White matter microstructure predicts effort and reward sensitivity

Neuroimage. 2026 Jan 16:121732. doi: 10.1016/j.neuroimage.2026.121732. Online ahead of print.

ABSTRACT

From rodents to humans, animals constantly face a central question: is the reward worth the effort? Effort and reward sensitivity in such situations vary substantially across individuals and ultimately shape goal-directed behavior. Yet, the neuroanatomical basis underlying this variability across individuals remain unclear. Here, we combined computational modeling of effort and reward sensitivity during decision-making with whole-brain diffusion MRI in 45 healthy participants to identify white matter substrates of individual effort and reward sensitivity. A data-driven, cluster-based analysis of fractional anisotropy and mean diffusivity revealed 12 clusters: five linked to effort sensitivity, all within tracts connected to major frontal valuation nodes (e.g., supplementary motor area [SMA], dorsal anterior cingulate cortex [dACC], orbitofrontal cortex [OFC]), and seven linked to reward sensitivity, spanning frontal valuation, fronto-parietal, and sensorimotor networks. The strongest associations involved two SMA-connected clusters, one shared across effort and reward sensitivity and another consistent across both microstructural metrics. Critically, microstructural features from the five effort-related and seven reward-related clusters reliably predicted graded individual differences in effort and reward sensitivity in out-of-sample, multi-class machine learning analyses, respectively, whereas randomly sampled clusters did not. SMA-connected tracts were the dominant predictors in these decoding analyses, with additional contributions from fronto-parietal and sensorimotor pathways for reward sensitivity. These findings reveal a distributed microstructure correlates underlying inter-individual differences in effort and reward sensitivity, with SMA pathways emerging as central hubs. They demonstrate that localized white matter microstructure can robustly predict these individual differences, offering a framework to forecast the impact of lesions or interventions on goal-directed behavior, including apathy and impulsivity.

PMID:41548823 | DOI:10.1016/j.neuroimage.2026.121732