Application Of Machine Learning Techniques In Investigating The Relationship Between Neuroimaging Dataset Measured By Functional Near Infra-Red Spectroscopy And Behavioral Dataset In A Moral Judgment Task

Coupling behavioral information with functional neuroimaging data sets promises to provide comprehensive insight into many medical data analyses. Analyzing the relationship of data sets of such diverse natures across multiple subjects requires special considerations. This enables a much more robust characterization of different data sets. Here, we investigate the relation between psychopathic traits quantified by the Psychopathic Personality Inventory-Revised [PPI-R]; (behavioral data set) and brain functional activities captured by functional near infra-red spectroscopy (fNIRS; neuroimaging data set). Particularly, we wanted to determine the psychopathic core traits most correlated with brain functional activation in personal (emotionally salient) and impersonal (more logical than emotional) moral judgment (MJ) decision-making. Our aim was to fill the gap in neuroimaging research between psychopathic traits and neuroimaging data during moral decision making using fNIRS. Applying Canonical Correlation Analysis (CCA) on brain functional activity recording from 30 healthy subjects and their psychopathic traits revealed coldheartedness and carefree non-planfulness to be highly correlated with prefrontal activation during personal (emotionally salient) MJ, while Machiavellian egocentricity, rebellious nonconformity, coldheartedness, and carefree non-planfulness were the core traits that exhibited the same dynamics as prefrontal activity during impersonal (more logical) MJ. Furthermore, ventromedial prefrontal cortex (vmPFC) and left lateral prefrontal cortex (PFC) were the prefrontal regions most positively correlated with psychopathic traits during personal MJ, and the right vmPFC and right lateral PFC were most correlated with impersonal MJ decision-making.