- Email: [email protected]
Memory for emotional events in violent offenders with antisocial personality disorder
Personality and Individual Differences 38 (2005) 1657–1667 www.elsevier.com/locate/paid
Memory for emotional events in violent offenders with antisocial personality disorder Mairead Dolan *, Rachael Fullam Edenfield Centre, Bolton Salford and Trafford Mental Health Partnership, Bury New Road Prestwich, Manchester M25 3BL, UK Received 10 March 2004; received in revised form 8 September 2004; accepted 30 September 2004
Abstract Eight-eight violent offenders meeting the criteria for DSM-IV antisocial personality disorder (ASPD) were assessed using the Psychopathy Checklist-Screening Version (PCL-SV) and compared with 20 healthy controls on an emotional memory task. All participants showed enhanced memory for the emotional phase of the task. On the free and cued recall components of the task high psychopathy scorers showed most impairment on the emotional phase compared with healthy controls. Analyses of psychopathy dimensions indicated no evidence of a dimensional relationship with emotional memory, but subjects categorised as scoring on the extremes of the psychopathy scale particularly those with high emotional detachment (Factor 1) and antisocial behaviour (Factor 2) had impairments in free recall an effect that was attenuated when subjects received cues to assist recall. The results suggest that emotional memory impairments in antisocial populations may be related to a variety of emotional and attentional processes linked with the prefrontallimbic neural circuitry. 2004 Elsevier Ltd. All rights reserved. Keywords: Emotional memory; Violence; Psychopathy; Antisocial personality
*
Corresponding author. Tel.: +44 161 772 3619; fax: +44 161 772 3446. E-mail address: mdolan@edenfield.bstmht.nhs.uk (M. Dolan).
0191-8869/$ - see front matter 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.paid.2004.09.028
1658
M. Dolan, R. Fullam / Personality and Individual Differences 38 (2005) 1657–1667
1. Introduction The antisocial personality disorders (antisocial personality disorder and psychopathy) are a group of overlapping disorders of personality that are associated with significant intra and interpersonal dysfunction. Antisocial Personality Disorder (ASPD) as defined in DSM-IV is more common than the dimensionally defined higher order construct of psychopathy as described by Cleckley (1976) and more recently by Hare (1991). Rates of these disorders are significantly higher in prison and forensic samples than in the general population (Hare, 1991). Half of prison inmates meet DSM-IV criteria for ASPD in the UK (Singleton, Meltzer, & Gatward, 1998). Coid (1992) reported a prevalence rate for psychopathy of 38% in maximum-security psychiatric hospital patients. Psychopathy assessed using HareÕs (1991) criteria is estimated to occur in less than 20% of prisoners with a diagnosis of ASPD (Hare, 1998). Although all these disorders are known to be a major cause of social distress—disruption to families, criminality and violence (Robins, 1990), very little is known about causation. Social factors are associated with antisocial personality disorders (Robins, 1990). However, they also occur in those with apparently normal family backgrounds and there is evidence of genetic liability (McGuffin & Thapar, 1992) suggesting a neurobiological basis for these disorders. There are a number of theories relating to the development of antisocial behaviour, the most prominent of which are the punishment/low fear theories (see Lykken, 1995); the Response Modulation deficit hypothesis (Patterson & Newman, 1993; see Newman, 1998, for a review) and more recently the ‘‘Violence Inhibition Mechanism’’(VIM) deficit proposed by Blair (1995) in which psychopathic behaviours and low empathy are perceived to be related to the failure of basic emotions (e.g. fear) to result in autonomic arousal and the inhibition of ongoing behaviour. Many of these theories focus on the behavioural components of antisocial behaviour and few focus on the interpersonal (low empathy, callousness) components of the disorder. Blair and Frith (2000), however, suggest that the amygdala may be a core component of the neural circuit that mediates the VIM and they have proposed that early amygdala dysfunction may result in the development of core psychopathic (affective–interpersonal) traits. Amygdala lesions in humans reduce the ability to acquire conditioned autonomic responses (Bechara et al., 1995) and impair the capacity to recall emotional material (Cahill, Babinsky, Markowitsch, & McGaugh, 1995). Functional imaging studies also confirm the notion that the amygdala is activated by affectively loaded visual stimuli (Blair, Morris, Frith, Perrett, & Dolan, 1999; Breiter et al., 1996; Morris et al., 1996; Phillips et al., 1997). There is now fairly extensive empirical evidence that psychopaths have attenuated electrodermal responses in anticipation of aversive stimuli (see Hare, 1998, for review) and attenuated startle responses to unpleasant slides (Lang, Bradley, & Cuthbert, 1990; Patrick, Bradley, & Lang, 1993; Levenston, Patrick, Bradley, & Lang, 2000). The latter findings are consistent with a hypothesised deficit in neural circuits involved in the processing of negative affect e.g. amygdala (Davis, 1989). To date most of the research literature on emotional information processing in antisocial samples has focussed on physiological responses to emotional visual stimuli and there has only been one study looking at memory for details of emotional events in psychopathic and
M. Dolan, R. Fullam / Personality and Individual Differences 38 (2005) 1657–1667
1659
non-psychopathic offenders (Christianson et al., 1996). In the latter study, psychopathic offenders did not show the normal memory bias for affectively laden material compared with non-psychopathic offenders. We tested the hypothesis that violent offenders meeting the criteria for ASPD would have impairments in emotional memory compared with healthy controls. We also examined emotional memory in relation to scores on the Psychopathy Checklist-Screening Version (PCL: SV, Hart, Cox, & Hare, 1995) in those rated as High or Low on Factor 1 (emotional detachment) as this measures the core affective traits that may be related to emotional information processing deficits (Patrick, 1994).
2. Method 2.1. Subjects Eighty-eight male, right-handed subjects, aged between 18 and 45 years were recruited from a category B prison and a maximum-security hospital as part of a larger scale study on neuropsychological functioning in DSM-IV criteria for antisocial personality disorder (ASPD, Structured Clinical Interview for Diagnosis [SCID-II] of Axis II disorder, First, Spitzer, Gibbon, & Williams, 1996). Subjects were detained for a minimum of 12 months and had a mean age of 32.33 years (SD 5.75 years) and, the mean number of years spent in education was 12.02 (SD 0.66). Full-scale NART IQ (Nelson, 1982) scores were within the normal range, see Table 1. Psychopathy ratings were made by a single trained rater and were based on file review and semi-structured interview using the Psychopathy Checklist: Screening Version (PCL: SV, Hart et al., 1995). Twenty male healthy volunteers were recruited from ancillary staff working in secure psychiatric hospital settings and prisons. Healthy controls were screened for axis II pathology (SCID-II, First et al., 1996). The mean age of controls was 31.65 years (SD 7.73 years), and mean years of education 12.20 (SD 0.61). All Subjects were screened for current Axis I disorder (SCID-I First et al., 1996), learning disability, significant head injury and drug or alcohol abuse. No subjects were on psychotropic medication, which might have affected neuropsychological test performance.
