Hair cortisol-related resting-state effective connectivity between the amygdala, hippocampus, and striatum: A...

Hair cortisol is a promising biomarker of chronic stress. However, few studies have investigated the resting-state neural correlates of subjective stress, and fewer still this biomarker. Characterizing brain function associated with hair cortisol offers insights into chronic stress through a psychoneuroendocrinological lens. Sixty-six participants (74.2% female; mean age = 20.64 ± 2.76 years, age range: 17-30 years) were included in the final analysis, all of whom completed resting-state functional magnetic resonance imaging (rs-fMRI), hair collection, and questionnaires. Spectral dynamic causal modeling (spDCM) characterized hair cortisol-related effective connectivity (EC), with support vector regression (SVR) in machine learning (ML) predicting psychological distress (i.e., perceived stress, state anxiety, depression severity, and negative affect) from the identified neural markers. Results indicated that hair cortisol concentrations (HCC) predominantly correlate positively with excitatory signaling from the amygdala and inhibitory signaling from striatal subregions, but negatively with excitatory signaling from the hippocampus and inhibitory signaling from striatal subregions and the hippocampus. The excitatory EC from the left-to-right hippocampus consistently predicted all psychological distress measures, emerging as a robust predictor. Our work pioneers the study of hair cortisol-related EC and its predictive effects on psychological distress. These findings may inform future research exploring the mesoscopic and microscopic neuroendocrine mechanisms underlying chronic stress.