with the
hypothesis, the hypothesis faces difficulties in providing
a full account of psychopathy.
septo-hippocampal position on psychopathy (Gorenstein
& Newman, 1980). Current formulations on the role of
either the form of decision making involved in the gambl-
ing task or the response reversal investigated with the
ID/ED task (LeDoux, 1998; O’Keefe, 1991). However,
the current results do have implications for the cogni-
tive, response set modulation model that was developed
from the septo-hippocampal position (Newman, 1998;
Newman, Schmitt, & Voss, 1997; Patterson & Newman,
1993). Response modulation involves “a rapid and rela-
tively automatic (i.e., non-effortful or involuntary) shift
of attention from the effortful organisation and imple-
mentation of goal-directed behaviour to its evaluation”
(Newman et al., 1997). Dysfunction within the system
responsible for response modulation will result in im-
poverished performance under conditions where a salient
stimulus should divert attention from ongoing behavior.
chopathy will be more likely than nonpsychopathic in-
dividuals to persist in a previously rewarded response,
even if the rate of punishment for this response increases.
Support for this prediction comes from a series of stud-
ies employing the one-pack card playing task mentioned
& Frick, 1996). However, the current data constrain the
response set modulation hypothesis. The ID/ED task re-
quired shifts of attention during both response reversal
and when shifting attention from the shape to the lines
in the extradimensional shift. The children with psycho-
in the Bechara task where the child had to shift away from
C and D. Perhaps, a crucial variable that must be consid-
ered is the frequency of punishment. In at least the begin-
ning of the one-pack card playing task and the Bechara
task, the punishments are infrequent. However, in the re-
sponse reversal phases of the ID/ED task, the previously
always rewarded stimulus suddenly becomes the always
punished stimulus. The current data suggest that in condi-
tions of certainty the children can alter response set. This
suggests in turn that, if we are to adopt the response set
modulation hypothesis, we should consider the impair-
ment to be in detecting the altered contingencies rather
than in shifting response. If the change in contingencies
is easy to detect, as in the ID/ED task, there should be no
An alternative way of conceptualizing the current re-
sults makes reference to the interconnections between ba-
solateral amygdala and orbitofrontal cortex and their crit-
ical role in encoding and using associative information
about the motivational significance of stimuli (Gallagher,
Gallagher, 1998, 2000). Within this circuit, the amygdala
has a critical role in forming associations between uncon-
ditioned and conditioned stimuli and between individual
conditioned stimuli (Killcross, Robbins, & Everitt, 1997;
LeDoux, 1998; Schoenbaum et al., 1998). Orbitofrontal
cortex encodes the motivational significance of the cues
and the incentive value of expected outcomes (Gallagher
et al., 1999; Schoenbaum et al., 2000; Thorpe, Rolls, &
orbitofrontal cortex are involved in encoding the value
of a stimulus, and lesions to either system impair re-
versals in responding following changes in the value of
508 Blair, Colledge, and Mitchell
reinforcement (Gallagher et al., 1999; Hatfield, Han,
both amygdala and orbitofrontal cortex dysfunction have
been found to result in poor performance on the gam-
bling task (Bechara et al., 1994, 1999). Moreover, there is
certainly mounting evidence of amygdala dysfunction in
psychopathic individuals (see, for a review, Blair & Frith,
ment contingencies can be detected. Dysfunction within
to changes in reinforcement contingencies and thus only
give rise to impairment on tasks such as the one-pack card
playing task and the gambling task where the changes
are less easy to detect. Of course, the above position is
highly speculative. However, it is important to note that
it gives rise to clear, testable predictions. While children
with psychopathic tendencies are not impaired on the cur-
rent version of the ID/ED task, if the above position is
correct, it can be predicted that they will be impaired on a
ules are less clearly indicated; for example, in Phase 1, if
Shape 1 was rewarded and Shape 2 punished only 80%
of the time and then, in Phase 2, Shape 1 punished and
are currently under investigation.
It is important to note some interesting similarities
and differences between adult psychopathic individuals
has shown strong similarities between the performance
of children with psychopathic tendencies on many neu-
Fisher & Blair, 1998; Newman et al., 1987; O’Brien &
Frick, 1996). Moreover, the current results echo previous
findings showing an impairment on the gambling task in
2001). However, there are clear differences between the
performance of children with psychopathic tendencies
and adult psychopathic individuals on the ID/ED task.