Table 1 Characteristics of the sample
Age (mean/SD) Years education Full scale IQ PCL:SV Total score PCL:SV Factor 1 PCL:SV Factor 2
Controls (20) Low PCL:SV (26) Medium High F PCL:SV (38) PCL:SV (20)
df
31.65 (7.73) 12.2 (0.63) 106.4 (11.8) – – –
3, 3, 3, 2, 2, 2,
34.15 (5.5) 12.03 (0.69) 102.3 (13.6) 12.3 (2.17) 5.22 (2.17) 7.03 (1.99)
31.79 (5.87) 12.05 (0.80) 101.5 (13.6) 16.34 (1.1) 7.84 (1.7) 8.50 (1.67)
31.20 (5.90) 11.95 (0.22) 107.0 (13.0) 19.65 (0.79) 9.57 (1.46) 10.0 (1.3)
1.14 0.51 1.25 145.12 35.6 18.9
Significance 105 105 105 85 85 85
N/S N/S N/S 0.001 0.001 0.000
1660
M. Dolan, R. Fullam / Personality and Individual Differences 38 (2005) 1657–1667
2.2. Procedure The North West Regions Multi-centre Ethics Committee approved the study. Participants were tested individually in an interview room attached to their ward/wing. The Cahill Emotional Memory task (Cahill & McGaugh, 1995) was completed as part of a larger battery of tests. Recall was assessed after 2 h of distraction using a computerised test battery. 2.2.1. Psychopathy assessment The Psychopathy Checklist: Screening Version (PCL: SV) (Hart et al., 1995) is a 12-item instrument that is quicker and easier to administer than the Psychopathy Checklist-Revised (PCL-R, Hare, 1991). It is conceptually and empirically related to the PCL-R and based on a similar symptom construct scale (Hart et al., 1995). It correlates with the PCL-R at 0.75 (Hart et al., 1995). PCL: SV items are scored as 0 = item not present, 1 = possible presence of item, 3 = item definitely present. Factor 1 of the PCL: SV reflects affective and interpersonal traits and Factor 2 reflects behavioural or social deviance components of psychopathy. As there are no agreed cut-off scores for assignment into ‘‘psychopathic’’ and ‘‘non-psychopathic’’ groups using the PCL: SV we categorised the offender sample into High (>75%), Medium (25–75%) and Low (<25%) scoring groups. 2.2.2. Cahill emotional memory task Emotional memory was assessed using the slide show developed by Cahill and McGaugh (1995), which was a modification of that produced by Heuer and Reisberg (1990). It consists of 11 slides coupled with a one-sentence narrative. The slide show tells the story of a mother and her young son leaving home and walking to the fatherÕs workplace. En route, the boy is struck by a car and taken to the hospital where surgeons successfully re-attach his severed feet. The mother then leaves the hospital to use the phone and collect her other child from school. The first four slides make up the ‘‘non-emotional’’ phase 1 of the task. The next four slides (phase 2) provide the ‘‘emotional arousal’’ component as the car strikes the boy and surgeons struggle to save his life. The final three slides (phase 3) offer a further emotionally neutral stage as the mother makes a phone call and hails a cab. The emotional phase is defined by the content of the slides and the narratives, and not by the emotional tone of the female narrator, which remains neutral across all three phases. Although there is some variation in the choice of slides in the emotional phase of the study across studies we used the method suggested by Cahill and McGaugh (1995). Participants viewed the slide show on a Satellite Pro 4200 series laptop computer with a 15 cm screen. Participants were instructed to watch the slide show and told that the task would be returned to later in the testing session. Memory for the slide show was re-tested 2 h later after completing a separate set of neuropsychological tasks. Initially participants were asked to free recall as much detail as possible relating to the slide show. One point was awarded for each slide correctly recalled. Participants were then asked 76 multiple choice questions (5–9 questions per slide), 29 questions related to phase 1, 28 questions related to phase 2 and 19 questions related to phase 3 of the slide show. Participants were advised that the questions were designed to be difficult but that they should answer every question even if they were forced to guess. The questions followed the order of the slide show and participants were advised each time the questions related to the next slide.
M. Dolan, R. Fullam / Personality and Individual Differences 38 (2005) 1657–1667
1661
2.2.3. Data analysis Data was analysed in SPSS version 9.0 (Chicago Illinois). The emotional story/slide component of the Cahill task was examined using free recall and recognition probes. The free recall components were rated as the number of slides correctly identified. The recognition components were scored out of the number of correctly identified probe questions, which varied between phases [phase 1 maximum 29; phase 2 maximum 28; phase 3 maximum 19]. A phase effect was examined using repeated measures analysis of variance for the whole sample and for individual groups to assess whether free and recognition task performance was better on phase 2 emotional elements of the task compared with phase 1 and 3. Correlations between NART IQ and test performance were examined using SpearmanÕs correlations. Group differences (controls, High, Medium and Low psychopathy scorers) on each phase of the task were examined using analysis of variance (ANOVA) and analysis of co-variance (ANCOVA) with NART IQ as covariate. Posthoc testing was conducted using Bonferroni corrections. In a secondary analysis the antisocial group were categorised into high and low scorers on Factor 1 (emotional detachment) based on the median score on this scale and performance on the task was examined in more detail. Two subjects in the Low psychopathy group did not recall any of the slides from the free recall component of the task so they were removed from the analysis.
3. Results 3.1. Characteristics of the sample The mean PCL: SV score for the offender sample was 15.82 (SD 3.12). Factor 1 (interpersonal/ affective items) mean score was 7.37 (SD 2.45). Factor 2 (social deviance) score was 8.42 (SD 2.01), see Table 1 for group mean scores. There were no significant differences between controls and the antisocial offenders on the NART, see Table 1. The NART score correlated significantly with Cahill total free recall (r = 0.26, n = 106, p < 0.01) and with recognition score (r = 0.43, n = 105, p < 0.001) and was used as a covariate in the analysis. There were no significant correlations between PCL: SV total and sub-factor scores and performance on the Cahill emotional memory test. 3.2. Performance on the Cahill emotional memory task 3.2.1. Recall test Overall, all participants showed a significant phase effect on the free recall task (F [2, 214] = 109.79, p < 0.001). Pair-wise comparisons revealed significant differences between phase 1 and 2 (mean difference 0.60, p < 0.001), phase 2 and 3 (mean difference 1.61, p < 0.001) and phase 1 and 3 (mean difference 1.01, p < 0.001). In general, subjects tended to remember phase 2, i.e. the emotional component of the task more readily. Analyses of each group separately indicated that there were significant differences (paired t test) in recall of phase 1 and 2 [controls (t = 3.03, df 19, p < 0.01); Low psychopathy (t = 1.84, df 26, 0.07); Medium psychopathy (t = 3.55, df 37, p < 0.001) but not in the High psychopathy
1662
M. Dolan, R. Fullam / Personality and Individual Differences 38 (2005) 1657–1667
group (t = 1.69, df 20, p = 0.10). A similar comparison of phase 2 and 3 indicated all 4 groups had better recall for phase 2 than 3 at the p < 0.001 level. A repeated measures ANOVA revealed a significant effect of group (F [3, 102] = 3.6, p < 0.01), phase (F [2, 204] = 105.6, p < 0.001) but no significant phase · group interaction (F [6, 204] = 0.86, p = 0.52). Posthoc testing indicated that the significant group differences were primarily between the healthy control comparison group and the High scoring psychopathic group (mean difference 0.71, p < 0.01). Table 2 shows the mean scores for the free recall component of the task. Analyses of covariance controlling for IQ indicated that there was a trend towards a significant group difference in phase 2 (the free recall of the emotional phase) of the task (p = 0.08) and a significant difference in phase 3 (p < 0.05). There were no significant group differences in phase 1 of the task, see Table 2. Key differences were between High Psychopathy scorers and healthy controls for phase 2. 3.2.2. Recognition On the recognition task a similar effect of phase was seen in the whole sample (F [2, 208] = 261, p < 0.001) with higher rates of recognition in phase 2 compared with phase 1 (paired t test t = 5.4, df 105, p < 0.001) and between phase 2 and 3 (paired t test, t = 22.3, df 104, p < 0.001). Analyses of each of the groups separately indicated that there were significant differences in recognition between phase 1 and 2 in healthy controls (t = 2.1, df 19, p < 0.05), the Low psychopathy group (t = 2.8, df 26, p < 0.01), Medium psychopathy group (t = 2.8, df 37, p < 0.01) and High psychopathy groups (t = 4.4, df 20, p < 0.001). There were significant differences in the recognition of phase 2 and 3 in all groups at the level of p < 0.001. A repeated measures ANOVA with phase and group as factors indicated a significant effect of phase (F [2, 202] = 264.8, p < 0.001), a significant effect of group (F [3, 101] = 4.29, p = 0.007) and a significant group · phase interaction (F [6, 202] = 2.21, p = 0.04). Posthoc testing indicated there were significant differences between controls and the Medium and High scoring psychopathy groups. Group differences in the correct recognition of each phase are shown in Table 2.
