Gray and White Matter Networks Predict Mindfulness and Mind Wandering Traits: A Data Fusion Machine Learning Approach

Abstract

ObjectivesThe main aim of our study was to find out which gray matter (GM) and white matter (WM) brain features are associated with mindfulness and mind wandering, and to investigate how mindfulness mediates deliberate and spontaneous mind wandering in terms of these associated brain components. MethodsStructural MRI scans of 76 individuals and self-reported questionnaires were included in this analysis. We applied unsupervised machine learning algorithms to the brain imaging data to decompose the fused GM and WM into naturally grouping covarying networks. We then conducted a mediation analysis to assess the influence of mindfulness on deliberate and spontaneous mind wandering based on this data fusion networks. Additionally, we investigated if certain mindfulness facets mediate the two forms of mind wandering traits. ResultsWe found GM and WM networks composed of structures that have been consistently linked to mindfulness (e.g., the cingulate, insula, basal ganglia) and fronto-parietal attentive regions exert direct effects on mindfulness, as well as on deliberate and spontaneous mind wandering. We also found an indirect mediating effect of the mindfulness facet acting with awareness on spontaneous mind wandering through one of the significant brain networks, we identified. ConclusionsThis study elicited the link between mind wandering and mindfulness, and expands our knowledge on the neural bases of these two psychological constructs.