Transcript
Journal of Abnormal Psychology
2013, Vol. 122, No. 1, 156 –166
© 2012 American Psychological Association
0021-843X/13/$12.00 DOI: 10.1037/a0030261
Impaired Cognitive Empathy in Criminal Psychopathy:
Evidence From a Laboratory Measure of Empathic Accuracy
Michael Brook
David S. Kosson
Northwestern University
Rosalind Franklin University of Medicine and Science
Empathy deficits feature prominently in theoretical accounts of psychopathy, yet studies that have
examined various aspects of emotional processing related to empathy have produced a mixed body of
findings. We created a laboratory measure of cognitive empathy based on the empathic accuracy
paradigm (i.e., the ability to accurately infer others’ emotions in a simulated interpersonal interaction)
and used it to examine relationships between psychopathy (assessed with the Psychopathy Checklist Revised) and cognitive empathy in a sample of incarcerated male offenders. Psychopathy was inversely
associated with empathic accuracy performance, as well as with the number of responses when rating the
emotional states of others. Empathic accuracy performance was most strongly associated with the
behavioral/antisocial and lifestyle features of psychopathy. When the emotional content of target
vignettes was examined, psychopathy was associated with poorer empathic accuracy for negatively
valenced emotions only (fear and sadness), although nonsignificant moderate effect sizes were also
observed for joy. Whereas the interpersonal/affective factor of psychopathy was associated with poor
empathic accuracy for joy, the behavioral/antisocial factor was associated with poor overall empathic
accuracy for negatively valenced emotions. At the psychopathy facet level, the interpersonal and lifestyle
features of psychopathy were associated with poor empathic accuracy for positively valenced emotions,
whereas the affective and antisocial features of psychopathy were inversely associated with empathic
accuracy for negatively valenced emotions. In contrast to its association with poor empathic accuracy
performance, psychopathy was not associated with ratings of perceived task difficulty.
Keywords: psychopathy, antisocial personality, empathy, empathic accuracy, emotion
The striking lack of empathy often described as characteristic of
psychopaths has received a great deal of attention in lay, clinical,
and theoretical accounts of psychopathy (Cleckley, 1941/1988;
Hare, 2006; Soderstrom, 2003). However, surprisingly few empirical investigations have directly examined relationships between
psychopathy and performance on laboratory measures of empathy.
To address this gap, we designed a laboratory measure of empathic
processing based on the well validated empathic accuracy paradigm and used this measure to elucidate relationships between
psychopathy and empathic processing.
Empathy and Empathic Accuracy
Empathy has been defined as a complex interpersonal phenomenon that involves the formation of an affective/motor/cognitive
connection between an individual displaying an emotional behavior (hereafter referred to as a target) and an individual observing
that emotional display (hereafter referred to as a perceiver). At a
very basic level, an empathic interaction involves a perceiver
internalizing and experiencing, at least to a limited degree, the
affective state of a target. This “affective contagion” is thought to
occur largely automatically and involuntarily and has been functionally linked to the limbic system and surrounding cortical
structures (Blair, 2005; Decety & Jackson, 2004; Hatfield, Cacioppo, & Rapson, 1994; Smith, 2006). It has also been reported
that a perceiver tends to synchronize his or her body language
(such as facial expressions, hand gestures, or posture) with that of
a target. As with affective contagion, this motor mimicry is
thought to occur predominantly on a subconscious level of awareness and appears to be functionally linked to the activity of mirror
neurons located in the parietal and inferior frontal cortical areas
(Blair, 2005; Carr, Iacoboni, Dubeau, Mazziotta, & Lenzi, 2003;
Decety & Jackson, 2006). Sharing in the affective state and motor
behavior of a target has been theorized to represent a lower-order,
rudimentary system of empathic processing sometimes referred to
as perception-action coupling (Preston & de Waal, 2002), which
evolved to facilitate a variety of prosocial and cooperative behaviors observed in the animal world.
This article was published Online First October 15, 2012.
Michael Brook, Stone Institute of Psychiatry, Northwestern University;
David S. Kosson, Department of Psychology, Rosalind Franklin University
of Medicine and Science.
This work was completed by the first author in partial fulfillment of the
requirements for the degree of Doctor of Philosophy at the Rosalind
Franklin University of Medicine and Science under the supervision of
David S. Kosson. We are indebted to Jean Decety, John Calamari, Kenneth
Kessler, and Gary Oltmans for their helpful comments on the dissertation.
We also express our gratitude to Jennifer Witherspoon, Richard Riddle,
and the rest of the Lake County Jail staff for their consistent support of this
research.
Correspondence concerning this article should be addressed to
Michael Brook, Northwestern University, Stone Institute of Psychiatry,
446 East Ontario Street 7-250, Chicago, IL 60611. E-mail: michael
[email protected]
156
COGNITIVE EMPATHY AND PSYCHOPATHY
In contrast, cognitive empathy is considered a higher-order
process that involves complex mentation, generally including conscious understanding of another’s internal states (Ickes, 1997), as
well as self-other awareness or the ability of a perceiver to mentally separate himself/herself from a target (Decety & Jackson,
2004). Unlike perception-action coupling, cognitive empathy is a
process that appears to require awareness and volition and can be
used by a perceiver to regulate his or her response in an empathic
interaction. Cognitive empathy involves complex interpersonal
judgment in addition to input from affective and episodic memory
stores and is thought to be functionally subserved by a diffuse
cortical network in the prefrontal and temporoparietal cortices
(Blair, 2005; Decety & Jackson, 2006). Two forms of cognitive
empathy, each with partially distinct neural correlates, have been
distinguished in the literature; the first involves inferences regarding others’ emotional states, and the second (termed Theory of
Mind) involves inferences regarding others’ complex intentions
and beliefs (Blair, 2006). Contemporary models of human empathy include perception-action coupling at the ontogenic core and
cognitive empathy as the more advanced evolutionary development (de Waal, 2005). Arguably, it is the cognitive component that
enables the distinction between human empathy and perceptionaction coupling displayed by nonhuman species, and it was the
focus of the present investigation.
Cognitive empathy has been operationalized in many studies
using self-report measures. However, the assumption that individuals are able to understand and report on subtle cognitive and
affective processes (such as empathy) is controversial (Nisbett &
Wilson, 1977; Sprangers, Van den Brink, Van Heerden, & Hoogstraten, 1987). Arguably the best validated laboratory measure of
cognitive empathy to date is the empathic accuracy paradigm
developed by Ickes and colleagues, which is designed to measure
the ability to accurately infer emotional states from videotaped
recordings of naïve (i.e., nonactor, nonconfederate) participants
(for details see Ickes, 1997). The empathic accuracy paradigm is
especially appealing for the study of human empathic processing
because of its use of seminaturalistic interpersonal interactions as
stimuli, thus emphasizing ecological validity in a standardized
laboratory task. This methodology has been validated in a variety
of normative samples (Hall & Schmid Mast, 2007; Ickes, Stinson,
Bissonnette, & Garcia, 1990; Ponnet, Buysse, Roeyers, & De
Corte, 2005; Simpson, Orina, & Ickes, 2003), but no prior study,
to our knowledge, has examined empathic accuracy in psychopathic offenders. In addition, the standard Ickes empathic accuracy
paradigm has substantial limitations, which we addressed in the
current study (see below).
Prior Research on Empathic Processing in
Psychopathy
Empathy dysfunction has been strongly implicated in theoretical
and clinical descriptions of psychopathy (Cleckley, 1941/1988;
Hare, 1999). In fact, negative associations have been consistently
reported between various self-report indices of empathy and psychopathic personality traits (e.g., Anastassiou-Hadjicharalambous
& Warden, 2008; Sandoval, Hancock, Poythress, Edens, & Lilienfeld, 2000)1. However, findings from prior self-report studies that
specifically examined cognitive empathy using the Perspective
Taking subscale of the Davis Interpersonal Reactivity Index (see
157
Method) are mixed, with evidence of both negative correlations
(Mullins-Nelson et al., 2006; Pardini et al., 2003) and no relationship (Williams & Paulhus, 2004; Zágon & Jackson, 1994) between
self-report measures of psychopathic traits and perspective taking.
Two other studies reported no significant associations between
clinical ratings of psychopathy (using the Hare Psychopathy
Checklist) and perspective taking scores (Book & Quinsey, 2004;
Dolan & Fullam, 2004). Furthermore, the conclusions that can be
drawn from these studies are limited, as self-report and observerreport empathy scores are informative about the perception of
individuals’ personality and behavioral traits related to empathy,
but they do not provide insight into the processes involved in
empathic interaction.
Laboratory studies that have examined relationships between
psychopathic traits and performance on tasks related to various
components of empathic processing have also produced mixed
findings. Several of these studies examined psychophysiological
responding (skin conductance, cardiac reactivity) to emotionally
laden stimuli. These paradigms can be said to bear on the lowerorder, affective component of empathy because they involve
largely automatic responses to emotionally evocative stimuli involving other people but have little implication for the cognitive
component of empathy. The reader is referred to Brook, Brieman,
and Kosson (2012) for a detailed review of these findings.