Table 2 Cahill emotional memory task Control
Low PCL:SV
Medium PCL:SV
High PCL:SV
F
Significance
Free recall Phase 1 Phase 2 Phase 3
20 2.84 (0.68) 3.47 (0.77) 1.68 (0.67)
27 2.37 (1.1) 2.81 (1.14) 1.55 (1.12)
38 2.23 (1.10) 3.00 (1.06) 1.23 (0.88)
21 2.19 (0.98) 2.66 (0.86)* 0.95 (0.74)*
1.49 2.26 3.9
N/S 0.08 0.05
Recognition Phase 1 Phase 2 Phase 3
19.47 (2.98) 21.10 (2.13) 11.47 (2.73)
16.70 (4.2) 18.17 (3.9) 10.33 (2.86)
15.78 (4.46)* 17.78 (3.23)* 10.78 (2.95)
16.00 (3.33)* 17.85 (2.86)* 11.00 (2.84)
3.7 5.1 0.28
0.05 0.01 N/S
df = (4, 105). * Different from controls, p < 0.05.
M. Dolan, R. Fullam / Personality and Individual Differences 38 (2005) 1657–1667
1663
Analysis of covariance indicated that there were significant group differences in phase 1 (p < 0.05), phase 2 (p < 0.01) but not in phase 3, see Table 2. 3.3. Emotional detachment in violent criminals In a secondary analyses we categorised our antisocial sample into High and Low scorers on Factor 1 (emotional detachment). A repeated measures ANOVA on free recall indicated that there was a significant effect of group [healthy controls, High Factor 1, Low Factor I] (F [2, 103] = 4.22, p < 0.05), a significant effect of phase (F [2, 206] = 101.4, p < 0.001) but no significant phase · group interaction (F [4, 206] = 0.24, N/S). Posthoc testing indicated that the differences were between controls and High Factor 1 scorers and between Low and High Factor 1 scorers (p < 0.05) for phase 2 of the task. A similar analysis on the cued recognition task indicated a significant effect of group (F [2, 102] = 6.2, p < 0.01) a significant effect of phase (F [2, 206] = 256.5, p < 0.001) and a significant group · phase interaction (F [4, 206] = 3.9, p < 0.05). Posthoc testing indicated that both High and Low factor 1 subjects were significantly worse that the healthy control comparison group for both phase 1 and 2 of the task but not phase 3, see Table 3. 3.4. Social deviance A similar analysis on the free recall task based on High and Low Factor 2 score (antisocial behaviour) revealed a significant effect of group (F [2, 102] = 4.6, p < 0.05), significant effect of phase (F [2, 206] = 101.8, p < 0.001) but no group · phase interaction (F [4, 206] = 0.42, N/S). Posthoc testing indicated that the differences were between High and Low Factor 2 scorers and controls. However, group differences were not significant after covarying for IQ, see Table 4. On the cued recall task there was a significant effect of group (F [2, 102] = 6.28, p < 0.01) a significant effect of phase (F [2, 206] = 6.28, p < 0.01) and a significant group · phase interaction (F [4, 206] = 3.40, p < 0.01). Both High and Low factor 2 scorers had lower performance than the healthy comparison group. These differences remained significant for phase I and 2 of the task after controlling for IQ differences, see Table 4. Table 3 High/low emotional detachment Variable
Controls
Low Factor 1
High Factor 1
F
Significance
Recall Phase 1 Phase 2 Phase 3
(20) 2.8 (0.69) 3.50 (0.7) 1.68 (0.6)
(44) 2.27 (1.08) 2.81 (1.14)* 1.25 (0.96)
(42) 2.260 (1.08) 2.90 (0.93)* 1.28 (0.94)
2.19 3.39 1.39
N/S 0.03 N/S
Recognition Phase 1 Phase 2 Phase 3
19.5 (2.9) 21.1 (2.1) 11.4 (2.7)
15.9 (3.9)* 18.18 (3.3)* 10.6 (2.69)
16.35 (4.3)* 18.0 (3.4)* 10.7 (3.0)
5.76 6.77 0.609
0.004 0.002 N/S
df = (3, 105). * Significantly different from controls, p < 0.05.
1664
M. Dolan, R. Fullam / Personality and Individual Differences 38 (2005) 1657–1667
Table 4 High and Low Factor 2 Variable
Controls
Low Factor 2
High Factor 2
F
Significance
Recall Phase 1 Phase 2 Phase 3
19 2.8 (0.68) 3.47 (0.77) 1.68 (0.67)
43 2.27 (1.18) 2.93 (1.07) 1.39 (1.04)
43 2.25 (0.97) 2.7 (1.01)* 1.13 (0.83)*
2.1 2.5 2.6
N/S 0.08 0.07
Recognition Phase 1 Phase 2 Phase 3
19.47 (2.9) 21.10 (2.1) 11.47 (2.7)
16.2 (4.7)* 18.53 (3.9)* 10.37 (3.2)
16.0 (3.4)* 17.6 (2.6)* 11.0 (2.4)
5.2 7.6 0.74
0.007 0.001 N/S
df = (3, 105). * Significantly different from controls, p < 0.05.
4. Discussion This study examines emotional memory in healthy controls and violent antisocial personality disorders using a task previously shown to enhance memory for emotional material in healthy controls (Cahill & McGaugh, 1995). We found a highly significant effect of the emotional phase for all subjects in the free recall component of the task although the relative improvement in free recall was non-significant in the High psychopathy scorers who showed no significant increase in free recall from phase 1 to 2. Overall, our findings support previous reports that emotional arousal influences conscious recall (e.g. Cahill et al., 1995; Heuer & Reisberg, 1990; Yuille & Cutshall, 1986) and that the central details of negative emotional events are better remembered than information preceding or succeeding them (Cahill & McGaugh, 1995; Christianson, 1992; Christianson & Loftus, 1991; Goodman, Hirschman, Hepps, & Rudy, 1991). This effect may be due to the attention capturing nature of a emotional compared with neutral events (Christianson & Safer, 1996; Craik & Blankstein, 1975) or arousal factors (Bradley et al., 1990; Cuthbert, Schupp, Bradley, Birbaumer, & Lang, 2000). Our findings largely fit with Christianson et al.Õs (1996) report that psychopathic and non-psychopathic criminals were more impaired in the recall of emotional material than controls. However, in this study the highly psychopathic group were impaired in free recall, in particular, suggesting that emotional information may not be as readily ‘‘burned into memory’’ in this group perhaps because they are less readily aroused by such scenes. In the cued recall task where memory prompts are given we found that all groups showed a relative improvement in recall between phase 1 and 2 although this effect was attenuated in both Medium and High psychopathy groups compared with healthy controls. It is possible that our selection of a violent offender sample may have reduced the likelihood that we would detect psychopathy-related emotional memory deficits as scenes of injury may cause less arousal in violent offenders. In this study we did not find striking differences in emotional memory when violent offenders were categorised on the emotional detachment factor and both high and low scorers were more impaired than healthy controls on free recall. Although this might suggest that emotional detach-
M. Dolan, R. Fullam / Personality and Individual Differences 38 (2005) 1657–1667
1665
ment does not account for the deficits in emotional memory it is possible that there was insufficient variance in Factor 1 scores to detect significant group differences. Previous studies (e.g. Patrick, 1994) have reported attenuated startle responses to aversive stimuli in highly emotionally detached (High Factor 1) antisocial individuals so future studies should also examine physiological responses to emotionally charged stimuli. In this study we did not find striking differences between high and low Factor 2 scorers on cued recognition although both did worse than controls. We did however, find that only high Factor 2 scores were impaired on free recall. Previous studies have shown evidence of similar memory deficits in impulsive aggressive offenders (Dolan & Anderson, 2002) so it is possible that the impulsivity component of psychopathy also makes a contribution to emotional deficits. Overall, our findings add weight to the notion that the emotional memory deficits seen in violent criminals may be due to a variety of mechanisms. High levels of emotional detachment (Factor 1) may be associated with aversive system dysfunction in which unpleasant slides fail to elicit a normal defensive response disposition (Blair, 1995; Lang et al., 1990). Similarly, high levels of social deviance may be due to higher-level associational deficits linked with impulsivity and attention. In support of the latter notion, Kosson (1996, 1998) showed that psychopaths have left hemisphere related deficits in attention. Future studies will need to further explore the relationship between emotional memory and key dimensional components of psychopathy e.g. empathy, anxiety and impulsivity. There is now a fairly extensive literature indicating that psychopathic criminals have attenuated electro-dermal responses to aversive stimuli (Hare, 1998) and fail to show potentiation of the startle reflex while viewing unpleasant slides (Lang et al., 1990; Patrick et al., 1993). Our findings add to the literature reporting deficits in emotional information processing in violent criminal samples. Studies on emotional information processing suggest a role for the amygdala, ventral striatum and prefrontal cortex in emotional information processing (Cardinal, Parkinson, Hall, & Everitt, 2002). While the ventromedial cortex (Damasio, Grabowski, Frank, Galaburda, & Damasio, 1994; Blair & Cipolotti, 2000) has been implicated in many of the behavioural/social deviance components of the antisocial personality disorders, there is increasing recognition of the potential role of the amygdala in the emotional information processing deficits reported in psychopathic offenders (Blair, 1995). Functional imaging studies indicate that the amygdala is activated in affectively loaded visual stimuli particularly fearful faces (e.g. Morris et al., 1996; Phillips et al., 1997). Ongoing work in the department, using neuroimaging techniques will clarify the neural circuits involved in emotional information processing in antisocial populations.
