A shared neural encoding model for the prediction of subject-specific fMRI response

TL;DR

This paper introduces a shared neural encoding model that leverages multi-subject fMRI data to improve prediction accuracy of individual responses to visual and auditory stimuli, capturing meaningful individual differences.

Contribution

The study presents a novel shared convolutional neural encoding approach that enhances subject-specific fMRI response prediction by utilizing inter-subject knowledge transfer.

Findings

Significant improvement over single-subject models in fMRI response prediction.

Effective capture of individual differences in response to facial and scene stimuli.

Demonstrated benefits of multi-subject data integration for neural encoding.

Abstract

The increasing popularity of naturalistic paradigms in fMRI (such as movie watching) demands novel strategies for multi-subject data analysis, such as use of neural encoding models. In the present study, we propose a shared convolutional neural encoding method that accounts for individual-level differences. Our method leverages multi-subject data to improve the prediction of subject-specific responses evoked by visual or auditory stimuli. We showcase our approach on high-resolution 7T fMRI data from the Human Connectome Project movie-watching protocol and demonstrate significant improvement over single-subject encoding models. We further demonstrate the ability of the shared encoding model to successfully capture meaningful individual differences in response to traditional task-based facial and scenes stimuli. Taken together, our findings suggest that inter-subject knowledge transfer can be beneficial to subject-specific predictive models.