Much of the laboratory evidence regarding the higher-order
cognitive component of empathy has come from behavioral paradigms that required participants to make conscious decisions regarding emotions or mental states displayed by target participants.
Findings suggest that, when asked to classify affect from static
pictures of faces, psychopaths do not commit an overall greater
number of classification errors than nonpsychopathic participants
(Blair & Coles, 2000; Book, Quinsey, & Langford, 2007; Glass &
Newman, 2006; Kosson, Suchy, Mayer, & Libby, 2002; Pham &
Philippot, 2010; Stevens, Charman, & Blair, 2001), with one
exception (cf. Habel, Kuhn, Salloum, Devos, & Schneider, 2002).
Instead, psychopaths’ recognition deficits tend to be circumscribed
to emotions of fear (Blair & Coles, 2000; Stevens et al., 2001),
sadness (Blair & Coles, 2000; Stevens et al., 2001), disgust (Kosson et al., 2002; Pham & Philippot, 2010), anger (Pham &
Philippot, 2010), and/or happiness (Pham & Philippot, 2010),
although studies vary with regard to which specific emotions are
associated with impaired recognition. Conversely, in paradigms
that use composite or morphed pictures of facial affect (such that
each successive slide provides more information regarding the
target emotion), psychopathy is typically associated with overall
recognition deficits (Dolan & Fullam, 2006; Hastings, Tangney, &
Stuewig, 2008), as well as deficits in recognizing fear (Blair,
Colledge, Murray, & Mitchell, 2001; Blair et al., 2004), sadness
(Blair et al., 2001; Dolan & Fullam, 2006; Hastings et al., 2008),
1
Negative associations have been reported between self-report measures of empathy and psychopathic traits in both incarcerated (Flight &
Forth, 2007; Pardini, Lochman, & Frick, 2003; Sandoval et al., 2000) and
community (Anastassiou-Hadjicharalambous & Warden, 2008; MullinsNelson, Salekin, & Leistico, 2006; Williams & Paulhus, 2004; Zágon &
Jackson, 1994) adult samples (cf. Book & Quinsey, 2004; Dolan & Fullam,
2004). Similarly, studies have reported significant negative associations
between psychopathic traits and observer ratings of empathy in children as
young as 4 years of age (Dadds et al., 2009).
158
BROOK AND KOSSON
and happiness (Hastings et al., 2008; Dolan & Fullam, 2006).
Similarly, studies examining vocal affect recognition have reported that individuals with psychopathic traits have overall affect
recognition deficits (Bagley, Abramowitz, & Kosson, 2009; Blair,
Budhani, Colledge, & Scott, 2005; Blair et al., 2002; Stevens et al.,
2001), in addition to deficits in recognizing fear (Blair et al., 2005;
Blair et al., 2002) and sadness (Stevens et al., 2001) from others’
speech. In one study, psychopathy was associated with overall
recognition deficits when emotion classifications were made solely
on the basis of prosodic speech cues, or solely on the basis of
semantic speech cues. Additionally, psychopaths were selectively
deficient in recognizing sadness and happiness from semantic cues
and in recognizing surprise from prosodic cues (Bagley et al.,
2009).
Importantly, not all behavioral research suggests that psychopaths are deficient in their ability to recognize emotion. For example, studies using Theory of Mind (ToM) tasks have consistently shown that, compared with nonpsychopathic controls,
psychopathic individuals are not deficient in correctly inferring
others’ mental states from simple stories (Blair, Sellars, Strickland,
& Clark, 1996; Dolan & Fullam, 2004; Jones, Happe, Gilbert,
Burnett, & Viding, 2010) or from static pictures of the eye region
(Dolan & Fullam, 2004; Richell et al., 2003). One study found that
inmates with elevated psychopathic traits were deficient in inferring affective states (emotions) but not cognitive states (beliefs) in
a modified ToM task (Shamay-Tsoory, Harari, Aharon-Peretz, &
Levkovitz, 2010). Moreover, two studies reported that psychopathic inmates did not differ significantly from community controls in their ability to categorize emotions from static pictures of
facial expressions, either overall or for specific emotions (Book et
al., 2007; Dolan & Fullam, 2004).
In fact, there is some evidence suggesting superior performance
of psychopaths on tasks related to cognitive empathy (Book et al.,
2007; Dolan & Fullam, 2004; Hansen, Johnsen, Hart, Waage, &
Thayer, 2008). Such findings appear incongruent with the view of
psychopathy as psychopathology; consequently, several theorists
have argued that superior cognitive empathy combined with behavioral hostility, disinhibition, and manipulativeness confers an
evolutionary advantage to psychopaths (Book & Quinsey, 2004).
This view stems from the evolutionary psychology perspective,
which holds that socially malignant traits (e.g., deceptiveness,
aggression, emotional callousness), when expressed at low and
constant frequency in a population, may confer reproductive and
other survival advantages on individuals in whom these traits are
prominently expressed (Mealey, 1995).
The findings reviewed above provide important insights into
specific abnormalities impacting the emotional functioning of psychopaths, yet they permit only a partial understanding of potential
empathy abnormalities in psychopathy, with behavioral paradigms
producing findings of deficient, equal, and superior processing of
emotional information in psychopathic versus nonpsychopathic
participants. This lack of consensus in the extant experimental
literature may be in part attributable to the substantial variance in
the methodology used in these studies, both in terms of psychopathy assessment and indices of emotional processing. Several prior
studies had derived psychopathy ratings from various self-report
measures, thus limiting the generalizability of findings to psychopathy as assessed via clinical ratings. Furthermore, the few studies
that had separately examined the specific dimensions of psychop-
athy (Bagley et al., 2009; Dolan & Fullam, 2006; Habel et al.,
2002) reported divergent relationships with measures of empathic
processing, highlighting the importance of examining the components of psychopathy in addition to total scores.
A more important gap in the literature stems from the fact that
no prior psychopathy study has addressed empathic processing in
a seminaturalistic paradigm consistent with contemporary definitions of cognitive empathy. Stimuli used in past investigations
(static pictures, recordings of human vocalizations, motion picture
clips) have limited ecological validity because they fail to capture
the complexity of human empathic interaction, which involves the
simultaneous interpretation by a perceiver of multiple verbal and
nonverbal cues displayed by a target. The empathic accuracy
paradigm, on the other hand, has the needed balance between
empirical rigor and ecological validity that could provide important insights into empathic processing in psychopathy.
To this end, we designed a behavioral measure of empathic
accuracy based on the Ickes et al. method and used this measure to
examine relationships between psychopathy and cognitive empathy. Two competing hypotheses were tested. The deficient empathy hypothesis was based on theoretical accounts of psychopaths as
emotionally underresponsive and empirical findings linking psychopathy to an impaired ability to discriminate emotions of others
and to physiological underresponsiveness to distress cues of others. According to this hypothesis, psychopathy is characterized by
one or more disruptions within the emotion processing network
and should thus be associated with poor cognitive empathy performance. Alternatively, the evolutionary fitness hypothesis was
based on the evolutionary psychology conceptualization of psychopathy as a survival strategy and on empirical findings linking
psychopathy to adequate or superior performance on emotion
processing tasks. According to the evolutionary fitness hypothesis,
psychopathy is characterized by at least an unimpaired ability to
understand others’ emotions, and should be associated with adequate or even superior cognitive empathy performance. In addition, we hypothesized that cognitive empathy performance should
correlate positively with trait cognitive empathy as measured by
self report.
A concomitant aim of this study was to design an empathic
accuracy task that improved on the original Ickes (1997) protocol,
which has been criticized for its reliance on a single target who
was relatively low in expressiveness and an open-ended response
format which makes completion of the paradigm and coding of the
responses somewhat cumbersome and arbitrary. Consequently, we
attempted to design a paradigm with multiple individuals serving
as targets to generate greater variation in target expressiveness. In
addition, our task used a standardized forced choice response
format to permit a standardized analysis of participant data.
Method
Participants
Participants were 103 adult male county jail inmates incarcerated for a felony or misdemeanor offense. Eligibility criteria for
participation were as follows: estimated IQ ⬎69, native English
speaker, reading level ⱖ4th grade, no history of severe traumatic
brain injury, no current diagnosis of a psychotic disorder, and no
current use of psychotropic medication. Participants’ ages ranged
COGNITIVE EMPATHY AND PSYCHOPATHY
from 18 to 51 years (M ⫽ 31.31, SD ⫽ 8.00). Sixty-two of the
participants (60.2%) identified themselves as African American,
33 (32.0%) as Caucasian, 7 (6.8%) as Latino,2 and 1 participant
(1.0%) identified himself as other or mixed. The number of years
of formal education ranged from 8 to 16 (M ⫽ 11.34, SD ⫽ 1.70).
