Prediction of life satisfaction from resting-state functional connectome

Abstract

Background: Better life satisfaction (LS) is associated with better psychological and psychiatric outcomes. To the best of our knowledge, no studies have examined prediction models for LS. Methods: Using resting-state functional magnetic resonance imaging (R-fMRI) data from the Human Connectome Project (HCP) Young Adult S1200 dataset, we examined whether LS is predictable from intrinsic functional connectivity (iFC). All the HCP data were subdivided into either discovery (n = 100) or validation (n = 766) datasets. Using R-fMRI data in the discovery dataset, we computed a matrix of iFCs between brain regions. Ridge regression, in combination with principal component analysis and 10-fold cross-validation, was used to predict LS. Prediction performance was evaluated by comparing actual and predicted LS scores. The generalizability of the prediction model obtained from the discovery dataset was evaluated by applying this model to the validation dataset. Results: The model was able to successfully predict LS in the discovery dataset (r = 0.381, p