Although the current study revealed no difficulty for chil-
dren with psychopathic tendencies on any phase of the
task, adult psychopathic individuals showed significant
impairment in response reversal. This suggests, following
is more severe in adult psychopathic individuals. Such in-
dividuals have difficulties even when the change in con-
tingencies is apparently very clear.
One brief caveat should be noted at this point: the
population studied in this paper overlapped considerably
tively small sample sizes, it will be important to replicate
these results in other settings. Moreover, it would be im-
portant to determine whether these results also extend to
girls with psychopathic tendencies. It might also be use-
ful to compare the performance of the two populations
in the present study with a third population. However, it
should be noted that the performance of the comparison
population on the four-pack card playing task was within
the range of healthy adult individuals (Bechara et al.,
1998) and performance of both groups on response rever-
healthy children of comparable age (Hughes, Russell, &
Robbins, 1994). Interestingly, children with ADHD, even
shift, unlike the populations in the current study, an error
et al., 1999).
An additional, interesting extension to this study
would be to investigate the relationship between perfor-
C/UN and I/CP). In this study, there was no overlap be-
tween the children with psychopathic tendencies and the
comparison children on either of these factors. However,
it would be predicted that measures of amygdala dysfunc-
tion should be more highly associated with the C/UN fac-
tor of psychopathy rather than with I/CP. In contrast, it
is possible to predict that some individuals with high lev-
els of impulsive, reactive aggression who do not present
bitofrontal cortex measures (see Blair & Cipolotti, 2000).
There are two main implications of the present re-
sults. First, the results have clear treatment implications.
If psychopathic tendencies reflect amygdala dysfunction,
as the current data suggest, neither empathy induction nor
anger management techniques are likely to be effective.
Empathy induction techniques will not be effective be-
cause the capacity to find the distress of others aversive
will be disrupted by the amygdala dysfunction (see Blair
& Frith, 2000). Anger management techniques will not
be effective because the disorder does not reflect an in-
ability to control reactive aggression (cf. Cornell et al.,
1996). In contrast, it reflects a heightened willingness to
engage in instrumental aggression. However, treatment
regimes based on altering the individual’s expectations
that aggressive action will have beneficial consequences
may prove useful (see Crick & Dodge, 1996). Second, the
differences between the child with psychopathic tenden-
cies and adult psychopaths, both showing impairment on
measures of amygdala dysfunction but the adults showing
Somatic Markers and Response Reversal509
far more pronounced impairment on measures reflecting
the interaction of orbitofrontal cortex and the amygdala
than children, have developmental implications. It is pos-
sible that because of the interconnections of the amyg-
dala and orbitofrontal cortex, a reduction in afferent in-
put from the amygdala (because of the primary amygdala
dysfunction) may, over time, have a negative impact on
the responsiveness of the orbitofrontal cortex. Accord-
ingly, the long-term effects of this dysfunction may not
be as apparent until later in the life span, thereby explain-
ing the lack of evidence for response reversal deficits in
children with psychopathic tendencies. A second possi-
bility is that the greater orbitofrontal cortex dysfunction
seen in the adults is a secondary consequence of some of
the behavioral characteristics of psychopathy. For exam-
ple, one of the criteria of psychopathy, stimulation seek-
ing, is often associated with drug use (Hare, 1991). Stud-
ies suggest that psychopathy is associated with higher
rates of drug abuse, dependence, and multiple drug use
(e.g., Smith & Newman, 1990; Hemphill, Hart, & Hare,
1994). Using a novel decision-making task, Rogers et al.
(1999) assessed the quality of decision making and de-
liberation time of individuals with focal orbitofrontal cor-
tex damage, and individuals who abused amphetamine or
opiates. All three groups showed impaired performance
on the task relative to comparison groups. Furthermore,
chronic amphetamine abusers showed a pattern of sub-
optimal decisionmaking that correlated with their years
of abuse. Given the neuro-cognitive impairments asso-
ciated with chronic drug abuse, and the data suggesting
higher rates of abuse and dependence among psycho-
pathic individuals, we cannot discount the possibility that
some of the decision-making impairments seen in psy-
chopathic individuals is acquired as a secondary conse-
quence of the stimulus-seeking behavior characteristic of
In conclusion, the current study found impaired per-
formance in children with psychopathic tendencies on the
gambling but not the ID/ED task. These data are in line
with the suggestion that amygdala dysfunction is the core
impairment in psychopathy (e.g., Blair & Frith, 2000).
