An event-related potential investigation of response inhibition in schizophrenia and psychopathy

Abstract

Background: Schizophrenia and psychopathy are both characterized by impulsive, poorly planned behavior. This behavior may originate from a weak or poorly coordinated response inhibition system. We tested the hypothesis that schizophrenia and psychopathy are associated with abnormal neural processing during the suppression of inappropriate responses.

Methods: The participants were schizophrenic patients, nonpsychotic psychopaths, and nonpsychotic, nonpsychopathic control subjects (defined by the Hare Psychopathy Checklist—Revised), all incarcerated in a maximum security psychiatric facility. We recorded behavioral responses and event-related potentials (ERPs) during a Go/No Go task.

Results: Schizophrenic patients made more errors of commission than did the nonpsychopathic offenders. As expected, the nonpsychopathic nonpsychotic participants showed greater frontal ERP negativity (N275) to the No Go stimuli than to the Go stimuli. This effect was small in the schizophrenic patients and absent in the psychopaths. For the nonpsychopaths, the P375 ERP component was larger on Go than on No Go trials, a difference that was absent in schizophrenic patients and in the opposite direction in psychopaths.

Conclusions: These findings support the hypothesis that the neural processes involved in response inhibition are abnormal in both schizophrenia and psychopathy; however, the nature of these processes appears to be different in the two disorders.

Introduction

Schizophrenia and psychopathy are complex conditions with heterogeneous clinical presentation. Although these conditions differ in many respects, impulsive, poorly planned behavior is characteristic of both schizophrenia Bleuler 1950, Kraepelin 1919 and psychopathy Cleckley 1976, Hare 1993. In addition, recent research suggests that psychopaths exhibit subtle disorders resembling the disorganization of thought and behavior that occur in schizophrenia Gillstrom 1995, Hare 1993, Williamson 1991.

The purpose of the present study was to examine the neural correlates of behavioral inhibition in psychopaths and nonpsychopaths, and chronic, medicated schizophrenic patients, all incarcerated in a maximum security forensic facility. We recorded behavioral responses and event-related potentials (ERPs) while the participants performed a Go/No Go task. The ERP Go/No Go paradigm allowed us to examine the neural processes involved with making, and inhibiting, behavioral responses. We hypothesized that the schizophrenic patients and psychopaths would show anomalies in the neural processes involved in behavioral inhibition. Comparison of the spatial and temporal characteristics of the ERPs generated during the Go/No Go task offers the possibility of establishing similarities and differences in the neural processes associated with response generation and inhibition in the two disorders. In general, such a strategy can help to distinguish abnormalities that are specific to a disorder from abnormalities that are a nonspecific correlate of the functional impairment that occurs in the disorder.

Many studies indicate that the characteristic symptoms of schizophrenia segregate into three groups: psychomotor poverty, disorganization, and reality distortion Andreasen et al 1995, Bilder et al 1985, Liddle 1987b, Malla et al 1993. Psychomotor poverty includes the core negative symptoms of poverty of speech, blunted affect, and reduced spontaneous movement Liddle 1987b, Malla et al 1993. Disorganization includes formal thought disorder, bizarre behavior, and inappropriate affect. Reality distortion includes delusions and hallucinations.

On the basis of the similarities between the clinical features of these three syndromes and the clinical features associated with focal brain lesions, Liddle (1987a) proposed that each of the three syndromes was associated with a specific pattern of disordered regional brain function. In particular, the disorganization syndrome resembles the pseudopsychopathic syndrome that arises from damage to the ventral prefrontal cortex Blumer and Benson 1975, Damasio et al 1990. The pseudopsychopathic syndrome is characterized by impulsive behavior, rambling speech, and fatuous affect. In view of the fact that both the disorganization syndrome in schizophrenia and the pseudopsychopathic syndrome entail dysregulation of behavior, thought, or speech and affect, Liddle (1987a) proposed that both might arise from disordered function of right ventral frontal and medial prefrontal regions involved in the inhibition of inappropriate mental activity.

In support of this hypothesis, Liddle and Morris (1991) demonstrated that severity of disorganization, but not reality distortion or psychomotor poverty, is correlated with impaired performance in tasks that involve the selection of an appropriate response, such as the Stroop Color-Word Test and the Reitan Trails B Test. Other investigators have confirmed the association between formal thought disorder and impaired Stroop performance (Baxter and Liddle 1998). Frith et al (1991) demonstrated that formal thought disorder and incongruity of affect, the cardinal features of the disorganization syndrome, were associated with errors of commission during a continuous performance test.