Measures
Psychopathy. The Psychopathy Checklist – Revised (PCL-R;
Hare, 1991) is the best validated and most widely used measure of
psychopathy in incarcerated populations. It consists of 20 items
rated on an ordinal scale (0 –2) according to the degree to which
the relevant disposition applies to the individual. Ratings were
made on the basis of a detailed semistructured interview and a
thorough review of available institutional files. In addition to the
total score, we examined PCL-R scores at the factor and facet
levels. The two-factor model (Harpur, Hare, & Hakstian, 1989)
consists of Factor 1 (F1) indexing the affective and interpersonal
traits of psychopathy and Factor 2 (F2) indexing the behavioral
and antisocial features of the disorder. The four-facet model (Hare,
2003; Neumann, Vitacco, Hare, & Wupperman, 2005) further
partitions F1 scores into the interpersonal (INT) and affective
(AFF) facets and F2 scores into the lifestyle (LIF) and antisocial
(ANT) facets.3 The PCL-R total, factor, and facet scores have been
extensively demonstrated to provide valid and reliable measures of
psychopathic traits in various adult incarcerated samples (Cooke,
Kosson, & Michie, 2001; Hemphill, Hare, & Wong, 1998; Sullivan, Abramowitz, Lopez, & Kosson, 2006). Psychopathy ratings
were made by clinical psychology graduate students with extensive training in PCL-R assessment. Interrater agreement for PCL-R
scores was good (average intraclass r ⫽ .83; one-way randomeffects model, n ⫽ 12 pairs of raters).
Cognitive empathy. The empathic accuracy task was designed in accordance with the Ickes (1997) methodology, while
improving on the original protocol in two major ways. First, the
stimulus recording was designed to include target participants of
both sexes and of varying age and ethnicity. Second, we introduced a standardized forced-choice response format as opposed to
the open response format in the original protocol that required
subjective scoring by independent judges.
Stimulus (target) recordings were created by videotaping adult
volunteers while they relayed past emotional experiences. To
simulate a naturalistic interpersonal interaction, target participants
were seated directly across a video camera, which was unobtrusively positioned behind the experimenter, at participants’ eye
level. The experimenter then randomly assigned to each volunteer,
one at a time, three primary emotions from the tripartite hierarchical inventory of emotion words outlined by Parrott (2001).
Briefly, the primary level of Parrott’s hierarchy consists of six
basic emotions (love, joy, surprise, anger, sadness, fear) commonly
described in the empirical psychology literature. The secondary
level comprises 25 more complex descriptors of emotional states
(e.g., affection, zest, irritation, disappointment, nervousness)
nested within each primary emotion. The tertiary level comprises
135 descriptors of complex mental states (e.g., longing, contentment, resentment, embarrassment) nested within each secondary
emotion.
For each of the assigned primary emotions, volunteers were
asked to recall and provide a brief but detailed verbal account of a
159
time in their lives when they felt the assigned emotion while
refraining from referring to the emotion by name. Upon completion of their stories, volunteers viewed their recordings and were
asked to identify and stop the playback at moments during which
they felt an emotion, even if that emotion was different from the
assigned emotion. For each such moment (or stop point), volunteers were asked to identify and rank-order (from most to least) the
emotions they felt at that moment as they were relating their story.
Potential responses consisted of the 25 secondary emotions from
the hierarchy described above, and each secondary emotion was
accompanied by a parenthesized list of its corresponding tertiary
emotions to aid in rating the secondary emotions by providing
additional emotional context. To optimize ecological validity and
allow us to examine the breadth of participants’ emotional responding, we did not restrict the number of emotions participants
could endorse. Primary emotions were omitted from the response
options to ensure that the task was not too easy. After the target
recording sessions, recordings were edited to isolate a number of
short video vignettes, each ending at the exact stop point identified
by each volunteer as the moment when they felt an emotion.
The resulting video vignettes were piloted with a small (n ⫽ 15)
sample of healthy participants to select the set of vignettes that
provided the best combination of: psychometric fit (internal consistency, discriminating ability), variation in length and affective
valence of the stories, and demographic characteristics of target
participants. The final recording was 15.7 minutes in length and
included 13 video vignettes from seven target participants (see the
Appendix for detailed description). Three participants were Caucasian (two female, one male), two were East Asian (one female,
one male), and two were African American (one female and one
male). Age of target participants ranged from 24 to 43. Six of the
seven target participants provided two vignettes, so as to enable an
examination of change in empathic accuracy with repeated exposure to one target participant. With regard to the emotion ranked
first by the target for each vignette, six of the vignettes portrayed
positively valenced emotions (corresponding to primary emotions
of love, joy, and surprise4), and seven portrayed negatively valenced emotions (corresponding to primary emotions of sadness
and fear).
Perceiver participants viewed each target recording and were
asked to rank-order (using the same response format as target
participants) the emotions experienced by each target at the end (or
stop point) of each vignette. Empathic accuracy was operationalized as the degree of correspondence (0 –2 Likert scale) between a
perceiver’s rating of the most salient emotion (i.e., the emotion
ranked #1) and each target’s rating of the most salient emotion. In
addition, we examined the total number of emotions endorsed by
perceiver participants as a measure of breadth of emotional responding. Finally, perceiver participants were asked to provide a
2
As our sample included only seven Latinos, we conducted all analyses
first including and then excluding the Latino participants. Because there
was no difference in the pattern of significance of findings, only the full
sample analyses are reported in the Results section.
3
Past factor analyses have also identified a three-factor solution that is
similar to the four-facet model but omits the antisocial facet (Cooke,
Michie, Hart, & Clark, 2004).
4
Although surprise is not intrinsically a positively or negatively valenced emotion, both vignettes that ranked surprise as #1 in the target
recording were positively valenced in content.
BROOK AND KOSSON
160
rating (on a 1–10 Likert scale, with 10 indicating most difficult) of
the difficulty of rating each vignette as a measure of participants’
own perception of their empathic accuracy performance.
Self-reported empathy. The Davis Interpersonal Reactivity
Index (IRI; Davis, 1983) is a self-report measure that has been
validated extensively in a variety of clinical and nonclinical samples, including psychopathic offenders (Jolliffe & Farrington,
2004; Mullins-Nelson et al., 2006; Rankin, Kramer, & Miller,
2005). The IRI is composed of 28 items that load on four conceptually independent scales: Perspective Taking (PT; the tendency to
understand and adopt others’ psychological states), Fantasy (FA;
the tendency to imaginatively transport self into fictional scenarios), Empathic Concern (EC; sympathy toward others’ misfortunes), and Personal Distress (PD; anxiety in emotional interpersonal situations). Of the four IRI scales, Perspective Taking is
conceptually most closely related to the construct of cognitive
empathy. Scores on the IRI subscales demonstrated acceptable
internal consistency in the current sample (Cronbach’s alpha: PT
⫽ .53, FA ⫽ .67, EC ⫽ .70, PD ⫽ .56).
General intelligence. Given the prior finding that differences
in self-reported empathy between psychopathic and nonpsychopathic inmates are apparent at higher (but not lower) levels of
intelligence (Heilbrun, 1982), we used the Shipley Institute of
Living Scale (SILS; Zachary, 1986) as a measure of general
intelligence. The SILS-estimated IQ scores (Weiss & Schell, 1991)
have been previously validated in forensic samples (Szyhowski,
2008) and correlate highly with scores on other brief measures of
intellectual functioning (Bowers & Pantle, 1998; Matthews, Lassiter, & Habedank, 2001; Villar, 2005).
Reading level. To examine whether any observed relationships between psychopathy and cognitive empathy were influenced by individual differences in reading ability, the Word Reading subtest from the Wide Range Achievement Test – Fourth
Edition (WRAT-4; Wilkinson & Robertson, 2006) was used as a
measure of word reading.
Emotional intelligence. To examine whether any observed
relationships between psychopathy and cognitive empathy were
influenced by individual differences in emotional intelligence, we
used the Trait Meta-Mood Scale (TMMS; Salovey, Mayer, Goldman, Turvey, & Palfai, 1995) as a measure of perceived emotional
intelligence. The TMMS has been validated in a variety of clinical
and nonclinical samples (Hodgson & Wertheim, 2007; Salovey,
Stroud, Woolery, & Epel, 2002). In the current sample, TMMS
scores were internally consistent (Cronbach’s alpha ⫽ .84).
p ⫽ .99 scales. In addition, consistent with prior research (Ickes
et al., 1990), participants exhibited greater empathic accuracy on
their second, M ⫽ 6.62, SD ⫽ 1.82, than their first, M ⫽ 5.57, SD
⫽ 1.75, exposure to the same target, t(102) ⫽ 4.67, p ⬍ .01,
consistent with prior findings of greater empathic accuracy with
increasing target familiarity. Finally, in line with prior findings of
greater empathic accuracy for female than male targets (Ickes et
al., 1990), participants exhibited greater empathic accuracy for
female, M ⫽ 61.23, SD ⫽ 14.13, than for male targets, M ⫽ 41.82,
SD ⫽ 15.39, t(102) ⫽ 10.20, p ⬍ .01.5
Psychopathy and Cognitive Empathy
To examine relationships between psychopathy and cognitive
empathy performance, we computed Pearson correlations between
psychopathy total and empathic accuracy task scores. As presented
in Table 1, psychopathy was inversely associated with empathic
accuracy, r ⫽ ⫺.29, p ⬍ .01. The correlation between psychopathy
total scores and the number of responses was marginally significant, r ⫽ ⫺.19, p ⫽ .06. Psychopathy was not associated with
perceived task difficulty.