References 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., & Cipolotti, L. (2000). Impaired social response reversal: A case of Ôacquired sociopathyÕ. Brain, 123, 1122–1141.
1666
M. Dolan, R. Fullam / Personality and Individual Differences 38 (2005) 1657–1667
Blair, J., & Frith, U. (2000). Neurocognitive explanations of the antisocial personality disorders. Criminal Behaviour and Mental Health, 10, 566–581. Blair, R. J., Morris, J. S., Frith, C. D., Perrett, D. I., & Dolan, R. J. (1999). Dissociable neural responses to facial expressions of sadness and anger. Brain, 122, 883–893. Breiter, H. C., Etcoff, N. L., Whalen, P. J., Kennedy, W. A., Rauch, S. L., Buckner, R. L., Strauss, M. M., Hyman, S. E., & Rosen, B. R. (1996). Response and habituation of the human amygdala during visual processing of facial expression. Neuron, 17, 875–887. Cahill, L., Babinsky, R., Markowitsch, H. J., & McGaugh, J. L. (1995). The amygdala and emotional memory. Nature, 377, 295–296. Cahill, L., & McGaugh, J. L. (1995). A novel demonstration of enhanced memory associated with emotional arousal. Consciousness and Cognition, 4, 410–421. Cardinal, R. N., Parkinson, J. A., Hall, J., & Everitt, B. J. (2002). Emotion and motivation: the role of the amygdala, ventral striatum, and prefrontal cortex. Neuroscience and Behavioural Reviews, 26, 321–352. Christianson, S. A. (1992). Emotional stress and eyewitness memory: A critical review. Psychological Bulletin, 112, 284–309. Christianson, S. A., Forth, A. E., Hare, R. D., Strachan, C., Lidberg, L., & Thorell, L. H. (1996). Remembering details of emotional events: a comparison between psychopathic and nonpsychopathic offenders. Personality and Individual Differences, 20, 437–443. Christianson, S. A., & Loftus, E. F. (1991). Remembering emotional events: the fate of detailed information. Cognition and Emotion, 5, 81–108. Christianson, S. A., & Safer, N. A. (1996). Emotional events and emotions in autobiographical memories. In D. C. Rubin (Ed.), Remembering our past: Studies in autobiographical memory. Cambridge: Cambridge University press. Cleckley, H. (1976). The Mask of Sanity (5th ed.). St. Louis, MO: Mosby. Coid, J. W. (1992). DSM-III diagnosis in criminal psychopaths: a way forward. Criminal and Mental Health, 2, 78–94. Craik, K., & Blankstein, K. (1975). Psychophysiology and human memory. In P. H. Venables & M. J. Christie (Eds.), Research in psychophysiology (pp. 389–417). London: Wiley. Cuthbert, B. N., Schupp, H. T., Bradley, M. M., Birbaumer, N., & Lang, P. J. (2000). Brain potentials in affective picture processing: covariation with autonomic arousal and affective report. Biological Psychology, 52, 95–111. Damasio, H., Grabowski, T., Frank, R., Galaburda, A. M., & Damasio, A. R. (1994). The return of Phineas Gage: clues about the brain from the skull of a famous patient. Science, 264, 1102–1105. Davis, M. (1989). Neural systems involved in fear-potentiated startle. Annals of the New York Academy of Sciences, 563, 165–183. Dolan, M., & Anderson, I. M. (2002). Executive and memory function and its relationship to impulsive aggression in personality disordered offenders and healthy controls. Journal of Forensic Psychiatry, 13(3), 503–526. First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J. B. W. (1996). Structured clinical interview for DSM-IV axis I disorders, clinician version (SCID-CV). Washington, DC: American Psychiatric Press, Inc. Goodman, G., Hirschman, J. E., Hepps, D. H., & Rudy, L. (1991). ChildrenÕs memory for stressful events. Merrill Palmer Quarterly, 37, 109–158. Hare, R. D. (1991). The Hare Psychopathy Checklist-Revised. Toronto: Multi-Health Systems. Hare, R. D. (1998). Psychopathy, affect and behaviour. In D. J. Cooke, A. E. Forth, & R. D. Hare (Eds.), Psychopathy: Theory, Research and Implications for society (pp. 105–137). Dordrecht, The Netherlands: Kluwer. Hart, S. D., Cox, D. N., & Hare, R. D. (1995). The Hare Psychopathy Checklist: Screening Version (PCL: SV). Toronto: Multihealth Systems Inc. Heuer, F., & Reisberg, D. (1990). Vivid memories of emotional events: the accuracy of remembered minutiae. Memory and Cognition, 18, 496–506. Kosson, D. S. (1996). Psychopathy and dual-task performance under focussing conditions. Journal of Abnormal Psychology, 105, 391–400. Kosson, D. S. (1998). Divided visual attention in psychopathic and nonpsychopathic offenders. Personality and Individual Differences, 24, 373–391. Lang, P. J., Bradley, M. M., & Cuthbert, B. N. (1990). Emotion, attention, and the startle reflex. Psychological Review, 97, 377–395.
M. Dolan, R. Fullam / Personality and Individual Differences 38 (2005) 1657–1667
1667
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. Lykken, D. T. (1995). The antisocial personalities. Hillsdale, NJ: Erlbaum. McGuffin, P., & Thapar, A. (1992). The genetics of personality disorder. British Journal of Psychiatry, 160, 12–23. Morris, J., Frith, C. D., Perrett, D. I., Rowland, D., Young, A. W., Calder, A. J., & Dolan, R. J. (1996). A differential neural response in the human amygdala to fearful and happy facial expressions. Nature, 383, 812–815. Nelson, H. E. (1982). National Adult Reading Test (NART). Test Manual. NFER-Nelson, Windsor. Newman, J. P. (1998). Psychopathic behaviour: an information processing perspective. In D. J. Cooke, A. E. Forth, & R. D. Hare (Eds.), Psychopathy: Theory, Research and Implications for society. Dordrecht, Netherlands: Kluwer Academic Publishers. 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 criminal 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. Phillips, M. L., Young, A. W., Senior, C., Brammer, M., Andrew, C., Calder, A. J., Bullmore, E. T., Perrett, D. I., Rowland, D., Williams, S. C. R., & David, A. S. (1997). A specific neural substrate for facial expressions of disgust. Nature, 389, 495–498. Robins, L. (1990). Deviant children grown up. Baltimore, MD: Williams & Williams. Singleton, N., Meltzer, M., & Gatward, R. (1998). Psychiatric Morbidity among Prisoners in England and Wales. Government Statistical Service. The Stationary Office: London. Yuille, J. C., & Cutshall, J. L. (1986). A case study of eyewitness memory. Journal of Applied Psychology, 71, 291–301.