To test the hypothesis that the disorganization syndrome is associated with aberrant activity in ventral and medial prefrontal cortex, Liddle et al (1992) used positron emission tomography (PET) to determine the pattern of correlation between regional cerebral blood flow (rCBF) and severity of the disorganization syndrome. In accordance with prediction, they found that the disorganization syndrome was correlated with underactivity in right ventral prefrontal cortex and contiguous insula, and with overactivity in right anterior cingulate. The site of overactivity in anterior cingulate cortex coincided with the site demonstrated to be maximally activated in healthy participants during the performance of the Stroop Test (Pardo et al 1990). In the Stroop Test, the participant is required to inhibit a tendency to respond to irrelevant aspects of a stimulus. The observation that disorganization is associated with overactivity in anterior cingulate is consistent with the proposal that, in patients with disorganization syndrome, inadequate activation of ventral prefrontal function results in the intrusion of inappropriate mental activity into current mental processing, leading to abnormal overactivity in the anterior cingulate region involved in suppressing inappropriate responses.

Subsequent studies of rCBF have provided partial confirmation of these findings, though there have been some discrepancies. For example, Ebmeier et al (1993), and Yuasa et al (1995) both confirmed that disorganization is associated with overactivity in anterior cingulate, though neither of these studies found an association with underactivity in the ventral frontal cortex.

Overall, the observed similarity between the pseudopsychopathic syndrome and the disorganization syndrome has proved to be a source of fruitful hypotheses regarding the nature of the cerebral abnormalities associated with the disorganization syndrome. The utility of the similarities between the two syndromes for generating further hypotheses about the underlying neural mechanisms is limited, however, by the fact that pseudopsychopathy arises from gross lesions to the brain Blumer and Benson 1975, Damasio et al 1990, whereas schizophrenia is not usually associated with gross structural lesions (Raz and Raz 1990). In contrast, psychopathy, which involves abnormalities of response inhibition similar in some respects to those occurring in pseudopsychopathy Cleckley 1976, Hare 1993, does not arise from a gross structural lesion of the brain, and hence is potentially a more appropriate condition to compare with schizophrenia.

Psychopathy is a personality disorder defined by a constellation of affective, interpersonal, and behavioral characteristics, including egocentricity, lack of empathy or guilt, shallow affect, impulsivity, sensation-seeking, poor behavioral controls, and a persistent violation of social norms and expectations. Psychopaths are more likely than normal individuals to act on the spur of the moment, without considering the possible consequences of their actions Cleckley 1976, Hare 1993.

Newman (1998) has argued that much of the impulsive, disinhibitory behavior of psychopaths, including difficulty in avoiding punishment and inhibiting dominant response sets, is the result of poor response modulation, defined as a cognitive processing deficit that hampers the ability to accommodate the meaning of contextual cues while engaged in the active organization and implementation of goal-directed behavior. Newman (1998) proposes that the poor response modulation of psychopaths represents a situation-specific failure to suspend ongoing behavior and to reallocate attentional resources.

Laboratory studies of response inhibition in psychopaths have produced mixed results. For example, Newman and his colleagues Newman and Kosson 1986, Newman et al 1985, Newman et al 1990, Patterson and Newman 1993 consistently have found that psychopaths are relatively poor at learning to inhibit reward-seeking behavior that results in monetary punishment. Lapierre et al (1995) reported that psychopaths made significantly more errors of commission when performing a Go/No Go task than did nonpsychopaths; however, this latter task also placed substantial demands on visuo-spatial processing, making interpretation of this finding difficult. Raine (1985) found that psychopaths did not make more errors of commission than did nonpsychopaths during performance of a continuous performance task.

Inconsistent results also have been obtained from studies of the performance of psychopaths on other tasks that require inhibition of dominant response sets and shifts in the allocation of attentional resources, including the Wisconsin Card Sorting Test (WCST), the Stroop Test, and the Trails B Test. Newman and Wallace (1993) reported that psychopaths performed more poorly than did nonpsychopaths on the Trails B Test. Similar results were obtained by Hart et al (1990) in one sample of offenders, but the results were not replicated in a larger sample of offenders. In addition, there is little evidence that psychopaths consistently perform poorly on the WCST or the Stroop Test Gorenstein 1982, Hare 1984.