Psychopathy Factor and Facet Relationships
At the level of psychopathy factors, F2 scores were inversely
associated with empathic accuracy, r ⫽ ⫺.26, p ⫽ .01, and the
number of responses (see Table 1). The relationship between F1
scores and empathic accuracy was marginally significant, r ⫽
⫺.19, p ⫽ .06. The F1 ⫻ F2 interaction did not account for a
significant proportion of variance in empathic accuracy scores, 
⫽ .63, p ⫽ .29. At the facet level (see Table 1), scores on both the
lifestyle and antisocial facets of psychopathy were inversely associated with empathic accuracy scores and the number of responses.
None of the correlations between interpersonal and affective facet
scores and cognitive empathy scores approached significance, and
none of the psychopathy scores were significantly associated with
perceived task difficulty rating scores.
To examine unique associations between psychopathy factor
and facet scores and cognitive empathy scores, we computed
partial correlations between each factor and facet score controlling
for the effects of the other relevant factor or facets. As shown
(above the diagonal) in Table 1, F2 scores were uniquely associated with empathic accuracy, r ⫽ ⫺.21, p ⫽ .03, and the number
of responses. Unique associations between the lifestyle facet and
empathic accuracy scores, r ⫽ ⫺.19, p ⫽ .06, and number of
responses, r ⫽ ⫺.18, p ⫽ .07, were marginally significant.
Results
Construct Validity of the Empathic Accuracy Task
As the empathic accuracy task was developed for the purposes of this study, several sets of preliminary analyses were
conducted to examine its validity as a measure of cognitive
empathy in the current sample. First, we examined the convergent and discriminant validity of empathic accuracy scores by
comparing their associations with various self-report indices of
empathy from the IRI. As a measure of cognitive empathy,
empathic accuracy scores evidenced expected associations with
scores on the PT scale, r ⫽ .21, p ⫽ .04, but not the FA, r ⫽
.14, p ⫽ .15, the EC, r ⫽ ⫺.05, p ⫽ .63, or the PD, r ⫽ ⫺.001,
5
To address target gender effects in the relationship between psychopathy and empathic accuracy, we examined empathic accuracy scores separately for female and male targets. In continuous analyses, total psychopathy scores were significantly associated with empathic accuracy for
female, r ⫽ ⫺.30, p ⬍ .01, but not male, r ⫽ ⫺.14, p ⫽ .16, targets. At
the psychopathy factor level, F1 scores showed significant associations
with empathic accuracy for female, r ⫽ ⫺.31, p ⬍ .01, but not male, r ⫽
.04, p ⫽ .70, targets, and F2 scores correlated significantly with empathic
accuracy for both female, r ⫽ ⫺.20, p ⫽ .05, and male, r ⫽ ⫺.19, p ⫽ .05,
targets. At the level of psychopathy facets, empathic accuracy for female
targets was associated with all four facet scores, interpersonal r ⫽ ⫺.23,
p ⫽ .02, affective r ⫽ ⫺.28, p ⫽ .01, lifestyle r ⫽ ⫺.20, p ⫽ .05, and
antisocial r ⫽ ⫺.20, p ⫽ .05. Conversely, empathic accuracy for male
targets was significantly associated only with the lifestyle features of
psychopathy, r ⫽ ⫺.24, p ⫽ .02.
COGNITIVE EMPATHY AND PSYCHOPATHY
161
Table 1
Intercorrelations Among Psychopathy, Cognitive Empathy, and Background Variables (All ns ⫽ 100 –103)
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
Psychopathy total
Factor 1
Factor 2
Interpersonal facet
Affective facet
Lifestyle facet
Antisocial facet
Empathic accuracy
Number of responses
Perceived task difficulty
Perspective taking
Emotional intelligence
IQ
Reading level
Age
Ethnicity
Education
1
2
3
4
—
.78ⴱⴱ
.84ⴱⴱ
.69ⴱⴱ
.58ⴱⴱ
.74ⴱⴱ
.76ⴱⴱ
⫺.29ⴱⴱ
⫺.19
⫺.05
⫺.34ⴱⴱ
⫺.11
.01
.06
.01
⫺.03
⫺.27ⴱⴱ
—
.35ⴱⴱ
.84ⴱⴱ
.80ⴱⴱ
.35ⴱⴱ
.33ⴱⴱ
⫺.19
.01
⫺.06
⫺.20ⴱ
.02
.07
.15
.13
⫺.06
⫺.09
—
.36ⴱⴱ
.23ⴱ
.87ⴱⴱ
.84ⴱⴱ
⫺.26ⴱⴱ
⫺.28ⴱⴱ
⫺.003
⫺.31ⴱⴱ
⫺.19
⫺.03
.003
⫺.12
.08
⫺.33ⴱⴱ
5
—
.35ⴱⴱ
.33ⴱⴱ
.37ⴱⴱ
⫺.16
.05
⫺.09
⫺.12
.13
.06
.14
.17
⫺.05
.01
6
—
.24ⴱ
.17
⫺.15
⫺.04
⫺.001
⫺.22ⴱ
⫺.10
.06
.11
.02
⫺.04
⫺.16
—
.50ⴱⴱ
⫺.29ⴱⴱ
⫺.24ⴱ
⫺.05
⫺.22ⴱ
⫺.18
⫺.07
.08
⫺.09
.003
⫺.30ⴱⴱ
7
8
9
10
—
⫺.21ⴱ
⫺.20ⴱ
⫺.05
⫺.40ⴱⴱ
⫺.16
.01
⫺.08
⫺.06
.09
⫺.22ⴱ
⫺.10
⫺.21ⴱ
⫺.03
⫺.07
⫺.19
⫺.07
—
.13
⫺.03
.21ⴱⴱ
⫺.04
.04
.05
⫺.14
.05
.15
.12
⫺.30ⴱⴱ
.17
⫺.03
⫺.18
⫺.14
⫺.07
.02
⫺.10
.02
.10
⫺.06
—
.02
.17
⫺.01
.03
.05
.07
⫺.07
.28ⴱⴱ
—
.11
⫺.06
⫺.13
⫺.01
⫺.06
⫺.03
.04
Note. Pearson two-tailed correlations. Zero order correlations are listed below the diagonal. Partial correlations are listed above the diagonal and represent
relationships between empathic accuracy and psychopathy factor and facet scores controlling for the effects of the other relevant factors and facets.
ⴱ
Correlation significant at p ⬍ .05 level. ⴱⴱ Correlation significant at p ⬍ .01 level.
Influence of Background Variables
As presented in Table 1, neither cognitive empathy nor
psychopathy scores were significantly associated with emotional intelligence, estimated IQ, reading level, age, or ethnicity. Conversely, number of years of education was inversely
associated with psychopathy total, F2, lifestyle, and antisocial
facet scores, and positively associated with the number of
responses. After partialing out the effects of education, relationships between empathic accuracy and psychopathy total,
r ⫽ ⫺.26, p ⫽ .01, F2, r ⫽ ⫺.22, p ⫽ .02, and lifestyle facet
scores, r ⫽ ⫺.26, p ⫽ .01, remained significant; however,
correlation with the antisocial facet dropped to trend level, r ⫽
⫺.19, p ⫽ .06. Number of responses remained significantly associated only with F2 scores, r ⫽ ⫺.20, p ⫽ .04; correlations with
the lifestyle facet, r ⫽ ⫺.12, p ⫽ .23, and antisocial facet scores,
r ⫽ ⫺.16, p ⫽ .12 became nonsignificant.
Empathic Accuracy by Emotion Type and Valence
To further explore observed relationships between cognitive
empathy and psychopathy, we computed empathic accuracy
separately for the positively valenced (love, joy, surprise) and
negatively valenced (sadness, fear) primary emotion categories.
Overall, participants exhibited greater empathic accuracy for
positive, M ⫽ 56.39, SD ⫽ 16.72, than for negative emotions,
M ⫽ 48.75, SD ⫽ 12.83, t(102) ⫽ 4.00, p ⬍ .01. Correlations
between psychopathy scores and empathic accuracy scores by
emotion valence and type are presented in Table 2. At the
zero-order level, total psychopathy scores were inversely asso-
Table 2
Relationships Between Psychopathy Scores and Empathic Accuracy for Each Emotion Type and Valence: Zero-Order and Partial
Correlations (All ns ⫽ 101–103)
Psychopathy scores
Factors
F1
Facets
F2
Interpersonal
Affective
Lifestyle
Antisocial
Empathic accuracy
scores
Total
Zero
order
Partial
Zero
order
Partial
Zero
order
Partial
Zero
order
Partial
Zero
order
Partial
Zero
order
Partial
Positive composite
Love
Joy
Surprise
Negative composite
Sadness
Fear
⫺.14
⫺.07
⫺.19
.004
⫺.31**
⫺.22*
⫺.23*
⫺.06
.03
⫺.23*
.13
⫺.24*
⫺.16
⫺.18
⫺.01
.07
⫺.22*
.18
⫺.16
⫺.10
⫺.12
⫺.14
⫺.10
⫺.07
⫺.10
⫺.26**
⫺.19
⫺.19
⫺.13
⫺.12
.01
⫺.15
⫺.20*
⫺.14
⫺.14
⫺.12
⫺.02
⫺.26**
.09
⫺.12
⫺.05
⫺.13
⫺.09
⫺.01
⫺.23*
.10
.07
.11
⫺.02
.04
.08
⫺.11
.13
⫺.28**
⫺.23*
⫺.17
.12
.10
⫺.02
.15
.24*
⫺.22*
⫺.12
⫺.22*
⫺.08
⫺.13
⫺.20*
⫺.21*
⫺.13
⫺.18
⫺.21*
⫺.05
⫺.05
⫺.29**
⫺.07
⫺.003
⫺.09
⫺.08
⫺.10
⫺.08
.02
⫺.26**
⫺.21*
⫺.16
.06
⫺.08
.04
.11
⫺.18
⫺.19
⫺.07
Note. Pearson two-tailed correlations. Partial correlations represent relationships between empathic accuracy and psychopathy factor and facet scores
controlling for the effects of the other relevant factors and facets.
