Psychopathy has been associated with increased putamen and striatum volumes. The nucleus accumbens – a key structure in reversal learning, less effective in psychopathy – has not yet received specific attention. Moreover, basal ganglia morphology has never been explored. We examined the morphology of the caudate, putamen and accumbens, manually segmented from magnetic resonance images of 26 offenders (age: 32.5 ± 8.4) with medium-high psychopathy (mean PCL-R = 30 ± 5) and 25 healthy controls (age: 34.6 ± 10.8). Local differences were statistically modeled using a surface-based radial distance mapping method (p < 0.05; multiple comparisons correction through permutation tests). In psychopathy, the caudate and putamen had normal global volume, but different morphology, significant after correction for multiple comparisons, for the right dorsal putamen (permutation test: p = 0.02). The volume of the nucleus accumbens was 13% smaller in psychopathy (p corrected for multiple comparisons < 0.006). The atypical morphology consisted of predominant anterior hypotrophy bilaterally (10–30%). Caudate and putamen local morphology displayed negative correlation with the lifestyle factor of the PCL-R (permutation test: p = 0.05 and 0.03). From these data, psychopathy appears to be associated with an atypical striatal morphology, with highly significant global and local differences of the accumbens. This is consistent with the clinical syndrome and with theories of limbic involvement.
Psychopathy is a personality disorder characterized by shallow affect, a callous lack of empathy or remorse, and an inclination to a parasitic and impulsive lifestyle, often leading to criminal behavior (Hare & Neumann, 2008).
Recent research on brain morphology has detected morphological differences in the brain structures of individuals with psychopathy, although differences in the methods used have led to inconsistent results (Koenigs, Baskin-Sommers, Zeier, & Newman, 2010). Differences have been reported in cortical and subcortical regions (Boccardi et al., 2011, de Oliveira-Souza et al., 2008, Muller et al., 2008, Tiihonen et al., 2008 and Yang, Raine, Colletti, Toga and Narr, 2009). Among these, the basal ganglia are crucial for emotional processing and behavioral planning, particularly via their connections with the amygdala and the frontal lobe, within the limbic circuit (Devinsky et al., 1995 and Vogt et al., 1992). Indeed, the hypothesis that psychopathic behavior may correlate with differences in the limbic structures and connections has been repeatedly corroborated (Craig et al., 2009, Kiehl, 2006 and Raine et al., 2010).
Some morphometric studies have found striatal volume alterations in conditions associated with violent behavior. One study found greater global volumes of the putamen in a group of subjects with antisocial personality disorder (ASPD) subjects (Barkataki, Kumari, Das, Taylor, & Sharma, 2006), which was interpreted as possibly related to impulsivity and poor behavioral control. A second study (Glenn, Raine, Yaralian, & Yang, 2010) examined the volume of the entire striatum, of the nucleus lenticularis (putamen and pallidus), and of the caudate head and body, in psychopathy. It found a 9.6% increase for the whole striatum, and similar enlargement of the nucleus lenticularis, bilaterally; findings were interpreted as connected to the increased stimulation seeking and impairments in reinforcement learning in psychopathy. A recent voxel-based morphometry (VBM) study detected greater local gray matter volume in the right caudate and in the left accumbens nuclei in offenders with sub-threshold psychopathy scores (Schiffer et al., 2011).
Nonetheless, only global volumes, or local gray matter density in voxel-based analyses, were examined, and information on the actual 3D shape of the basal ganglia is lacking to date. Moreover, the previous studies did not carry out specific region-of-interest analyses for the nucleus accumbens, the critical basal ganglia nucleus connecting to the amygdala and the orbitofrontal cortex within the anterior limbic circuit (Devinsky et al., 1995 and Vogt et al., 1992), which is known to be involved in psychopathy (Boccardi et al., 2011, de Oliveira-Souza et al., 2008, Kiehl, 2006, Raine et al., 2010 and Yang, Raine, Narr, Colletti and Toga, 2009). The accumbens is a key structure in reversal learning, i.e., the ability to overwrite on previously learned knowledge as context conditions change. Particularly, individuals need to take experience into account, and update the values that they attributed to stimuli or goals, as well as knowledge about the effects of their own behavior. A goal, or a behavior, that turned out to be good value in the past may not be worth or useful in a different time or context, so the individual must overwrite on prior knowledge in order to engage in more adaptive actions. Considering the limited ability of people with psychopathy to learn from experience, and that many are refractory to different kinds of treatment or correctional contexts, the accumbens is a particularly interesting target for neuroscientific investigation, to assess the neurobiological correlates of the cognitive and volitional capacities generating behavior in psychopathy.
Here we aimed to quantify the global volume of the caudate, putamen and accumbens nuclei, and to map local morphological differences using the radial distance mapping algorithm. We examined a sample of criminal offenders with psychopathy as defined by the Psychopathy Checklist Revised (PCL-R), lacking disorders in the schizophrenia spectrum, which are associated with independent brain alterations in individuals with violent behavior (Narayan et al., 2007).
This study detects a peculiar striatal morphology, characterized by severe hypotrophy of the nucleus accumbens, in offenders with psychopathy. Replication is required due to many limitations of the study. Anyway this finding is consistent with the clinical and neuropsychological features of psychopathy, as well as the known unresponsiveness to treatment or correction of these subjects.
Psychobiological correlates of psychopathy point to the involvement of the paralimbic network, a set of cerebral structures connecting emotional information to the structures processing cognitive information and controlling behaviour. Anyway, scientific investigations did not achieve definitive conclusions about this condition yet, and the transfer of this kind of data to everyday practice, especially at the juridical level, may not straightforward for a number of reasons, and should be considered as premature to date.
