Brain structure correlates of social information use: an exploratory machine learning approach

Area of Science:

• Neuroscience
• Cognitive Science
• Social Psychology

Background:

• Individual differences in social learning significantly influence societal behaviors, including voting and polarization.
• Previous research confirms stable individual differences in social information use, but underlying neural mechanisms remain unclear.

Purpose of the Study:

• To identify specific brain volumes associated with individual differences in social information use.
• To explore both linear and non-linear brain-behavior relationships using machine learning.

Main Methods:

• Employed two complementary machine learning techniques: lasso regression and random forest regression.
• Analyzed brain volumes in relation to individual variations in social information use.

Main Results:

• Confirmed a positive relationship between left pars triangularis volume and social information use.
• Identified known social brain network regions (mPFC, STS, TPJ, ACC) and novel regions (postcentral gyrus, caudal middle frontal gyrus, pallidum, entorhinal cortex).

Conclusions:

• Machine learning approaches successfully captured complex brain-behavior relationships in social learning.
• Findings offer novel insights into the neural basis of individual differences in social learning and suggest future research directions.