Recent research has provided insight into how the brain's structural networks influence the personality trait of narcissism. Using advanced machine learning techniques, researchers have identified specific patterns of grey matter that predict narcissistic tendencies. The results of this study were recently published in European Journal of Neuroscience.
Individuals with high levels of narcissism often exhibit a grandiose sense of self-importance, a need for excessive praise, and a lack of empathy for others. These traits can have significant effects on mental health and interpersonal relationships, leading to a range of social and personal difficulties. Narcissistic personality disorder, a more extreme manifestation, affects approximately 1% to 15% of the U.S. population in clinical settings and presents unique challenges in mental health care due to its complex symptomatology and overlap with other personality disorders.
Despite the prevalence of narcissistic traits and NPD, research on their neurological underpinnings remains limited. Existing studies primarily focus on either gray or white matter individually and often use univariate methods that do not consider the complex interactions within the brain. This study aimed to fill this gap by using a multivariate machine learning approach to examine the joint contributions of both gray and white matter to narcissistic traits.
“In the Clinical Affective Neuroscience Laboratory, we are working on creating neural predictive models of personality that in future can guide clinical classification and the prevention of disease progression. Personality is who we are. However, our understanding of its neural basis and how to predict personality from neural traits is still insufficient. In our lab, we are paving the way to achieve this goal,” said Alessandro Grecucci, professor of Affective Neuroscience and Neuroengineering at the University of Trento and study author.
To investigate the neural underpinnings of narcissistic personality traits, the researchers utilized data from the MPI-Leipzig Mind Brain Body dataset, which contains MRI and behavioral data from 318 participants. The analysis focused on 135 healthy individuals, consisting of 64 women and 71 men, with a mean age of 31.94 years. These participants were selected on the basis that they were in good health, did not use drugs, and had no history of substance abuse or neurological disorders.
Personality traits were assessed using the Personality Style and Disorders Inventory (PSDI), a validated self-report inventory that can measure a range of personality traits and indicate potential personality disorders. The researchers focused on the narcissistic, histrionic, anxious/avoidant, and paranoid subscales to differentiate neural networks associated with these traits.
To analyze the data, the research team used parallel independent component analysis (p-ICA), a machine learning technique that identifies independent components in multimodal data. This method allowed them to examine the covariation of gray and white matter. They then applied stepwise regression and random forest regression to predict narcissistic traits based on these brain networks.
The researchers identified eight separate networks in both grey and white matter. Grey matter is primarily made up of neuronal cell bodies, dendrites and synapses and is involved in processing and interpreting information in the brain. White matter, on the other hand, is made up of bundles of myelinated axons that connect different grey matter regions and facilitate communication between them.
Of these networks, one in particular stood out for its strong association with narcissistic traits: This network encompassed regions of the frontal, temporal, and parietal lobes, including the superior temporal gyrus, angular gyrus, and middle temporal gyrus, areas involved in social cognition and empathy, as well as white matter regions of the cerebellum and thalamus, which are important for cognitive and emotional processing.
Notably, the identified grey matter regions largely overlapped with the default mode network (DMN), which is known for its role in self-referential thinking, social cognition and processing emotional experiences. This overlap suggests that the DMN plays an important role in the expression of narcissistic traits and reinforces the idea that these traits are deeply rooted in brain structures and functions related to self-awareness and social interactions.
The study's predictive model, developed using random forest regression, confirmed the robustness of these findings: the model demonstrated that the identified network could reliably predict narcissism traits in new individuals, highlighting the potential for using brain imaging data to assess personality traits.
“We are trying to develop a neural model of narcissistic personality disorder,” Grecucci told PsyPost, “a psychiatric illness characterized by pervasive delusions of grandiosity, a need for constant admiration, and a lack of empathy for others. The disorder can lead to significant impairments in personal and professional relationships and overall functioning. In our previous work, we focused on gray matter aspects. In this new paper, we extend our findings to include white matter contributions.”
“Thanks to innovative data fusion machine learning methods, we discovered a covariant grey matter circuit that encodes enough information to predict narcissistic personality. We then extracted a predictive model from this circuit that can predict levels of narcissism.”
“This circuitry overlaps with the default mode network, a network of interacting brain regions that is highly active when people are at rest and is associated with mindfulness, introspection and thought,” explains Grecci. “We argue that this may be one of the hubs that encodes our personality, and we have demonstrated that it is also involved in other personality traits such as borderline, antisocial and, more recently, obsessive-compulsive disorder.”
This study sheds light on the neural underpinnings of narcissistic personality traits, but it also has limitations that should be considered: it relies solely on structural data and does not take into account functional brain data that could provide a more comprehensive understanding of neural mechanisms.
Although the sample size is larger than many previous studies, it can still be increased for a more robust whole-brain association analysis. Another limitation is the assessment of narcissistic traits using the PSDI. The PSDI does not distinguish between vulnerable and grandiose subtypes of narcissism. Future studies could investigate these subtypes separately to see if different brain networks are involved.
The study, “Reflecting Narcissus: Gray and white matter features jointly contribute to the default mode network predicting narcissistic personality traits,” was authored by Kanithin Jornkokgood, Teresa Baggio, Richard Bakiaj, Peela Wongupparaj, Remo Jobe, and Alessandro Grecucci.