In general, it appears that impaired response inhibition in laboratory tasks is not nearly as pronounced in psychopaths as it is in schizophrenic patients. It is worth noting that in schizophrenic patients impairment on response inhibition tasks is associated with formal thought disorder and incongruity of affect Frith et al 1991, Liddle and Morris 1991, both key elements of the disorganization syndrome (Liddle 1987a). Interestingly, there is some evidence that psychopathy also may be associated with thought disorder Cleckley 1976, Gillstrom 1995, Hare 1993, Williamson 1991. Although thought disorder in psychopathy is more subtle than that observed in schizophrenia, thorough investigation of psychopaths’ speech and language reveals substantial abnormalities, including derailment, tangentiality and a lack of coherence (Williamson 1991). Williamson (1991) also found that 20 of 21 psychopaths, assessed with the Psychopathy Checklist—Revised (PCL-R; Hare 1991), met the Thought, Language and Communication (TLC; Andreasen 1979) criteria for thought disorder. B.J. Gillstrom (unpublished data, 1994) found that psychopaths performed poorly the Proverbs test, a test that reflects a breakdown in thought processes and is sensitive to the thought disorder associated with schizophrenia (Gorham 1956). Recently, we have shown that the visual P3 elicited by target stimuli is reduced in psychopaths compared to nonpsychopaths (Kiehl et al 1999b). One interpretation of these P3 effects in psychopaths is that it reflects subtle thought disorder (Ward et al 1992). Thus, there is accumulating evidence that psychopathy is associated with subtle thought disorder.

Of course, psychopathy and schizophrenia also differ in many respects. For example, lack of conscience and an ability to manipulate and control others are features of psychopathy but not of schizophrenia, while many features of schizophrenia, including delusions, hallucinations, and the core negative symptoms (poverty of speech and decreased voluntary activity) are not characteristic of psychopathy. Nonetheless, the possibility that the disorganization syndrome found in schizophrenics may also occur in psychopaths, though to a less pronounced and dramatic degree, implies that similar neurologic mechanisms may underlie some features of the two disorders. In the present study, we investigated neural mechanisms associated with response selection in a Go/No Go paradigm. Our goal was to distinguish those neural abnormalities that are specific to each of the two disorders from those that reflect a nonspecific neural disorder of response selection.

Go/No Go paradigms have been used in both animal and human experimentation to assess impulsivity, learning, and response to reward and punishment Brown et al 1989, Schroger 1993, Watanabe 1986. The neural correlates of several different Go/No Go tasks have been studied with ERPs. The most common ERP components to show modulation during a visually presented Go/No Go task include a negative potential with a fronto-central distribution at approximately 275 msec poststimulus (N2) and a centro-parietal positive potential at 300 msec (P3). The anterior N2 is larger for No Go trials than for Go trials, and has been interpreted as a reflection of early response inhibition processes Eimer 1993, Gemba and Sasaki 1989, Jodo and Kayam 1992, Kok 1986, Naito and Matsumura 1994, Pfefferbaum et al 1985, Simson et al 1977. The posterior P3 is larger for Go trials than for No Go trials and the scalp topography of the P3 has parietal maximum for Go trials and a centro-parietal distribution for the No Go trials. Hillyard et al 1976, Jodo and Inoue 1990, Karlin et al 1969, Kok 1986, Pfefferbaum and Ford 1988, Pfefferbaum et al 1985, Simson et al 1977. These morphological and topographical P3 differences between Go and No Go trials are thought to reflect late (e.g., 300 msec or later) decision-making processes. We expected that schizophrenic patients and psychopaths would show reduced amplitude of the early No Go N2 “response inhibition” potential. Furthemore, in light of the evidence that the P3 component is abnormal during the classical oddball task in both schizophrenic patients (McCarley et al 1993) and in psychopaths (Kiehl et al 1999a), we further expected that the morphology and topography of the P3 “decision-making” potential would be abnormal in both schizophrenic patients and psychopaths. Specifically, we predicted that the posterior P3 to target stimuli would be reduced in psychopaths and in schizophrenic patients relative to the nonpsychopathic control subjects.