Predicting apathy using connectome-based models derived from static and dynamic brain connectivity.

Abstract

Apathy is a prevalent neuropsychiatric symptom across various neurological and psychiatric disorders. Despite its significant impact on functional outcomes, quality of life, and caregiver burden, the neural mechanisms underlying apathy remain poorly understood. Static and dynamic functional connectivity serve as neural fingerprints for personalized predictions, capturing complementary aspects of brain function and engaging distinct networks. This study developed predictive models of apathy using both static and dynamic functional connectivity to elucidate network-level mechanisms in healthy university students. Static connectome-based predictive modeling (CPM) demonstrated that disrupting the default mode network significantly impaired prediction, which may be related to deficits in internal motivation associated with apathy. Dynamic CPM revealed that lesioning the medial frontal, fronto-parietal, and visual II networks diminished accuracy, suggesting that impairments in behavioral initiation and execution in apathy. By integrating static and dynamic connectivity in predictive models, this study uncovers complementary network dynamics underlying apathy and highlights the potential neural basis of apathy.