The Disconnection of Psychopaths

Functional connectivity between the right amygdala and anterior vmPFC is reduced in psychopaths. From Fig. 2 of Motzkin et al., (2011).


The last post discussed the case of a 14 yr old boy with congenital brain abnormalities and severe antisocial behavior said to be "consistent with" psychopathy. This label is quite stigmatizing and the diagnosis is a controversial one (Skeem et al., 2011),¹ particularly in children. What is psychopathy, exactly? According to Ermer and colleagues (2011),

Psychopathy is a serious personality disorder marked by affective and interpersonal deficiencies, as well as behavioral problems and antisocial tendencies (Cleckley, 1976). Affective and interpersonal traits (termed Factor 1) include callousness and a profound inability to experience remorse, guilt, and empathy; antisocial and behavioral problems (termed Factor 2) include impulsivity, stimulation seeking, and irresponsibility. These symptoms tend to manifest at an early age, continue throughout adulthood, and pervade numerous aspects of psychopaths’ daily functioning.

As for the brain regions implicated in psychopathy, dysfunction in the amygdala and ventromedial prefrontal cortex (vmPFC) have been suspected for quite some time (Abbott, 2001; Blair, 2007; Koenigs et al., 2011). From this perspective, a recent study on the structural and functional connectivity of these two regions (Motzkin et al., 2011) isn't entirely groundbreaking. However, the logistics of conducting those experiments were anything but simple: the participants were male inmates of a medium security prison in Wisconsin.

Kent Kiehl outside the mobile scanner he has used to look at the brains of inmates at a New Mexico prison. Credit: Nature News.

Dr. Kiehl's work with criminal psychopaths has been featured in the New Yorker. In the present study, diffusion tensor imaging (DTI) and resting state fMRI were used to examine the structural and functional connectivity of the vmPFC (Motzkin et al., 2011), which has been associated with decision making and the regulation of emotional behavior. The Psychopathy Checklist-Revised (PCL-R) (Hare, 2003)² was administered to the participants. Those with scores >30 were classified as psychopaths, while the non-psychopaths scored <20.

In the DTI study of structural connectivity, 13 non-psychopaths were compared to 14 psychopaths, 7 of whom were low-anxious or primary psychopaths and 7 were high-anxious/secondary psychopaths.³ The uncinate fasciculus (UF) is the main pathway connecting the vmPFC and the anterior temporal lobe (including the amygdala). Fractional anisotropy (FA), a measure of white matter structural integrity, was compared between the groups for regions of interest (after scaling for overall whole brain FA, which was reduced in the psychopaths). For comparison, FA in other frontal-temporal tracts was examined. Results indicated that FA was indeed reduced in the right UF, which replicates an earlier study (Craig et al., 2009).

Fig. 1 (modified from Motzkin et al., 2011). DTI results: reduced white matter integrity is specific to the right UF in psychopaths. b, The UF ROI (red) superimposed on an entire UF tract, as computed with tractography. f, Bar plots of mean scaled FA values in the UF. Psychopaths exhibited significantly lower scaled FA values only in right UF. Error bars indicate SEM. *p < 0.05.

Resting state fMRI, which measures spontaneous, low-frequency fluctuations in BOLD signal, was used to examine functional connectivity. In this experiment, 20 prisoners who were not psychopaths were compared to 20 prisoners who were considered psychopaths. Based on the DTI study, the right amygdala was used as a "seed region" to examine functional connectivity with vmPFC. In addition, the medial parietal cortex (precuneus and posterior cingulate) was used as a seed region, because this area is also reciprocally interconnected with vmPFC. If the amygdala-vmPFC and medial parietal-vmPFC circuits are not functioning properly, it was expected that correlated activity between regions would be lower. And indeed, that's what was observed in the psychopaths: resting vmPFC BOLD signal was less correlated with amygdala and medial parietal activity.

Fig. 3 (modified from Motzkin et al., 2011). Functional connectivity between medial parietal cortex and vmPFC is reduced in psychopaths. [...] The group difference map indicates two separate clusters within vmPFC where non-psychopaths have significantly greater connectivity with the precuneus/PCC seed than psychopaths (vmPFC and rACC). b, Group differences in connectivity were assessed in the vmPFC mask (red) for correlation coefficients computed using the mean time series extracted from the PCC seed (blue; x = −5, y = +49, z = +40). c, The bar plots depict group differences in connectivity estimates (Fisher z-transformed correlation coefficients) within each significant cluster.

