Skip to main content

A machine learning approach to determine functional biomarkers associated with psychopathic personality traits in a moral judgment task

Thursday, September 14, 2017 — Poster Session III

12:00 p.m. – 1:30 p.m.
FAES Terrace


  • H Dashtestani
  • R Zaragoza
  • R Kermanian
  • F Chowdhry
  • AH Gandjbakhche


Antisocial personality disorder (ASPD) is characterized by a violation of the rights of others and lack of conformity to social norms. While ASPD is traditionally diagnosed through psychiatric evaluation in accordance with the symptoms outlined in DSMV, ASPD patients can be extremely manipulative, resulting in controversial diagnoses by the subjective measures. However, understanding the neural basis behind ASPD can greatly enhance traditional diagnostic methods. Scientists have also found correlations between psychopathic personality traits and responses to moral judgment (MJ) tasks. Our study is the first neuroimaging study that has implemented the MJ task with personality assessments of psychopathic traits in a cost-effective and patient-friendly environment. We utilize functional near infrared spectroscopy (fNIRS), which is portable and tolerable to patient movements. fNIRS measures brain activation by monitoring changes of oxygenated hemoglobin in the brain similar to fMRI-BOLD. The task used in our study of MJ is based on series of questions which examines personal versus impersonal dilemmas, defined as emotionally salient scenarios versus more distant ones. We hypothesized that the brain exhibits distinguishable hemodynamic patterns for each category. Using the hemodynamic responses of 20 typical subjects, we analyzed the fNIRS data using a non-linear classification method called cubic SVM. Our results show that we can differentiate between the personal and impersonal hemodynamic responses with mean accuracy of 83%. This confirms our hypothesis of distinguishable hemodynamic patterns by category and suggests that it is possible to classify degrees of psychopathy based on neural activity.

Category: Biomedical Engineering and Biophysics