Memory for emotional events in violent offenders with antisocial personality disorder Mairead Dolan *, Rachael Fullam Edenfield Centre, Bolton Salford and Trafford Mental Health Partnership, Bury New Road Prestwich, Manchester M25 3BL, UK Received 10 March 2004; received in revised form 8 September 2004; accepted 30 September 2004
Abstract Eight-eight violent offenders meeting the criteria for DSM-IV antisocial personality disorder (ASPD) were assessed using the Psychopathy Checklist-Screening Version (PCL-SV) and compared with 20 healthy controls on an emotional memory task. All participants showed enhanced memory for the emotional phase of the task. On the free and cued recall components of the task high psychopathy scorers showed most impairment on the emotional phase compared with healthy controls. Analyses of psychopathy dimensions indicated no evidence of a dimensional relationship with emotional memory, but subjects categorised as scoring on the extremes of the psychopathy scale particularly those with high emotional detachment (Factor 1) and antisocial behaviour (Factor 2) had impairments in free recall an effect that was attenuated when subjects received cues to assist recall. The results suggest that emotional memory impairments in antisocial populations may be related to a variety of emotional and attentional processes linked with the prefrontallimbic neural circuitry. 2004 Elsevier Ltd. All rights reserved. Keywords: Emotional memory; Violence; Psychopathy; Antisocial personality
*
Corresponding author. Tel.: +44 161 772 3619; fax: +44 161 772 3446. E-mail address: mdolan@edenfield.bstmht.nhs.uk (M. Dolan).
0191-8869/$ - see front matter 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.paid.2004.09.028
1658
M. Dolan, R. Fullam / Personality and Individual Differences 38 (2005) 1657–1667
1. Introduction The antisocial personality disorders (antisocial personality disorder and psychopathy) are a group of overlapping disorders of personality that are associated with significant intra and interpersonal dysfunction. Antisocial Personality Disorder (ASPD) as defined in DSM-IV is more common than the dimensionally defined higher order construct of psychopathy as described by Cleckley (1976) and more recently by Hare (1991). Rates of these disorders are significantly higher in prison and forensic samples than in the general population (Hare, 1991). Half of prison inmates meet DSM-IV criteria for ASPD in the UK (Singleton, Meltzer, & Gatward, 1998). Coid (1992) reported a prevalence rate for psychopathy of 38% in maximum-security psychiatric hospital patients. Psychopathy assessed using HareÕs (1991) criteria is estimated to occur in less than 20% of prisoners with a diagnosis of ASPD (Hare, 1998). Although all these disorders are known to be a major cause of social distress—disruption to families, criminality and violence (Robins, 1990), very little is known about causation. Social factors are associated with antisocial personality disorders (Robins, 1990). However, they also occur in those with apparently normal family backgrounds and there is evidence of genetic liability (McGuffin & Thapar, 1992) suggesting a neurobiological basis for these disorders. There are a number of theories relating to the development of antisocial behaviour, the most prominent of which are the punishment/low fear theories (see Lykken, 1995); the Response Modulation deficit hypothesis (Patterson & Newman, 1993; see Newman, 1998, for a review) and more recently the ‘‘Violence Inhibition Mechanism’’(VIM) deficit proposed by Blair (1995) in which psychopathic behaviours and low empathy are perceived to be related to the failure of basic emotions (e.g. fear) to result in autonomic arousal and the inhibition of ongoing behaviour. Many of these theories focus on the behavioural components of antisocial behaviour and few focus on the interpersonal (low empathy, callousness) components of the disorder. Blair and Frith (2000), however, suggest that the amygdala may be a core component of the neural circuit that mediates the VIM and they have proposed that early amygdala dysfunction may result in the development of core psychopathic (affective–interpersonal) traits. Amygdala lesions in humans reduce the ability to acquire conditioned autonomic responses (Bechara et al., 1995) and impair the capacity to recall emotional material (Cahill, Babinsky, Markowitsch, & McGaugh, 1995). Functional imaging studies also confirm the notion that the amygdala is activated by affectively loaded visual stimuli (Blair, Morris, Frith, Perrett, & Dolan, 1999; Breiter et al., 1996; Morris et al., 1996; Phillips et al., 1997). There is now fairly extensive empirical evidence that psychopaths have attenuated electrodermal responses in anticipation of aversive stimuli (see Hare, 1998, for review) and attenuated startle responses to unpleasant slides (Lang, Bradley, & Cuthbert, 1990; Patrick, Bradley, & Lang, 1993; Levenston, Patrick, Bradley, & Lang, 2000). The latter findings are consistent with a hypothesised deficit in neural circuits involved in the processing of negative affect e.g. amygdala (Davis, 1989). To date most of the research literature on emotional information processing in antisocial samples has focussed on physiological responses to emotional visual stimuli and there has only been one study looking at memory for details of emotional events in psychopathic and
M. Dolan, R. Fullam / Personality and Individual Differences 38 (2005) 1657–1667
1659
non-psychopathic offenders (Christianson et al., 1996). In the latter study, psychopathic offenders did not show the normal memory bias for affectively laden material compared with non-psychopathic offenders. We tested the hypothesis that violent offenders meeting the criteria for ASPD would have impairments in emotional memory compared with healthy controls. We also examined emotional memory in relation to scores on the Psychopathy Checklist-Screening Version (PCL: SV, Hart, Cox, & Hare, 1995) in those rated as High or Low on Factor 1 (emotional detachment) as this measures the core affective traits that may be related to emotional information processing deficits (Patrick, 1994).
2. Method 2.1. Subjects Eighty-eight male, right-handed subjects, aged between 18 and 45 years were recruited from a category B prison and a maximum-security hospital as part of a larger scale study on neuropsychological functioning in DSM-IV criteria for antisocial personality disorder (ASPD, Structured Clinical Interview for Diagnosis [SCID-II] of Axis II disorder, First, Spitzer, Gibbon, & Williams, 1996). Subjects were detained for a minimum of 12 months and had a mean age of 32.33 years (SD 5.75 years) and, the mean number of years spent in education was 12.02 (SD 0.66). Full-scale NART IQ (Nelson, 1982) scores were within the normal range, see Table 1. Psychopathy ratings were made by a single trained rater and were based on file review and semi-structured interview using the Psychopathy Checklist: Screening Version (PCL: SV, Hart et al., 1995). Twenty male healthy volunteers were recruited from ancillary staff working in secure psychiatric hospital settings and prisons. Healthy controls were screened for axis II pathology (SCID-II, First et al., 1996). The mean age of controls was 31.65 years (SD 7.73 years), and mean years of education 12.20 (SD 0.61). All Subjects were screened for current Axis I disorder (SCID-I First et al., 1996), learning disability, significant head injury and drug or alcohol abuse. No subjects were on psychotropic medication, which might have affected neuropsychological test performance.