* Correlation significant at p ⬍ .05 level. ** Correlation significant at p ⬍ .01 level.
BROOK AND KOSSON
162
ciated with the negative emotion composite score, r ⫽ ⫺.31, p
⬍ .01, and with scores for the specific negative emotions of
sadness, r ⫽ ⫺.22, p ⫽ .03, and fear, r ⫽ ⫺.23, p ⫽ .02.
Although psychopathy scores were unrelated to positive emotion composite scores, the association with joy scores approached statistical significance, r ⫽ ⫺.19, p ⫽ .06.
Psychopathy factor and facet scores (see Table 2) showed
divergent relationships with empathic accuracy for positive and
negative emotions. At the psychopathy factor score level, F1
was inversely associated with joy scores, r ⫽ ⫺.23, p ⫽ .02,
and negative emotion composite scores, r ⫽ ⫺.24, p ⫽ .02,
with the latter becoming nonsignificant after controlling for the
effects of F2. Conversely, F2 was inversely associated with
negative emotion composite scores, r ⫽ ⫺.26, p ⫽ .01, and
evidenced trend-level relationships with sadness, r ⫽ ⫺.19, p ⫽
.06, and fear scores, r ⫽ ⫺.19, p ⫽ .05, and the correlation with
the negative emotion composite remained significant after controlling for the effects of F1. At the facet level, interpersonal features
of psychopathy were inversely associated only with joy scores, r ⫽
⫺.26, p ⫽ .01. Affective facet scores were inversely associated
with the negative emotion composite, r ⫽ ⫺.28, p ⫽ .01, and
sadness scores, r ⫽ ⫺.23, p ⫽ .02. Lifestyle facet scores were
inversely associated with positive emotion composite, r ⫽ ⫺.22,
p ⫽ .03, and surprise scores, r ⫽ ⫺.20, p ⫽ .04. Antisocial facet
scores were inversely associated with the negative emotion composite, r ⫽ ⫺.26, p ⫽ .01, and sadness scores, r ⫽ ⫺.21, p ⫽ .03.
After partialing out the effects of the other facets, lifestyle facet
scores were no longer associated with the negative composite
score, and associations between antisocial facet and negative emotion composite, r ⫽ ⫺.18, p ⫽ .07, and sadness, r ⫽ ⫺.19, p ⫽
.06, scores dropped to trend level.
Psychopathy Group Analyses
Group analyses using traditional psychopathy total score cutoffs
examined empathic accuracy performance in psychopathic
(PCL-R ⬎ 29; n ⫽ 29), nonpsychopathic (PCL-R ⬍ 21; n ⫽ 25),
and middle-scoring (PCL-R ⫽ 21–29; n ⫽ 49) participants. In
brief, group comparisons revealed significantly worse empathic
accuracy performance in the psychopathic than in the nonpsychopathic group, both overall and for negatively valenced (but not
positively valenced) emotions, and these group differences were
associated with medium to large effect sizes.6
Discussion
The current study is the first, to our knowledge, to demonstrate
impaired cognitive empathy in psychopathy using a seminaturalistic empathic accuracy paradigm. Our findings clearly corroborate
the deficient empathy hypothesis, as evidenced by inverse associations between psychopathy and empathic accuracy scores, as well
as robust group differences between psychopathic and nonpsychopathic inmates, which held independent of demographic factors,
general intellectual and reading abilities, and perceived emotional
intelligence. The current results corroborate and extend prior findings examining the performance of individuals with psychopathic
traits on tasks related to cognitive empathy that found overall
impaired recognition of emotion from morphed affective faces and
speech cues. Both facial and vocal affect recognition are presum-
ably important in naturalistic empathic interaction, and previously
reported impairments in these skills point to possible etiologic
mechanisms underlying performance deficits in our study. However, as outlined in the introduction section, overall deficits are not
typically observed in facial affect recognition paradigms that use
static faces or in ToM tasks. One possible explanation for this
pattern of findings is that the real-time analysis of emotion from
sequences of morphed faces and speech cues and from naturalistic
interactions requires a more complex or dynamic level of emotion
processing than the analysis of emotion from static stimuli. Such
an explanation may be interpreted as consistent with the notion
that psychopaths are capable of hearing the words but not the
music in the full spectrum of affect displayed by a target in a
naturalistic human interaction (Johns & Quay, 1962). In other
words, psychopaths may not consistently evidence deficits in appreciating specific, concrete aspects of human affective displays
but may fail when judging more complex and integrative components of emotion, such as dynamic facial expressions and voice
prosody during naturalistic interpersonal interactions. This notion
is further corroborated by recent evidence suggesting that psychopaths’ emotional processing deficits are most evident in situations
presenting greater stimulus complexity (Sadeh & Verona, 2012).
Psychopathy was also associated with endorsing fewer emotions
in response to others’ emotional displays, suggesting that in addition to being less accurate in judging others’ most salient emotion
in interpersonal interactions, psychopaths also have difficulty appreciating the full spectrum of emotions displayed by people. This
is consistent with Cleckley’s (1941/1988) notion that psychopaths
exhibit a general poverty of affect in interpersonal interactions.
This is a novel finding, as, to our knowledge, no prior study of
cognitive empathy in psychopathy has examined the ability to
recognize emotion in community (nonactor) participants given a
6
Analysis of variance on the empathic accuracy scores revealed a significant effect of psychopathy group, F(2,100) ⫽ 5.67, p ⬍ .01, 2 ⫽ .10.
Follow-up analyses revealed lower accuracy scores in the psychopathic than
the nonpsychopathic group, t(52) ⫽ ⫺3.12, p ⫽ .003, d ⫽ .85. The nonpsychopathic group also had higher empathic accuracy scores than the middlescoring group, t(72) ⫽ 2.68, p ⫽ .009, d ⫽ .64. The difference between the
psychopathic and the middle-scoring group was smaller and non-significant,
t(76) ⫽ ⫺.93, p ⫽ .36, d ⫽ .22. At the level of specific emotion type and
valence, the effect of psychopathy group was significant for the negative
emotion accuracy composite, F(2,100) ⫽ 6.93, p ⫽ .002, 2 ⫽ .12, as well
as for the specific primary emotions of sadness, F(2,100) ⫽ 3.68, p ⫽ .03,
2 ⫽ .07, and fear, F(2,100) ⫽ 3.40, p ⫽ .04, 2 ⫽ .06. Post hoc
comparisons revealed that psychopathic participants had significantly
poorer accuracy than nonpsychopathic participants for the negative emotion composite score t(52) ⫽ ⫺3.38, p ⫽ .001, d ⫽ .92, as well as for fear
t(52) ⫽ ⫺2.27, p ⫽ .03, d ⫽ .62 and sadness t(52) ⫽ ⫺2.61, p ⫽ .01, d
⫽ .72. Nonpsychopaths performed better than middle-scoring participants
on accuracy for the negative emotion composite score, t(72) ⫽ 2.80, p ⫽
.01, d ⫽ .67, and for fear, t(72) ⫽ 2.28, p ⫽ .03, d ⫽ .56, but not for
sadness, t(72) ⫽ 1.64, p ⫽ .11, d ⫽ .41. There were no statistically
significant differences between the psychopathic and the middle-scoring
groups on any of the negative emotion scores, all ts ⬍ 1.49, all ps ⬎ .14.
Likewise, no statistically significant group differences were found for
composite positive emotion accuracy scores, F(2,100) ⫽ 1.26, p ⫽ .29, 2
⫽ .03, or at the level of the specific primary emotions of love, F(2,100) ⫽
1.16, p ⫽ .85, 2 ⬍ .01, joy, F (2,100) ⫽ 2.34, p ⫽ .10, 2 ⫽ .05, or
surprise, F(2,100) ⫽ 1.48, p ⫽ .23, 2 ⫽ .03, despite the medium effect
size for the group difference in empathic accuracy for joy between the
psychopathic and nonpsychopathic groups (d ⫽ .60).
