Neural predictors of hidden, persistent psychological states at work

Abstract

Common workplace challenges such as feeling overwhelmed, burned out, or disengaged often remain hidden due to fear of judgment or social norms, contributing to rising mental health crises and organizational dysfunction. This study presents a brain-based framework for predicting these hidden and persistent psychological states through noninvasive neuroimaging. We used functional near-infrared spectroscopy to record neural activity from 67 executives in the field as they watched a video about workplace attitudes. We then applied a multitimepoint pattern analysis (MTPA) approach to reduce timeseries dimensionality and successfully classify whether individuals were feeling overwhelmed (72.8% accuracy) or in need of a new or different challenge (79.1% accuracy) in their careers using the temporal parietal junction (TPJ) and dorsal medial prefrontal cortex (dmPFC), respectively. The MTPA framework also allowed us to reverse-engineer specific thematic properties of the stimulus that evoked differential neural responses linked to these predicted outcomes. Emotional content in the video (e.g., reported distress) corresponded to the selected TPJ features that predicted whether someone was overwhelmed, while socially relevant content (e.g., missing social gatherings) aligned with the selected features from the dmPFC that predicted the need for a new or different challenge. These findings demonstrate the ability of neural measures to unobtrusively identify hidden and persistent psychological states in real-world settings, enabling targeted interventions that can improve well-being and engagement.