Because of this dysfunction, the system crucial for repre-
senting the motivational value of stimuli that is mediated
orbitofrontal cortex, is disrupted. The dysfunction in this
(Mitchell et al., Submitted, 2001).
This work was supported by a Medical Research
Council grant [Ref. G9716841] and the Department of
Health (VISPED initiative). The authors would like to
thank the staff and students at the Batchwood, Boxmoor,
Larwood, and Rushden Schools.
Adolphs, R., Cahill, L., Schul, R., & Babinsky, R. (1997). Impaired
declarative memory for emotional material following bilateral
amygdala damage in humans. Learning and Memory, 4, 291–300.
Adolphs, R., Tranel, D., Young, A. W., Calder, A. J., Phelps, E. A.,
Anderson, A. K., Lee, G. P., & Damasio, A. R. (1999). Recogni-
tion of facial emotion in nine individuals with bilateral amygdala
damage. Neuropsychologia, 37, 1111–1117.
Anderson, S. W., Bechara, A., Damasio, H., Tranel, D., & Damasio,
A. R. (1999). Impairment of social and moral behavior related to
early damage in human prefrontal cortex. Nature Neuroscience, 2,
Angrilli, A., Mauri, A., Palomba, D., Flor, H., Birhaumer, N., Sartori,
G., & di Paola, F. (1996). Startle reflex and emotion modulation
impairment after a right amygdala lesion. Brain, 119, 1991–2000.
Aniskiewicz, A. S. (1979). Autonomic components of vicarious condi-
tioning and psychopathy. Journal of Clinical Psychology, 35, 60–
Barry, C. T., Frick, P. J., DeShazo, T. M., McCoy, M. G., Ellis, M., &
Loney, B. R. (2000). The importance of callous-unemotional traits
for extending the concept of psychopathy to children. Journal of
Abnormal Psychology, 109, 335–340.
Bechara, A., Damasio, H., & Damasio, A. R. (2000). Emotion, decision
making and the orbitofrontal cortex. Cerebral Cortex, 10, 295–
Bechara, A., Damasio, A. R., Damasio, H., & Anderson, S. W. (1994).
Insensitivity to future consequences following damage to human
prefrontal cortex. Cognition, 50, 7–15.
Bechara, A., Damasio, H., Damasio, A. R., & Lee, G. P. (1999). Dif-
ferent contributions of the human amygdala and ventromedial pre-
frontal cortex to decision-making. Journal of Neuroscience, 19,
Bechara, A., Damasio, H., Tranel, D., & Anderson, S. W. (1998). Disso-
prefrontal cortex. Journal of Neuroscience, 18, 428–437.
Bechara, A., Tranel, D., Damasio, H., Adolphs, R., Rockland, C., &
Damasio, A. R. (1995). Double dissociation of conditioning and
declarative knowledge relative to the amygdala and hippocampus
in humans. Science, 269, 1115–1118.
Blair, R. J. R. (1995). A cognitive developmental approach to morality:
Investigating the psychopath. Cognition, 57, 1–29.
Blair, R. J. R. (1997). Moral reasoning in the child with psychopathic
tendencies. Personality and Individual Differences, 22, 731–739.
Blair, R. J. R. (1999). Responsiveness to distress cues in the child with
psychopathic tendencies. Personality and Individual Differences,
Blair, R. J. R., & Cipolotti, L. (2000). Impaired social response reversal:
A case of “acquired sociopathy.” Brain, 123, 1122–1141.
Blair, R. J. R., & Coles, M. (2000). Expression recognition and Behav-
ioral problems in early adolescence. Cognitive Development, 15,
Blair, R. J. R., Colledge, E., Murray, L., & Mitchell, D. (2001). A selec-
tive impairment in the processing of sad and fearful expressions in
children with psychopathic tendencies. Journal of Abnormal Child
Blair, R. J. R., & Frith, U. (2000). Neuro-cognitive explanations of the
Antisocial Personality Disorders. Criminal Behavior and Mental
510 Blair, Colledge, and Mitchell
individual: A lack of responsiveness to distress cues? Psychophys-
iology, 34, 192–198.
Dissociable neural responses to facial expressions of sadness and
anger. Brain, 122, 883–893.
Blumer, D., & Benson, D. F. (1975). Personality changes with frontal
and temporal lobe lesions. In D. F. Benson & D. Blumer (Eds.),
Psychiatric aspects of neurological disease (pp. 151–170). New
York: Grune & Stratton.