COGNITIVE EMPATHY AND PSYCHOPATHY
seminaturalistic interpersonal interaction or unrestricted response
format.
The finding that poor empathic accuracy performance is most
strongly associated with behavioral/antisocial (F2) and lifestyle
features of psychopathy is intriguing. Theoretically, to the extent
that the interpersonal/affective components of psychopathy represent the personality core of the disorder, psychopaths’ performance
on empathy-related tasks would be expected to correlate most
strongly with F1 scores. In fact, evidence regarding relationships
between specific psychopathy dimensions and indices of emotion
processing has been mixed. Whereas research has linked F1 scores
to psychophysiological (Vanman, Mejia, Dawson, Schell, &
Raine, 2003; Verona, Patrick, Curtin, Bradley, & Lang, 2004) and
lexical decision (Lorenz & Newman, 2002) abnormalities when
processing emotional information, studies have also reported
unique associations between emotional processing indices and
scores on the behavioral/antisocial components of psychopathy
(Dolan & Fullam, 2006; Fullam & Dolan, 2006; Patrick, Cuthbert,
& Lang, 1994; Verona et al., 2004). Although it could be argued
that poor empathic accuracy partially reflects externalizing psychopathology that is not specific to psychopathy, our findings of
significant negative correlations between empathic accuracy and
psychopathy total scores, as well as a marginally significant correlation with F1 scores (and some significant correlations between
scores on interpersonal and affective facets and empathy for
more specific emotion categories), indicate that the associations
observed in this study are not entirely attributable to the lifestyle and antisocial features of psychopathy. Even so, theoretical implications of associations between poor empathic accuracy and the lifestyle features of psychopathy warrant further
research. If cognitive empathy is as crucial in the development
of prosocial interpersonal behavior as some theorists have posited (Preston & de Waal, 2002), it stands to reason that a
developmental lack of cognitive empathy could contribute to
the expression of lifestyle features of psychopathy such as
parasitic and irresponsible behavior.
With regard to the valence of target emotions, our findings are
consistent with prior research suggesting decreased sensitivity to
negatively valenced affective stimuli in psychopathic individuals,
including abnormal startle response to negatively valenced emotional scenes (Levenston, Patrick, Bradley, & Lang, 2000; Patrick,
1994), and deficient recognition of fear and sadness from static and
morphed faces and speech cues. It should be noted, however, that
these findings are not always specific to fear and sadness and are
not entirely consistent across studies (see Introduction for review).
One prominent explanation for psychopaths’ relative insensitivity
to fear and sadness is the violence inhibition model (Blair, 1995,
2001), according to which humans have evolved a cognitiveaffective mechanism that becomes activated when faced with
interpersonal displays of distress, which serves to inhibit violent
behavior and promote development of “moral” decision making,
empathy, and pro-social behavior. Blair has argued that psychopaths are characterized by deficient activation of the violence
inhibition system and should therefore be relatively insensitive to
displays of fear and sadness in others. The violence inhibition
model is partially corroborated by our findings of significant
relationships for negative emotions but not for positive emotions.
However, prior studies have reported affect recognition deficits for
emotions other than fear and sadness (Bagley et al., 2009; see also
163
Pham and Philippot, 2010), and current results revealed marginally
significant associations between psychopathy and accuracy for joy,
as well as significant relationships between interpersonal and
lifestyle features and empathic accuracy for positively valenced
emotions. These findings suggest that the violence inhibition
model, like other theories that posit exclusive deficits in processing
negative affect, cannot explain the entire pattern of findings for
individuals with psychopathic features. The evidence for significant associations between the interpersonal and affective components of psychopathy and valence-specific impairments in cognitive empathy also provide additional evidence that the core
features of psychopathy are related to specific impairments in
cognitive empathy. Such findings highlight the importance of
examining factor and facet scores separately when investigating
relationships between psychopathy and emotion processing.
Finally, findings of no relationship between empathic accuracy
performance and subjective ratings of task difficulty are consistent
with prior research suggesting a discrepancy between objective
indices and subjective perception of emotion processing in psychopathic participants (Brook et al., 2012; Pham & Philippot,
2010).
Limitations
There are several limitations to this research. Because the emphasis in our design was on ecological validity, we did not control
for the length of target vignettes and, as noted above, did not
restrict the number of responses participants could make. Standardizing these factors could have strengthened the internal validity of the task; however, our goal was to design a task resembling
a naturalistic interpersonal interaction. Likewise, our conclusions
regarding specific emotion deficits are limited by the naturalistic
nature of our task, as the target recording comprised unequal
numbers of vignettes portraying specific primary emotions (e.g.,
two vignettes portraying surprise vs. four vignettes portraying
sadness). Further, the absence of any vignettes in which the primary emotion was anger (or the related secondary emotion of
disgust) prevents us from relating our findings to prior research
reporting impaired processing of these emotions in psychopathic
offenders.
Although an attempt was made to include an ethnically diverse
and representative sample of target participants in the stimulus
recording, the final stimulus set did not include any Latino participants. Given the preponderance of evidence from social psychology and affective neuroscience research that interpersonal identification is greatest for members of one’s own racial or ethnic
group, stimulus materials would optimally include a range of
targets resembling the ethnic/racial composition of the potential
perceiver sample. Nonetheless, because only seven participants in
the current sample were Latino, this limitation did not significantly
bias the results (see Footnote #2).
Finally, the generalizability of our findings is limited by our use
of incarcerated male offenders as participants. Whereas the ethnic,
educational, and socioeconomic characteristics of our sample appear representative of the jail and prison population in the United
States, they are not representative of the population at large, and
further research is needed to replicate these findings in community
participants of both sexes.
BROOK AND KOSSON
164
References
Anastassiou-Hadjicharalambous, X., & Warden, D. (2008).
Physiologically-indexed and self-perceived affective empathy in
Conduct-Disordered children high and low on Callous-Unemotional
traits. Child Psychiatry and Human Development, 39, 503–517. doi:10
.1007/s10578-008-0104-y
Bagley, A. D., Abramowitz, C. S., & Kosson, D. S. (2009). Vocal affect
recognition and psychopathy: Converging findings across traditional and
cluster analytic approaches to assessing the construct. Journal of Abnormal Psychology, 118, 388 –398. doi:10.1037/a0015372
Blair, R. J. R. (1995). A cognitive developmental approach to morality:
Investigating the psychopath. Cognition, 57, 1–29. doi:10.1016/00100277(95)00676-P
Blair, R. J. R. (2001). Neurocognitive models of aggression, the antisocial
personality disorders, and psychopathy. Journal of Neurology, Neurosurgery & Psychiatry, 71, 727–731. doi:10.1136/jnnp.71.6.727
Blair, R. J. R. (2005). Responding to the emotions of others: Dissociating
forms of empathy through the study of typical and psychiatric populations. Consciousness and Cognition, 14, 698 –718. doi:10.1016/j.concog
.2005.06.004
Blair, R. J. R. (2006). Dissociable systems for empathy. In G. Bock & J.
Goode (Eds.), Empathy and Fairness (pp. 134 –145). Chichester, UK:
John Wiley and Sons. doi:10.1002/9780470030585.ch10
Blair, R. J. R., Budhani, S., Colledge, E., & Scott, S. (2005). Deafness to
fear in boys with psychopathic tendencies. Journal of Child Psychology
and Psychiatry, 46, 327–336. doi:10.1111/j.1469-7610.2004.00356.x
Blair, R. J. R., & Coles, M. (2000). Expression recognition and behavioural
problems in early adolescence. Cognitive Development, 15, 421– 434.
doi:10.1016/S0885-2014(01)00039-9
Blair, R. J. R., Colledge, E., Murray, L., & Mitchell, D. G. V. (2001). A
selective impairment in the processing of sad and fearful expressions in
children with psychopathic tendencies. Journal of Abnormal Child Psychology, 29, 491– 498. doi:10.1023/A:1012225108281
Blair, R. J. R., Mitchell, D. G. V., Peschardt, K. S., Colledge, E., Leonard,
R. A., Shine, J. H., . . ., & Perrett, D. I. (2004). Reduced sensitivity to
others’ fearful expressions in psychopathic individuals. Personality and
Individual Differences, 37, 1111–1122. doi:10.1016/j.paid.2003.10.008
Blair, R. J. R., Mitchell, D. G. V., Richell, R. A., Kelly, S., Leonard, A.,
Newman, C., & Scott, S. K. (2002). Turning a deaf ear to fear: Impaired
recognition of vocal affect in psychopathic individuals. Journal of Abnormal Psychology, 111, 682– 686. doi:10.1037/0021-843X.111.4.682
Blair, R. J. R., Sellars, C., Strickland, I., & Clark, F. (1996). Theory of
mind in the psychopath. Journal of Forensic Psychiatry, 7, 15–25.
doi:10.1080/09585189608409914
Book, A. S., & Quinsey, V. L. (2004). Psychopaths: Cheaters or warriorhawks? Personality and Individual Differences, 36, 33– 45. doi:10.1016/
S0191-8869(03)00049-7
Book, A. S., Quinsey, V. L., & Langford, D. (2007). Psychopathy and the
perception of affect and vulnerability. Criminal Justice and Behavior,
34, 531–544. doi:10.1177/0093854806293554
Bowers, T. L., & Pantle, M. L. (1998). Shipley Institute for Living Scale and
the Kaufman Brief Intelligence Test as screening instruments for intelligence. Assessment, 5, 187–195. doi:10.1177/107319119800500209
Brook, M., Brieman, C. L., & Kosson, D. S. (2012). Emotion processing in
psychopathy checklist - assessed psychopathy: A review of the literature.