Table 1 Characteristics of the sample
Age (mean/SD) Years education Full scale IQ PCL:SV Total score PCL:SV Factor 1 PCL:SV Factor 2
Controls (20) Low PCL:SV (26) Medium High F PCL:SV (38) PCL:SV (20)
df
31.65 (7.73) 12.2 (0.63) 106.4 (11.8) – – –
3, 3, 3, 2, 2, 2,
34.15 (5.5) 12.03 (0.69) 102.3 (13.6) 12.3 (2.17) 5.22 (2.17) 7.03 (1.99)
31.79 (5.87) 12.05 (0.80) 101.5 (13.6) 16.34 (1.1) 7.84 (1.7) 8.50 (1.67)
31.20 (5.90) 11.95 (0.22) 107.0 (13.0) 19.65 (0.79) 9.57 (1.46) 10.0 (1.3)
1.14 0.51 1.25 145.12 35.6 18.9
Significance 105 105 105 85 85 85
N/S N/S N/S 0.001 0.001 0.000
1660
M. Dolan, R. Fullam / Personality and Individual Differences 38 (2005) 1657–1667
2.2. Procedure The North West Regions Multi-centre Ethics Committee approved the study. Participants were tested individually in an interview room attached to their ward/wing. The Cahill Emotional Memory task (Cahill & McGaugh, 1995) was completed as part of a larger battery of tests. Recall was assessed after 2 h of distraction using a computerised test battery. 2.2.1. Psychopathy assessment The Psychopathy Checklist: Screening Version (PCL: SV) (Hart et al., 1995) is a 12-item instrument that is quicker and easier to administer than the Psychopathy Checklist-Revised (PCL-R, Hare, 1991). It is conceptually and empirically related to the PCL-R and based on a similar symptom construct scale (Hart et al., 1995). It correlates with the PCL-R at 0.75 (Hart et al., 1995). PCL: SV items are scored as 0 = item not present, 1 = possible presence of item, 3 = item definitely present. Factor 1 of the PCL: SV reflects affective and interpersonal traits and Factor 2 reflects behavioural or social deviance components of psychopathy. As there are no agreed cut-off scores for assignment into ‘‘psychopathic’’ and ‘‘non-psychopathic’’ groups using the PCL: SV we categorised the offender sample into High (>75%), Medium (25–75%) and Low (<25%) scoring groups. 2.2.2. Cahill emotional memory task Emotional memory was assessed using the slide show developed by Cahill and McGaugh (1995), which was a modification of that produced by Heuer and Reisberg (1990). It consists of 11 slides coupled with a one-sentence narrative. The slide show tells the story of a mother and her young son leaving home and walking to the fatherÕs workplace. En route, the boy is struck by a car and taken to the hospital where surgeons successfully re-attach his severed feet. The mother then leaves the hospital to use the phone and collect her other child from school. The first four slides make up the ‘‘non-emotional’’ phase 1 of the task. The next four slides (phase 2) provide the ‘‘emotional arousal’’ component as the car strikes the boy and surgeons struggle to save his life. The final three slides (phase 3) offer a further emotionally neutral stage as the mother makes a phone call and hails a cab. The emotional phase is defined by the content of the slides and the narratives, and not by the emotional tone of the female narrator, which remains neutral across all three phases. Although there is some variation in the choice of slides in the emotional phase of the study across studies we used the method suggested by Cahill and McGaugh (1995). Participants viewed the slide show on a Satellite Pro 4200 series laptop computer with a 15 cm screen. Participants were instructed to watch the slide show and told that the task would be returned to later in the testing session. Memory for the slide show was re-tested 2 h later after completing a separate set of neuropsychological tasks. Initially participants were asked to free recall as much detail as possible relating to the slide show. One point was awarded for each slide correctly recalled. Participants were then asked 76 multiple choice questions (5–9 questions per slide), 29 questions related to phase 1, 28 questions related to phase 2 and 19 questions related to phase 3 of the slide show. Participants were advised that the questions were designed to be difficult but that they should answer every question even if they were forced to guess. The questions followed the order of the slide show and participants were advised each time the questions related to the next slide.
M. Dolan, R. Fullam / Personality and Individual Differences 38 (2005) 1657–1667
1661
2.2.3. Data analysis Data was analysed in SPSS version 9.0 (Chicago Illinois). The emotional story/slide component of the Cahill task was examined using free recall and recognition probes. The free recall components were rated as the number of slides correctly identified. The recognition components were scored out of the number of correctly identified probe questions, which varied between phases [phase 1 maximum 29; phase 2 maximum 28; phase 3 maximum 19]. A phase effect was examined using repeated measures analysis of variance for the whole sample and for individual groups to assess whether free and recognition task performance was better on phase 2 emotional elements of the task compared with phase 1 and 3. Correlations between NART IQ and test performance were examined using SpearmanÕs correlations. Group differences (controls, High, Medium and Low psychopathy scorers) on each phase of the task were examined using analysis of variance (ANOVA) and analysis of co-variance (ANCOVA) with NART IQ as covariate. Posthoc testing was conducted using Bonferroni corrections. In a secondary analysis the antisocial group were categorised into high and low scorers on Factor 1 (emotional detachment) based on the median score on this scale and performance on the task was examined in more detail. Two subjects in the Low psychopathy group did not recall any of the slides from the free recall component of the task so they were removed from the analysis.
3. Results 3.1. Characteristics of the sample The mean PCL: SV score for the offender sample was 15.82 (SD 3.12). Factor 1 (interpersonal/ affective items) mean score was 7.37 (SD 2.45). Factor 2 (social deviance) score was 8.42 (SD 2.01), see Table 1 for group mean scores. There were no significant differences between controls and the antisocial offenders on the NART, see Table 1. The NART score correlated significantly with Cahill total free recall (r = 0.26, n = 106, p < 0.01) and with recognition score (r = 0.43, n = 105, p < 0.001) and was used as a covariate in the analysis. There were no significant correlations between PCL: SV total and sub-factor scores and performance on the Cahill emotional memory test. 3.2. Performance on the Cahill emotional memory task 3.2.1. Recall test Overall, all participants showed a significant phase effect on the free recall task (F [2, 214] = 109.79, p < 0.001). Pair-wise comparisons revealed significant differences between phase 1 and 2 (mean difference 0.60, p < 0.001), phase 2 and 3 (mean difference 1.61, p < 0.001) and phase 1 and 3 (mean difference 1.01, p < 0.001). In general, subjects tended to remember phase 2, i.e. the emotional component of the task more readily. Analyses of each group separately indicated that there were significant differences (paired t test) in recall of phase 1 and 2 [controls (t = 3.03, df 19, p < 0.01); Low psychopathy (t = 1.84, df 26, 0.07); Medium psychopathy (t = 3.55, df 37, p < 0.001) but not in the High psychopathy
1662
M. Dolan, R. Fullam / Personality and Individual Differences 38 (2005) 1657–1667
group (t = 1.69, df 20, p = 0.10). A similar comparison of phase 2 and 3 indicated all 4 groups had better recall for phase 2 than 3 at the p < 0.001 level. A repeated measures ANOVA revealed a significant effect of group (F [3, 102] = 3.6, p < 0.01), phase (F [2, 204] = 105.6, p < 0.001) but no significant phase · group interaction (F [6, 204] = 0.86, p = 0.52). Posthoc testing indicated that the significant group differences were primarily between the healthy control comparison group and the High scoring psychopathic group (mean difference 0.71, p < 0.01). Table 2 shows the mean scores for the free recall component of the task. Analyses of covariance controlling for IQ indicated that there was a trend towards a significant group difference in phase 2 (the free recall of the emotional phase) of the task (p = 0.08) and a significant difference in phase 3 (p < 0.05). There were no significant group differences in phase 1 of the task, see Table 2. Key differences were between High Psychopathy scorers and healthy controls for phase 2. 3.2.2. Recognition On the recognition task a similar effect of phase was seen in the whole sample (F [2, 208] = 261, p < 0.001) with higher rates of recognition in phase 2 compared with phase 1 (paired t test t = 5.4, df 105, p < 0.001) and between phase 2 and 3 (paired t test, t = 22.3, df 104, p < 0.001). Analyses of each of the groups separately indicated that there were significant differences in recognition between phase 1 and 2 in healthy controls (t = 2.1, df 19, p < 0.05), the Low psychopathy group (t = 2.8, df 26, p < 0.01), Medium psychopathy group (t = 2.8, df 37, p < 0.01) and High psychopathy groups (t = 4.4, df 20, p < 0.001). There were significant differences in the recognition of phase 2 and 3 in all groups at the level of p < 0.001. A repeated measures ANOVA with phase and group as factors indicated a significant effect of phase (F [2, 202] = 264.8, p < 0.001), a significant effect of group (F [3, 101] = 4.29, p = 0.007) and a significant group · phase interaction (F [6, 202] = 2.21, p = 0.04). Posthoc testing indicated there were significant differences between controls and the Medium and High scoring psychopathy groups. Group differences in the correct recognition of each phase are shown in Table 2.