Cleckley, H. (1967). The mask of sanity (4th ed.). St Louis, MO: Mosby.
Cornell, D. G., Warren, J., Hawk, G., Stafford, E., Oram, G., & Pine, D.
Journal of consulting and clinical psychology, 64, 783–790.
Crick, N. R., & Dodge, K. A. (1996). Social information-processing
mechanisms on reactive and proactive aggression. Child Develop-
ment, 67(3), 993–1002.
Damasio, A. R. (1994). Descartes’ error: Emotion, rationality and the
human brain. New York: Putnam (Grosset Books).
Damasio, A. R., Tranel, D., & Damasio, H. (1990). Individuals with so-
ciopathic behavior caused by frontal damage fail to respond auto-
nomically to social stimuli. Behavioral Brain Research, 41, 81–94.
Dias, R., Robbins, T. W., & Roberts, A. C. (1996). Dissociation in pre-
Dunn, L. M., Wheklan, C., & Pintillie, D. (1982). British Picture Vocab-
ulary Scale. Windsor, UK: NFER-Nelson.
Fine, C., & Blair, R. J. R. (2000). The cognitive and emotional effects of
amygdala damage. Neurocase, 6, 435–450.
Fisher, L., & Blair, R. J. R. (1998). Cognitive impairment and its rela-
tionship to psychopathic tendencies in children with emotional and
behavioral difficulties. Journal of Abnormal Child Psychology, 26,
Frick, P. J. (1995). Callous–unemotional traits and conduct problems:
A two-factor model of psychopathy in children. Issues in Crimino-
logical and Legal Psychology, 24, 47–51.
chopathy and Conduct Problems in Children. Journal of Abnormal
Psychology, 103, 700–707.
Gallagher, M., McMahan, R. W., & Schoenbaum, G. (1999). Orbito-
frontal cortex and representation of incentive value in associative
learning. Journal of Neuroscience, 19, 6610–6614.
Gorenstein, E. E., & Newman, J. P. (1980). Disinhibitory Psychopathol-
ogy: A new perspective and a model for research. Psychological
Review, 37, 301–315.
Grafman, J., Schwab, K., Warden, D., Pridgen, B. S., & Brown, H. R.
(1996). Frontal lobe injuries, violence, and aggression: A report of
the Vietnam head injury study. Neurology, 46, 1231–1238.
Gray, J. A. (1987). The psychology of fear and stress (2nd ed.).
Cambridge: University of Cambridge Press.
Hare, R. D. (1998). Psychopathy, affect and behavior. In D. J. Cooke,
A. E. Forth, & R. D. Hare (Eds.), Psychopathy: Theory, research
and implications for society (pp. 81–105). Dordrecht: Kluwer.
Hare, R. D., & Jutai, J. W. (1983). Criminal history of the male psy-
chopath: Some preliminary data. In K. T. Van Dusen & S. A.
Mednick (Eds.), Prospective studies of crime and delinquency
(pp. 225–236). Boston: Kluwer-Nijhoff.
Hatfield, T., Han, J. S., Conley, M., Gallagher, M., & Holland, P. (1996).
Neurotoxic lesions of basolateral, but not central, amygdala inter-
fere with Pavlovian second-order conditioning and reinforcer de-
valuation effects. Journal of Neuroscience, 16, 5256–5265.
Hemphill, J. F., Hart, S. D., & Hare, R. D. (1994). Psychopathy and
substance use. Journal of Personality Disorders, 8, 169–180.
distress in prison psychopaths. Journal of Personality and Social
Psychology, 34, 556–560.
dysfunction in autism. Neuropsychologia, 32, 477–492.
Kempton, S., Vance, A., Maruff, P., Luk, E., Costin, J., & Pantelis, C.
(1999). Executive function and attention deficit hyperactivity dis-
order: Stimulant medication and better executive function perfor-
mance in children. Psychological Medicine, 29, 527–538.
Killcross, S., Robbins, T. W., & Everitt, B. J. (1997). Different types
of fear-conditioned behavior mediated by separate nuclei within
amygdala. Nature, 388, 377–380.
LaBar, K. S., LeDoux, J. E., Spencer, D. D., & Phelps, E. A. (1995).
in humans. Journal of Neuroscience, 15, 6846–6855.
LaPierre, D., Braun, C. M. J., & Hodgins, S. (1995). Ventral frontal
deficits in psychopathy: Neuropsychological test findings. Neu-
ropsychologia, 33, 139–151.