Clinical Psychology Review (under review).
Carr, L., Iacoboni, M., Dubeau, M. C., Mazziotta, J. C., & Lenzi, G. L.
(2003). Neural mechanisms of empathy in humans: A relay from neural
systems for imitation to limbic areas. Proceedings of the National
Academy of Sciences of the United States of America, 100, 5497–5502.
doi:10.1073/pnas.0935845100
Cleckley, H. M. (1941/1988). The Mask of sanity (5th ed.). Augusta, GA:
Emily S. Cleckley.
Cooke, D. J., Kosson, D. S., & Michie, C. (2001). Psychopathy and
ethnicity: Structural, item, and test generalizability of the Psychopathy
Checklist - Revised (PCL-R) in Caucasian and African American participants. Psychological Assessment, 13, 531–542. doi:10.1037/10403590.13.4.531
Cooke, D. J., Michie, C., Hart, S. D., & Clark, D. A. (2004). Reconstructing psychopathy: Clarifying the significance of antisocial and socially
deviant behavior in the diagnosis of psychopathic personality disorder.
Journal of Personality Disorders, 18, 337–357.
Dadds, M. R., Hawes, D. J., Frost, A. D., Vassallo, S., Bunn, P., Hunter,
K., & Merz, S. (2009). Learning to ’talk the talk: The relationship of
psychopathic traits to deficits in empathy across childhood. Journal of
Child Psychology and Psychiatry, 50, 599 – 606. doi:10.1111/j.14697610.2008.02058.x
Davis, M. H. (1983). Measuring individual differences in empathy: Evidence for a multidimensional approach. Journal of Personality and
Social Psychology, 44, 113–126. doi:10.1037/0022-3514.44.1.113
Decety, J., & Jackson, P. L. (2004). The functional architecture of human
empathy. Behavioral and Cognitive Neuroscience Reviews, 3, 71–100.
doi:10.1177/1534582304267187
Decety, J., & Jackson, P. L. (2006). A social-neuroscience perspective on
empathy. Current Directions in Psychological Science, 15, 54 –58. doi:
10.1111/j.0963-7214.2006.00406.x
de Waal, F. B. M. (2005). Our inner ape: A leading primatologist explains
why we are who we are. New York, NY: Riverhead Books.
Dolan, M., & Fullam, R. (2004). Theory of mind and mentalizing ability in
antisocial personality disorders with and without psychopathy. Psychological Medicine, 34, 1093–1102. doi:10.1017/S0033291704002028
Dolan, M., & Fullam, R. (2006). Face affect recognition deficits in
personality-disordered offenders: Association with psychopathy. Psychological Medicine, 36, 1563–1569. doi:10.1017/S0033291706008634
Flight, J. I., & Forth, A. E. (2007). Instrumentally violent youths: The roles
of psychopathic traits, empathy, and attachment. Criminal Justice and
Behavior, 34, 739 –751. doi:10.1177/0093854807299462
Fullam, R., & Dolan, M. (2006). Emotional information processing in
violent patients with schizophrenia: Association with psychopathy and
symptomatology. Psychiatry Research, 141, 29 –37. doi:10.1016/j
.psychres.2005.07.013
Glass, S. J., & Newman, J. P. (2006). Recognition of facial affect in
psychopathic offenders. Journal of Abnormal Psychology, 115, 815–
820. doi:10.1037/0021-843X.115.4.815
Habel, U., Kuhn, E., Salloum, J. B., Devos, H., & Schneider, F. (2002).
Emotional processing in psychopathic personality. Aggressive Behavior,
28, 394 – 400. doi:10.1002/ab.80015
Hall, J. A., & Schmid Mast, M. (2007). Sources of accuracy in the
empathic accuracy paradigm. Emotion, 7, 438 – 446. doi:10.1037/15283542.7.2.438
Hansen, A. L., Johnsen, B. H., Hart, S., Waage, L., & Thayer, J. F. (2008).
Brief communication: Psychopathy and recognition of facial expressions
of emotion. Journal of Personality Disorders, 22, 639 – 645. doi:
10.1521/pedi.2008.22.6.639
Hare, R. D. (1991). The Hare Psychopathy Checklist-Revised. Toronto,
Ontario, Canada: Multi-Health Systems.
Hare, R. D. (1999). Without conscience: The disturbing world of the
psychopaths among us. New York, NY: Guilford Press.
Hare, R. D. (2003). Hare Psychopathy Checklist - Revised (PCL-R) technical manual (2nd ed.). Toronto, Ontario, Canada: Multi-Health Systems
Inc.
Hare, R. D. (2006). Psychopathy: A clinical construct whose time has
come. In C. R. Bartol & A. M. Bartol (Eds.), Current perspectives in
forensic psychology and criminal justice (pp. 107–117). Thousand Oaks,
CA: Sage Publications.
Harpur, T. J., Hare, R. D., & Hakstian, A. R. (1989). Two-factor conceptualization of psychopathy: Construct validity and assessment implica-
COGNITIVE EMPATHY AND PSYCHOPATHY
tions. Psychological Assessment: A Journal of Consulting and Clinical
Psychology, 1, 6 –17. doi:10.1037/1040-3590.1.1.6
Hastings, M. E., Tangney, J. P., & Stuewig, J. (2008). Psychopathy and
identification of facial expressions of emotion. Personality and Individual Differences, 44, 1474 –1483. doi:10.1016/j.paid.2008.01.004
Hatfield, E., Cacioppo, J. T., & Rapson, R. L. (1994). Emotional contagion: Studies in emotion and social interaction. New York, NY: Cambridge University Press.
Heilbrun, A. B. (1982). Cognitive models of criminal violence based upon
intelligence and psychopathy levels. Journal of Consulting and Clinical
Psychology, 50, 546 –557. doi:10.1037/0022-006X.50.4.546
Hemphill, J. F., Hare, R. D., & Wong, S. (1998). Psychopathy and
recidivism: A review. Legal and Criminological Psychology, 3, 139 –
170. doi:10.1111/j.2044-8333.1998.tb00355.x
Hodgson, L. K. & Wertheim, E. H. (2007). Does good emotion management aid forgiving? Multiple dimensions of empathy, emotion management and forgiveness of self and others. Journal of Social and Personal
Relationships, 24, 931–949. doi:10.1177/0265407507084191
Ickes, W. J. (1997). Empathic accuracy. New York, NY: Guilford Publications.
Ickes, W. J., Stinson, L., Bissonnette, V., & Garcia, S. (1990). Naturalistic
social cognition: Empathic accuracy in mixed-sex dyads. Journal of
Personality and Social Psychology, 59, 730 –742. doi:10.1037/00223514.59.4.730
Johns, J. H., & Quay, H. C. (1962). The effect of social reward on verbal
conditioning in psychopathic and neurotic military offenders. Journal of
Consulting Psychology, 26, 217–220. doi:10.1037/h0048399
Jolliffe, D., & Farrington, D. P. (2004). Empathy and offending: A systematic review and meta-analysis. Aggression and Violent Behavior, 9,
441– 476. doi:10.1016/j.avb.2003.03.001
Jones, A. P., Happe, F. G. E., Gilbert, F., Burnett, S., & Viding, E. (2010).
Feeling, caring, knowing: Different types of empathy deficit in boys
with psychopathic tendencies and autism spectrum disorder. Journal of
Child Psychology and Psychiatry, 51, 1188 –1197. doi:10.1111/j.14697610.2010.02280.x
Kosson, D. S., Suchy, Y., Mayer, A. R., & Libby, J. (2002). Facial affect
recognition in criminal psychopaths. Emotion, 2, 398 – 411. doi:10.1037/
1528-3542.2.4.398
Levenston, G. K., Patrick, C. J., Bradley, M. M., & Lang, P. J. (2000). The
psychopath as observer: Emotion and attention in picture processing.
Journal of Abnormal Psychology, 109, 373–385. doi:10.1037/0021843X.109.3.373
Lorenz, A. R., & Newman, J. P. (2002). Deficient response modulation and
emotion processing in low-anxious Caucasian psychopathic offenders:
Results from a lexical decision task. Emotion, 2, 91–104. doi:10.1037/
1528-3542.2.2.91
Matthews, T. D., Lassiter, K., & Habedank, H. (2001). Validity of two
brief measures: The General Abilities Measure for Adults and the
Shipley Institute of Living Scale. Perceptual and Motor Skills,
92(3,Pt1), 881– 887.