Reduced connectivity between vmPFC and precuneus/posterior cingulate cortex (PCC)? What might this mean? Those two regions are major nodes in the default mode network (DMN), and activity in this circuit might be related to "self-reflective cognition" (Qin & Northoff, 2011), which might be defective in psychopaths. But the authors are cautious in further interpreting this result:

However, at this point it may be premature to speculate on the precise functional significance of reduced vmPFC–medial parietal connectivity in psychopathy.

But if one isn't so circumspect, it would be easy to postulate that the reduction in vmPFC–medial parietal connectivity is a reflection of a fragmented sense of self in individuals with psychopathy. Here we can veer way off course into object relations theory, which is exactly what Gullhaugen and Nøttestad (2011) did in their review paper on case studies of psychopaths.

Object relations theory is a collective term for theories focusing on how early interpersonal interaction, or object relations, construct psychological structure (an inner experience of a self and an other; St. Claire, 2000), which serves as a foundation for future relationships. ...

According to this theory, “narcissism and disorders of the self imply that the very central structures of the personality (inner experiences of self and others) are defective” (St. Claire, 2000, p. 139). In the PCL-R, narcissism is covered by Item 2: Grandiose sense of self-worth, and research has demonstrated that the PCL-R Factor 1 (the interpersonal and affective domain) is correlated with narcissistic personality disorder. Individuals with defective personality structure/pathologically developed object relations are distinguished by a tendency to define others as one-dimensional need-satisfying objects (part–objects), deviances in self- and affect regulation, primitive feelings such as anxiety and rage, and projection and splitting as primitive defense.

Now before you say, "Really!?! I did not know that The Neurocritic was a proponent of neuropsychoanalysis," let me state that I am not a proponent of neuropsychoanalysis,⁴ although I have written about it in the past. I just happened to be reading these two papers at the same time and wondered how they might be related (odd as that may seem). Clearly, Gullhaugen and Nøttestad (2011) themselves are not interested in the neural correlates of psychopathy, and Motzkin et al. (2011) aren't especially curious about the psychopath's early interpersonal interactions. The true model for this sort of mapping is Professor Karl Friston, one of the most prominent (and prolific) researchers in the field of neuroimaging. In collaboration with Robin Carhart-Harris, he developed a neurobiological account of Freudian ideas (Carhart-Harris & Friston, 2010).

...Freud argued that the ego modulates and suppresses both exogenous and endogenous signals. In neurobiological terms, exogenous signals correspond to interoceptive and exteroceptive signals from thalamic and unimodal sensory areas that convey sensory signals (prediction-errors) to polymodal and medial temporal lobe structures. Endogenous signals could be equated with subsequent bottom-up prediction errors (excitation) arising in limbic and paralimbic systems, which are passed up to high-level polymodal cortical areas that comprise the nodes of the default-mode.

The ego (normal adult human consciousness) was more or less equated with activity in the brain's "default mode" network (Raichle et al., 2001), measured as spontaneous fluctuations in the BOLD signal during unconstrained cognition. Activity in the DMN is anti-correlated with activity in "active task" networks, which are engaged by typical neuroimaging experiments. Although they didn't mention psychopathy or antisocial personality disorder specifically, reduced activity in the DMN was associated with primitive thought and emotion. See Friston is Freudian and A Bayesian Brain is a Freudian Brain for further details.

Returning to the Gullhaugen and Nøttestad review, Table 2 summarized the case histories of 11 severely psychopathic offenders, which included your classic serial killers with horrific childhoods [you can skip the Hannibal Lecter analysis]. One of the most salient things (in my view) was the extensive co-morbidity with other disorders, both Axis II (borderline, narcissistic, histrionic, and schizoid personality disorders) and Axis I (major depression, anxiety, substance abuse, schizophreniform disorder). Although these individuals were completely devoid of remorse and empathy for other people, evidence that they experienced psychological pain included self-injury, suicidal thoughts and plans, very low self-esteem, hypervigilance, and "feelings of emptiness and longings for love."

This brings up the possible distinction between primary (low-anxious) and secondary (high-anxious) psychopathy. In the neuroimaging study, the primary psychopaths showed significantly greater functional connectivity than secondary psychopaths for the vmPFC–amygdala correlation map but not for the vmPFC–medial parietal one (Motzkin et al., 2011). The authors did not interpret this finding, but perhaps the disinhibited and antisocial behavior associated with PCL-R Factor 2 goes more awry in secondary psychopaths, whereas the "fragmented sense of self" is more similar in the two subtypes. The observation that led me to the uneasy DMN/object relations synthesis was the prevalence of identity disturbances in the psychoanalytic case material (Gullhaugen and Nøttestad, 2011), which included descriptors like fragmented personality, reality-testing deficit, derealization, and unstable self-image.