LeDoux, J. E. (1995). Emotion: Clues from the brain. Annual Review of
Psychology, 46, 209–235.
LeDoux, J. (1998). The emotional brain. New York: Weidenfeld &
Lykken, D. T. (1957). A study of anxiety in the sociopathic personality.
Journal of Abnormal and Social Psychology, 55, 6–10.
Lykken, D. T. (1995). The antisocial personalities. Hillsdale, NJ:
markers and response reversal: Is there evidence of orbitofrontal
cortex dysfunction in psychopathic individuals? Manuscript sub-
mitted for publication.
Newman, J. P. (1998). Psychopathic behavior: An information pro-
cessing perspective. In D. J. Cooke, A. E. Forth, & R. D. Hare
(Eds.), Psychopathy: Theory, research and implications for society
(pp. 81–105). Dordrecht: Kluwer.
Newman, J. P., Patterson, C. M., & Kosson, D. S. (1987). Response
perseveration in psychopaths. Journal of Abnormal Psychology,
Newman, J. P., Schmitt, W. A., & Voss, W. D. (1997). The impact of
the generality of the response modulation hypothesis. Journal of
Abnormal Psychology, 106, 563–575.
O’Brien, B. S., & Frick, P. J. (1996). Reward dominance: Associa-
tions with anxiety, conduct problems, and psychopathy in children.
Journal of Abnormal Child Psychology, 24, 223–240.
O’Keefe, J. (1991). The hippocampal cognitive map and navigational
strategies. In J. Paillard (Ed.), Brain and space (pp. 273–295).
Oxford: Oxford University Press.
Owen, A. M., Roberts, A. C., Polkey, C. E., Sahakian, B. J., & Robbins,
performance following frontal lobe excisions, temporal lobe exci-
sions or amygdalo-hippocampectomy in man. Neuropsychologia,
Patrick, C. J. (1994). Emotion and psychopathy: Startling new insights.
Psychophysiology, 31, 319–330.
Patrick, C. J., Bradley, M. M., & Lang, P. J. (1993). Emotion in the crim-
inal psychopath: Startle reflex modulation. Journal of Abnormal
Psychology, 102, 82–92.
Patterson, C. M., & Newman, J. P. (1993). Reflectivity and learning
from aversive events: Toward a psychological mechanism for the
syndromes of disinhibition. Psychological Review, 100, 716–736.
Rahman, S., Sahakian, B. J., Hodges, J. R., Rogers, R. D., & Robbins,
totemporal dementia. Brain, 122, 1469–1493.
Rogers, R. D., Blackshaw, A. J., Middleton, H. C., Mathews, K.,
Hawtin, K., Crowley, C., Hopwood, A., Wallace, C., Deakin, J. F.,
Sahakian, B. J., & Robbins, T. W. (1999). Tryptophan depletion
impairs stimulus-reward learning while methylphenidate disrupts
attentional control in healthy young adults: Implications for the
Berl, 146, 482–491.
of the Royal Scoiety B, 351, 1433–1443.
Somatic Markers and Response Reversal511 Download full-text
learning in patients with social and emotional changes associated
Psychiatry, 57, 1518–1524.
Schmitt, W. A., Brinkley, C. A., & Newman, J. P. (1999). Testing Dama-
sio’s somatic marker hypothesis with psychopathic individuals:
Risk takers or risk averse? Journal of Abnormal Psychology, 108,
Schoenbaum, G., Chiba, A. A., & Gallagher, M. (1998). Orbitofrontal
cortex and basolateral amygdala encode expected outcomes during
learning. Nature Neuroscience, 1, 155–159.
during learning and reversal training. Journal of Neuroscience, 20,
Smith, S. S., & Newman, J. P. (1990). Alcohol and drug abuse-
dependence disorders in psychopathic and nonpsychopathic
criminal offenders. Journal of Abnormal Psychology, 4, 430–
Stevens, D., Charman, T., & Blair, R. J. R. (2001). Recognition of emo-
tion in expressions and vocal tones in children with psychopathic
tendencies. Journal of Genetic Psychology, 162(2), 201–211.
Thorpe, S. J., Rolls, E. T., & Maddison, S. (1983). The orbitofrontal
cortex: Neuronal activity in the behaving monkey. Experimental
Brain Research, 49, 93–115.
Williamson, S., Hare, R. D., & Wong, S. (1987). Violence: Criminal
psychopaths and their victims. Canadian Journal of Behavioral
Science, 19, 454–462.