Poster F19, Tuesday, March 28, 8:00 – 10:00 am, Pacific Concourse
Resting-state functional connectivity in large-scale brain networks predicts neuroticism and extraversion in novel individuals
Wei-Ting Hsu1, Monica D. Rosenberg1, Dustin Scheinost1, Emily S. Finn1, R. Todd Constable1, Marvin M. Chun1; 1Yale University
The personality dimensions of neuroticism and extraversion are strongly associated with emotional experience. Previous studies reported fMRI activity correlates of these traits, but no study has used resting-state brain connectivity to predict these traits. Here, using a fully cross-validated approach, we predict novel individuals’ neuroticism and extraversion from functional connectivity observed as they simply rested during fMRI scanning. To this end, we applied a novel technique, connectome-based predictive modeling (CPM; Finn et al. 2015; Rosenberg et al., 2016), to resting-state fMRI data and personality scores (self-reported NEO Five Factor Inventory) from 125 subjects of the Nathan Kline Institute Rockland sample. Using a predefined 268-node whole-brain functional atlas (Shen et al., 2012), we calculated for each individual a 268-by-268 “connectivity matrix,” where each cell consisted of the functional time-course correlation (“edge”) between one node and another. Within this full connectivity matrix, CPM identified functional connectivity networks whose strengths positively or negatively correlated with neuroticism and extraversion. The edges from these networks form a weight matrix that can be applied to a novel subject’s connectivity matrix, and combined with a general linear model to generate a predicted personality score. Using leave-one-out cross validation, we then correlated predicted scores with observed scores, and found that CPM predicted neuroticism (r=0.27, p<0.01) and extraversion scores (r=0.24, p<0.01). CPM previously predicted fluid intelligence (Finn et al., 2015) and sustained attention (Rosenberg et al., 2016); this study extends the method to predict personality traits from resting state functional connectivity data.
Topic Area: EMOTION & SOCIAL: Emotional responding