A huge caveat is in order here, as the the populations of Motzkin et al. (male prisoners in a medium-security prison) and Gullhaugen and Nøttestad (serial killers and other severe offenders, with women grossly overrepresented [3 of 11]) were entirely different. My limited knowledge led me to believe that the majority of the latter population were secondary psychopaths with multiple diagnoses. Although Motzkin et al. matched their psychopath subtypes for substance abuse history, the co-morbidity of other mental disorders is unknown [except for the exclusion of psychosis and bipolar disorder].

In the end, we might say that psychopathic individuals show both literal (white matter)⁵ and metaphorical (interpersonal relations) disconnections.


Footnotes

¹ To quote from the paper (Skeem et al., 2011):

Few psychological concepts evoke simultaneously as much fascination and misunderstanding as psychopathic personality, or psychopathy. Typically, individuals with psychopathy are misconceived as fundamentally different from the rest of humanity and as inalterably dangerous. Popular portrayals of “psychopaths” are diverse and conflicting, ranging from uncommonly impulsive and violent criminal offenders to corporate figures who callously and skillfully manuever their way to the highest rungs of the social ladder.

Thanks to @Keith_Laws for linking to this 68 page tome while I happened to be working on this post. No, I haven't read the whole thing.

² Forensic psychologist Dr. Karen Franklin has written about multiple controversies surrounding the PCL-R, including the failure of Factor 1 to predict violence and Dr. Hare's attempt to block publication of a critical article. Also see this NPR series on Weighing The Value Of A Test For Psychopaths.

³ For a thorough discussion of primary vs. secondary psychopathy and the dual process-model, see Skeem et al. (2011):

...the dual-process model is an etiologic theory that posits distinct mechanisms underlying the interpersonal-affective (i.e., meanness/boldness) and antisocial (i.e., disinhibition) features of psychopathy. From the dual-process perspective, boldness and, to a lesser extent, meanness reflect a weakness in emotional reactivity, particularly defensive (fear) reactivity. This temperamental deficit is presumed to go hand-in-hand with differences in the functioning of affective-motivational systems including the amygdala and affiliated brain structures. [. . .] ...a second temperamental process underpins the disinhibition component of psychopathy: externalizing-propensity, or the liability toward impulse-control problems of various types, including antisocial behavior and substance use. This propensity may be associated with dysfunction in fronto-cortical brain systems that help to regulate emotion and guide decision making and action.

⁴ OMG! Has the Frontiers franchise gone too far? Introducing Frontiers in Psychoanalysis and Neuropsychoanalysis. Really!?!

⁵ Whole brain FA was reduced in the psychopaths, but functional connectivity maps from resting state fMRI (i.e., correlated BOLD fluctuations) do not necessarily imply direct anatomical connections. "White matter" sounded better in that context, though. For more on white matter, see the Neurological Correlates blog.


References

Abbott A. (2001). Into the mind of a killer. Nature. 410:296-8.

Blair RJ. (2007). The amygdala and ventromedial prefrontal cortex in morality and psychopathy. Trends Cogn Sci. 11:387-92.

Carhart-Harris R, Friston,K. (2010). The default-mode, ego-functions and free-energy: a neurobiological account of Freudian ideas. Brain 33:1265-1283.

Ermer E, Cope LM, Nyalakanti PK, Calhoun VD, Kiehl KA. (2011). Aberrant paralimbic gray matter in criminal psychopathy. J Abnorm Psychol. Dec 12. [Epub ahead of print].

Sundt Gullhaugen A, & Aage Nøttestad J (2011). Looking for the hannibal behind the cannibal: current status of case research. International journal of offender therapy and comparative criminology, 55 (3), 350-69 PMID: 20413645

Koenigs M, Baskin-Sommers A, Zeier J, Newman JP. (2011). Investigating the neural correlates of psychopathy: a critical review. Mol Psychiatry 16(8):792-9.

Motzkin JC, Newman JP, Kiehl KA, & Koenigs M (2011). Reduced prefrontal connectivity in psychopathy. The Journal of neuroscience : the official journal of the Society for Neuroscience, 31 (48), 17348-57 PMID: 22131397

Qin P, Northoff G (2011) How is our self related to midline regions and the default-mode network? Neuroimage 57:1221–1233.

Raichle ME, MacLeod AM, Snyder AZ, Powers WJ, Gusnard DA, Shulman GL. (2001). A default mode of brain function. Proc. Natl. Acad. Sci. 98: 676–682.

Skeem JL, Polaschek DLL, Patrick CJ, Lilienfeld SO (2011). Psychopathic Personality: Bridging the Gap Between Scientific Evidence and Public Policy. Psychological Science in the Public Interest 12:95-162.