Table 2 Cahill emotional memory task Control
Low PCL:SV
Medium PCL:SV
High PCL:SV
F
Significance
Free recall Phase 1 Phase 2 Phase 3
20 2.84 (0.68) 3.47 (0.77) 1.68 (0.67)
27 2.37 (1.1) 2.81 (1.14) 1.55 (1.12)
38 2.23 (1.10) 3.00 (1.06) 1.23 (0.88)
21 2.19 (0.98) 2.66 (0.86)* 0.95 (0.74)*
1.49 2.26 3.9
N/S 0.08 0.05
Recognition Phase 1 Phase 2 Phase 3
19.47 (2.98) 21.10 (2.13) 11.47 (2.73)
16.70 (4.2) 18.17 (3.9) 10.33 (2.86)
15.78 (4.46)* 17.78 (3.23)* 10.78 (2.95)
16.00 (3.33)* 17.85 (2.86)* 11.00 (2.84)
3.7 5.1 0.28
0.05 0.01 N/S
df = (4, 105). * Different from controls, p < 0.05.
M. Dolan, R. Fullam / Personality and Individual Differences 38 (2005) 1657–1667
1663
Analysis of covariance indicated that there were significant group differences in phase 1 (p < 0.05), phase 2 (p < 0.01) but not in phase 3, see Table 2. 3.3. Emotional detachment in violent criminals In a secondary analyses we categorised our antisocial sample into High and Low scorers on Factor 1 (emotional detachment). A repeated measures ANOVA on free recall indicated that there was a significant effect of group [healthy controls, High Factor 1, Low Factor I] (F [2, 103] = 4.22, p < 0.05), a significant effect of phase (F [2, 206] = 101.4, p < 0.001) but no significant phase · group interaction (F [4, 206] = 0.24, N/S). Posthoc testing indicated that the differences were between controls and High Factor 1 scorers and between Low and High Factor 1 scorers (p < 0.05) for phase 2 of the task. A similar analysis on the cued recognition task indicated a significant effect of group (F [2, 102] = 6.2, p < 0.01) a significant effect of phase (F [2, 206] = 256.5, p < 0.001) and a significant group · phase interaction (F [4, 206] = 3.9, p < 0.05). Posthoc testing indicated that both High and Low factor 1 subjects were significantly worse that the healthy control comparison group for both phase 1 and 2 of the task but not phase 3, see Table 3. 3.4. Social deviance A similar analysis on the free recall task based on High and Low Factor 2 score (antisocial behaviour) revealed a significant effect of group (F [2, 102] = 4.6, p < 0.05), significant effect of phase (F [2, 206] = 101.8, p < 0.001) but no group · phase interaction (F [4, 206] = 0.42, N/S). Posthoc testing indicated that the differences were between High and Low Factor 2 scorers and controls. However, group differences were not significant after covarying for IQ, see Table 4. On the cued recall task there was a significant effect of group (F [2, 102] = 6.28, p < 0.01) a significant effect of phase (F [2, 206] = 6.28, p < 0.01) and a significant group · phase interaction (F [4, 206] = 3.40, p < 0.01). Both High and Low factor 2 scorers had lower performance than the healthy comparison group. These differences remained significant for phase I and 2 of the task after controlling for IQ differences, see Table 4. Table 3 High/low emotional detachment Variable
Controls
Low Factor 1
High Factor 1
F
Significance
Recall Phase 1 Phase 2 Phase 3
(20) 2.8 (0.69) 3.50 (0.7) 1.68 (0.6)
(44) 2.27 (1.08) 2.81 (1.14)* 1.25 (0.96)
(42) 2.260 (1.08) 2.90 (0.93)* 1.28 (0.94)
2.19 3.39 1.39
N/S 0.03 N/S
Recognition Phase 1 Phase 2 Phase 3
19.5 (2.9) 21.1 (2.1) 11.4 (2.7)
15.9 (3.9)* 18.18 (3.3)* 10.6 (2.69)
16.35 (4.3)* 18.0 (3.4)* 10.7 (3.0)
5.76 6.77 0.609
0.004 0.002 N/S
df = (3, 105). * Significantly different from controls, p < 0.05.
1664
M. Dolan, R. Fullam / Personality and Individual Differences 38 (2005) 1657–1667
Table 4 High and Low Factor 2 Variable
Controls
Low Factor 2
High Factor 2
F
Significance
Recall Phase 1 Phase 2 Phase 3
19 2.8 (0.68) 3.47 (0.77) 1.68 (0.67)
43 2.27 (1.18) 2.93 (1.07) 1.39 (1.04)
43 2.25 (0.97) 2.7 (1.01)* 1.13 (0.83)*
2.1 2.5 2.6
N/S 0.08 0.07
Recognition Phase 1 Phase 2 Phase 3
19.47 (2.9) 21.10 (2.1) 11.47 (2.7)
16.2 (4.7)* 18.53 (3.9)* 10.37 (3.2)
16.0 (3.4)* 17.6 (2.6)* 11.0 (2.4)
5.2 7.6 0.74
0.007 0.001 N/S
df = (3, 105). * Significantly different from controls, p < 0.05.
4. Discussion This study examines emotional memory in healthy controls and violent antisocial personality disorders using a task previously shown to enhance memory for emotional material in healthy controls (Cahill & McGaugh, 1995). We found a highly significant effect of the emotional phase for all subjects in the free recall component of the task although the relative improvement in free recall was non-significant in the High psychopathy scorers who showed no significant increase in free recall from phase 1 to 2. Overall, our findings support previous reports that emotional arousal influences conscious recall (e.g. Cahill et al., 1995; Heuer & Reisberg, 1990; Yuille & Cutshall, 1986) and that the central details of negative emotional events are better remembered than information preceding or succeeding them (Cahill & McGaugh, 1995; Christianson, 1992; Christianson & Loftus, 1991; Goodman, Hirschman, Hepps, & Rudy, 1991). This effect may be due to the attention capturing nature of a emotional compared with neutral events (Christianson & Safer, 1996; Craik & Blankstein, 1975) or arousal factors (Bradley et al., 1990; Cuthbert, Schupp, Bradley, Birbaumer, & Lang, 2000). Our findings largely fit with Christianson et al.Õs (1996) report that psychopathic and non-psychopathic criminals were more impaired in the recall of emotional material than controls. However, in this study the highly psychopathic group were impaired in free recall, in particular, suggesting that emotional information may not be as readily ‘‘burned into memory’’ in this group perhaps because they are less readily aroused by such scenes. In the cued recall task where memory prompts are given we found that all groups showed a relative improvement in recall between phase 1 and 2 although this effect was attenuated in both Medium and High psychopathy groups compared with healthy controls. It is possible that our selection of a violent offender sample may have reduced the likelihood that we would detect psychopathy-related emotional memory deficits as scenes of injury may cause less arousal in violent offenders. In this study we did not find striking differences in emotional memory when violent offenders were categorised on the emotional detachment factor and both high and low scorers were more impaired than healthy controls on free recall. Although this might suggest that emotional detach-
M. Dolan, R. Fullam / Personality and Individual Differences 38 (2005) 1657–1667
1665
ment does not account for the deficits in emotional memory it is possible that there was insufficient variance in Factor 1 scores to detect significant group differences. Previous studies (e.g. Patrick, 1994) have reported attenuated startle responses to aversive stimuli in highly emotionally detached (High Factor 1) antisocial individuals so future studies should also examine physiological responses to emotionally charged stimuli. In this study we did not find striking differences between high and low Factor 2 scorers on cued recognition although both did worse than controls. We did however, find that only high Factor 2 scores were impaired on free recall. Previous studies have shown evidence of similar memory deficits in impulsive aggressive offenders (Dolan & Anderson, 2002) so it is possible that the impulsivity component of psychopathy also makes a contribution to emotional deficits. Overall, our findings add weight to the notion that the emotional memory deficits seen in violent criminals may be due to a variety of mechanisms. High levels of emotional detachment (Factor 1) may be associated with aversive system dysfunction in which unpleasant slides fail to elicit a normal defensive response disposition (Blair, 1995; Lang et al., 1990). Similarly, high levels of social deviance may be due to higher-level associational deficits linked with impulsivity and attention. In support of the latter notion, Kosson (1996, 1998) showed that psychopaths have left hemisphere related deficits in attention. Future studies will need to further explore the relationship between emotional memory and key dimensional components of psychopathy e.g. empathy, anxiety and impulsivity. There is now a fairly extensive literature indicating that psychopathic criminals have attenuated electro-dermal responses to aversive stimuli (Hare, 1998) and fail to show potentiation of the startle reflex while viewing unpleasant slides (Lang et al., 1990; Patrick et al., 1993). Our findings add to the literature reporting deficits in emotional information processing in violent criminal samples. Studies on emotional information processing suggest a role for the amygdala, ventral striatum and prefrontal cortex in emotional information processing (Cardinal, Parkinson, Hall, & Everitt, 2002). While the ventromedial cortex (Damasio, Grabowski, Frank, Galaburda, & Damasio, 1994; Blair & Cipolotti, 2000) has been implicated in many of the behavioural/social deviance components of the antisocial personality disorders, there is increasing recognition of the potential role of the amygdala in the emotional information processing deficits reported in psychopathic offenders (Blair, 1995). Functional imaging studies indicate that the amygdala is activated in affectively loaded visual stimuli particularly fearful faces (e.g. Morris et al., 1996; Phillips et al., 1997). Ongoing work in the department, using neuroimaging techniques will clarify the neural circuits involved in emotional information processing in antisocial populations.