Mealey, L. (1995). The sociobiology of sociopathy: An integrated evolutionary model. Behavioral and Brain Sciences, 18, 523–599. doi:10
.1017/S0140525X00039595
Mullins-Nelson, J. L., Salekin, R. T., & Leistico, A. R. (2006). Psychopathy, empathy, and perspective-taking ability in a community sample:
Implications for the successful psychopathy concept. The International
Journal of Forensic Mental Health, 5, 133–149. doi:10.1080/14999013
.2006.10471238
Neumann, C. S., Vitacco, M. J., Hare, R. D., & Wupperman, P. (2005).
Reconstruing the “reconstruction” of psychopathy: A comment on
Cooke, Michie, Hart, and Clark. Journal of Personality Disorders, 19,
624 – 640. doi:10.1521/pedi.2005.19.6.624
165
Nisbett, R. E. & Wilson, T. D. (1977). The halo effect: Evidence for
unconscious alteration of judgments. Journal of Personality and Social
Psychology, 35, 250 –256. doi:10.1037/0022-3514.35.4.250
Pardini, D. A., Lochman, J. E., & Frick, P. J. (2003). Callous/unemotional
traits and social-cognitive processes in adjudicated youths. Journal of
the American Academy of Child & Adolescent Psychiatry, 42, 364 –371.
doi:10.1097/00004583-200303000-00018
Parrott, W. G. (Ed.). (2001). Emotions in social psychology: Essential
readings. New York, NY: Psychology Press.
Patrick, C. J. (1994). Emotion and psychopathy: Startling new insights.
Psychophysiology, 31, 319 –330. doi:10.1111/j.1469-8986.1994
.tb02440.x
Patrick, C. J., Cuthbert, B. N., & Lang, P. J. (1994). Emotion in the
criminal psychopath: Fear image processing. Journal of Abnormal Psychology, 103, 523–534. doi:10.1037/0021-843X.103.3.523
Pham, T. H., & Philippot, P. (2010). Decoding of facial expression of
emotion in criminal psychopaths. Journal of Personality Disorders, 24,
445– 459. doi:10.1521/pedi.2010.24.4.445
Ponnet, K., Buysse, A., Roeyers, H., & De Corte, K. (2005). Empathic
accuracy in adults with a pervasive developmental disorder during an
unstructured conversation with a typically developing stranger. Journal
of Autism and Developmental Disorders, 35, 585– 600. doi:10.1007/
s10803-005-0003-z
Preston, S. D., & de Waal, F. B. M. (2002). Empathy: Its ultimate and
proximate bases. Behavioral and Brain Sciences, 25, 1–20.
Rankin, K. P., Kramer, J. H., & Miller, B. L. (2005). Patterns of cognitive
and emotional empathy in frontotemporal lobar degeneration. Cognitive
and Behavioral Neurology, 18, 28 –36. doi:10.1097/01.wnn.0000152225
.05377.ab
Richell, R. A., Mitchell, D. G. V., Newman, C., Leonard, A., Baron-Cohen,
S., & Blair, R. J. R. (2003). Theory of mind and psychopathy: Can
psychopathic individuals read the ‘language of the eyes’? Neuropsychologia, 41, 523–526. doi:10.1016/S0028-3932(02)00175-6
Sadeh, N., & Verona, E. (2012). Visual complexity attenuates emotional
processing in psychopathy: Implications for fear-potentiated startle deficits. Cognitive, Affective, & Behavioral Neuroscience, 12, 346 –360.
doi:10.3758/s13415-011-0079-1
Salovey, P., Mayer, J. D., Goldman, S. L., Turvey, C., & Palfai, T. P.
(1995). Emotional attention, clarity, and repair: Exploring emotional
intelligence using the trait meta-mood scale. In J. Pennebaker (Ed.),
Emotion, disclosure, & health. Washington, DC: American Psychological Association. doi:10.1037/10182-006
Salovey, P., Stroud, L. R., Woolery, A., & Epel, E. S. (2002). Perceived
emotional intelligence, stress reactivity, and symptom reports: Further
explorations using the trait meta-mood scale. Psychology & Health, 17,
611– 627. doi:10.1080/08870440290025812
Sandoval, A. M., Hancock, D., Poythress, N., Edens, J. F., & Lilienfeld, S.
(2000). Construct validity of the Psychopathic Personality Inventory in
a correctional sample. Journal of Personality Assessment, 74, 262–281.
doi:10.1207/S15327752JPA7402_7
Shamay-Tsoory, S. G., Harari, H., Aharon-Peretz, J., & Levkovitz, Y.
(2010). The role of the orbitofrontal cortex in affective theory of mind
deficits in criminal offenders with psychopathic tendencies. Cortex: A
Journal Devoted to the Study of the Nervous System and Behavior, 46,
668 – 677. doi:10.1016/j.cortex.2009.04.008
Simpson, J. A., Orina, M. M., & Ickes, W. J. (2003). When accuracy hurts,
and when it helps: A test of the empathic accuracy model in marital
interactions. Journal of Personality and Social Psychology, 85, 881–
893. doi:10.1037/0022-3514.85.5.881
Smith, A. M. (2006). Cognitive empathy and emotional empathy in human
behavior and evolution. The Psychological Record, 56, 3–21.
Soderstrom, H. (2003). Psychopathy as a disorder of empathy. European
Child & Adolescent Psychiatry, 12, 249 –252. doi:10.1007/s00787-0030338-y
BROOK AND KOSSON
166
Sprangers, M., Van den Brink, W., Van Heerden, J., & Hoogstraten, J.
(1987). A constructive replication of White’s alleged refutation of Nisbett and Wilson and of Bem: Limitations on verbal reports of internal
events. Journal of Experimental Social Psychology, 23, 302–310. doi:
10.1016/0022-1031(87)90042-4
Stevens, D., Charman, T., & Blair, R. J. R. (2001). Recognition of emotion
in facial expressions and vocal tones in children with psychopathic
tendencies. The Journal of Genetic Psychology: Research and Theory on
Human Development, 162, 201–211. doi:10.1080/00221320109597961
Sullivan, E. A., Abramowitz, C. S., Lopez, M., & Kosson, D. S. (2006).
Reliability and construct validity of the psychopathy checklist-revised
for Latino, European American, and African American male inmates.
Psychological Assessment, 18, 382–392. doi:10.1037/1040-3590.18.4
.382
Szyhowski, D. (2008). A psychometric study of the Shipley Institute of
Living Scale. Dissertation Abstracts International: Section B: The Sciences and Engineering, 69(5-B), 3315.
Vanman, E. J., Mejia, V. Y., Dawson, M. E., Schell, A. M., & Raine, A.
(2003). Modification of the startle reflex in a community sample: Do one
or two dimensions of psychopathy underlie emotional processing? Personality and Individual Differences, 35, 2007–2021. doi:10.1016/S01918869(03)00052-7
Verona, E., Patrick, C. J., Curtin, J. J., Bradley, M. M., & Lang, P. J.
(2004). Psychopathy and physiological response to emotionally evocative sounds. Journal of Abnormal Psychology, 113, 99 –108. doi:10
.1037/0021-843X.113.1.99
Villar, O. M. (2005). Concurrent validity between the Shipley Institute of
Living Scale and the WASI in the assessment of intellectual functioning.
Dissertation Abstracts International: Section B: The Sciences and Engineering, 65(10-B), 5457.
Weiss, J. L., & Schell, R. E. (1991). Estimating WAIS-R IQ from the
Shipley Institute of Living Scale: A replication. Journal of Clinical
Psychology, 47, 558 –562. doi:10.1002/1097-4679(199107)47:4⬍558::
AID-JCLP2270470414⬎3.0.CO;2-W
Wilkinson, G. S., & Robertson, G. J. (2006). Wide Range Achievement Test
4 professional manual. Lutz, FL: Psychological Assessment Resources.
Williams, K. M., & Paulhus, D. L. (2004). Factor structure of the SelfReport Psychopathy scale (SRP-II) in non-forensic samples. Personality
and Individual Differences, 37, 765–778. doi:10.1016/j.paid.2003.11
.004
Zachary, R. A. (1986). Shipley Institute of Living Scale: Revised manual.
Los Angeles, CA: Eastern Psychological Services.
Zágon, I. K., & Jackson, H. J. (1994). Construct validity of a psychopathy
measure. Personality and Individual Differences, 17, 125–135. doi:10
.1016/0191-8869(94)90269-0
Appendix
Descriptive Characteristics of the Emotion Vignettes in the Target Recording
Vignette no
1
2
3
4
5
6
7
8
9
10
11
12
13
Target gender
Target ethnicity
Female
Male
Female
Male
Female
Female
Male
Same
Same
Same
Same
Same
Same
Caucasian
African American
Asian
Caucasian
Caucasian
African American
Asian
vignette # 2
vignette # 3
vignette # 4
vignette # 5
vignette # 6
vignette # 7
target
target
target
target
target
target
as
as
as
as
as
as
Vignette length (seconds)
Target emotion (primary level)
40
78
38
91
54
19
116
81
80
85
70
80
112
Sadness
Sadness
Fear
Fear
Joy
Joy
Surprise (positive)
Love
Fear
Surprise (positive)
Sadness
Sadness
Joy
Received December 27, 2011
Revision received August 9, 2012
Accepted August 15, 2012 䡲