References 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., & Cipolotti, L. (2000). Impaired social response reversal: A case of Ôacquired sociopathyÕ. Brain, 123, 1122–1141.
1666
M. Dolan, R. Fullam / Personality and Individual Differences 38 (2005) 1657–1667
Blair, J., & Frith, U. (2000). Neurocognitive explanations of the antisocial personality disorders. Criminal Behaviour and Mental Health, 10, 566–581. Blair, R. J., Morris, J. S., Frith, C. D., Perrett, D. I., & Dolan, R. J. (1999). Dissociable neural responses to facial expressions of sadness and anger. Brain, 122, 883–893. Breiter, H. C., Etcoff, N. L., Whalen, P. J., Kennedy, W. A., Rauch, S. L., Buckner, R. L., Strauss, M. M., Hyman, S. E., & Rosen, B. R. (1996). Response and habituation of the human amygdala during visual processing of facial expression. Neuron, 17, 875–887. Cahill, L., Babinsky, R., Markowitsch, H. J., & McGaugh, J. L. (1995). The amygdala and emotional memory. Nature, 377, 295–296. Cahill, L., & McGaugh, J. L. (1995). A novel demonstration of enhanced memory associated with emotional arousal. Consciousness and Cognition, 4, 410–421. Cardinal, R. N., Parkinson, J. A., Hall, J., & Everitt, B. J. (2002). Emotion and motivation: the role of the amygdala, ventral striatum, and prefrontal cortex. Neuroscience and Behavioural Reviews, 26, 321–352. Christianson, S. A. (1992). Emotional stress and eyewitness memory: A critical review. Psychological Bulletin, 112, 284–309. Christianson, S. A., Forth, A. E., Hare, R. D., Strachan, C., Lidberg, L., & Thorell, L. H. (1996). Remembering details of emotional events: a comparison between psychopathic and nonpsychopathic offenders. Personality and Individual Differences, 20, 437–443. Christianson, S. A., & Loftus, E. F. (1991). Remembering emotional events: the fate of detailed information. Cognition and Emotion, 5, 81–108. Christianson, S. A., & Safer, N. A. (1996). Emotional events and emotions in autobiographical memories. In D. C. Rubin (Ed.), Remembering our past: Studies in autobiographical memory. Cambridge: Cambridge University press. Cleckley, H. (1976). The Mask of Sanity (5th ed.). St. Louis, MO: Mosby. Coid, J. W. (1992). DSM-III diagnosis in criminal psychopaths: a way forward. Criminal and Mental Health, 2, 78–94. Craik, K., & Blankstein, K. (1975). Psychophysiology and human memory. In P. H. Venables & M. J. Christie (Eds.), Research in psychophysiology (pp. 389–417). London: Wiley. Cuthbert, B. N., Schupp, H. T., Bradley, M. M., Birbaumer, N., & Lang, P. J. (2000). Brain potentials in affective picture processing: covariation with autonomic arousal and affective report. Biological Psychology, 52, 95–111. Damasio, H., Grabowski, T., Frank, R., Galaburda, A. M., & Damasio, A. R. (1994). The return of Phineas Gage: clues about the brain from the skull of a famous patient. Science, 264, 1102–1105. Davis, M. (1989). Neural systems involved in fear-potentiated startle. Annals of the New York Academy of Sciences, 563, 165–183. Dolan, M., & Anderson, I. M. (2002). Executive and memory function and its relationship to impulsive aggression in personality disordered offenders and healthy controls. Journal of Forensic Psychiatry, 13(3), 503–526. First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J. B. W. (1996). Structured clinical interview for DSM-IV axis I disorders, clinician version (SCID-CV). Washington, DC: American Psychiatric Press, Inc. Goodman, G., Hirschman, J. E., Hepps, D. H., & Rudy, L. (1991). ChildrenÕs memory for stressful events. Merrill Palmer Quarterly, 37, 109–158. Hare, R. D. (1991). The Hare Psychopathy Checklist-Revised. Toronto: Multi-Health Systems. Hare, R. D. (1998). Psychopathy, affect and behaviour. In D. J. Cooke, A. E. Forth, & R. D. Hare (Eds.), Psychopathy: Theory, Research and Implications for society (pp. 105–137). Dordrecht, The Netherlands: Kluwer. Hart, S. D., Cox, D. N., & Hare, R. D. (1995). The Hare Psychopathy Checklist: Screening Version (PCL: SV). Toronto: Multihealth Systems Inc. Heuer, F., & Reisberg, D. (1990). Vivid memories of emotional events: the accuracy of remembered minutiae. Memory and Cognition, 18, 496–506. Kosson, D. S. (1996). Psychopathy and dual-task performance under focussing conditions. Journal of Abnormal Psychology, 105, 391–400. Kosson, D. S. (1998). Divided visual attention in psychopathic and nonpsychopathic offenders. Personality and Individual Differences, 24, 373–391. Lang, P. J., Bradley, M. M., & Cuthbert, B. N. (1990). Emotion, attention, and the startle reflex. Psychological Review, 97, 377–395.
M. Dolan, R. Fullam / Personality and Individual Differences 38 (2005) 1657–1667
1667
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. Lykken, D. T. (1995). The antisocial personalities. Hillsdale, NJ: Erlbaum. McGuffin, P., & Thapar, A. (1992). The genetics of personality disorder. British Journal of Psychiatry, 160, 12–23. Morris, J., Frith, C. D., Perrett, D. I., Rowland, D., Young, A. W., Calder, A. J., & Dolan, R. J. (1996). A differential neural response in the human amygdala to fearful and happy facial expressions. Nature, 383, 812–815. Nelson, H. E. (1982). National Adult Reading Test (NART). Test Manual. NFER-Nelson, Windsor. Newman, J. P. (1998). Psychopathic behaviour: an information processing perspective. In D. J. Cooke, A. E. Forth, & R. D. Hare (Eds.), Psychopathy: Theory, Research and Implications for society. Dordrecht, Netherlands: Kluwer Academic Publishers. 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 criminal 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. Phillips, M. L., Young, A. W., Senior, C., Brammer, M., Andrew, C., Calder, A. J., Bullmore, E. T., Perrett, D. I., Rowland, D., Williams, S. C. R., & David, A. S. (1997). A specific neural substrate for facial expressions of disgust. Nature, 389, 495–498. Robins, L. (1990). Deviant children grown up. Baltimore, MD: Williams & Williams. Singleton, N., Meltzer, M., & Gatward, R. (1998). Psychiatric Morbidity among Prisoners in England and Wales. Government Statistical Service. The Stationary Office: London. Yuille, J. C., & Cutshall, J. L. (1986). A case study of eyewitness memory. Journal of Applied Psychology, 71, 291–301.