Linking Subjective Experience of Anxiety to Brain Function using Natural Language Processing

Background: Mechanistic research on anxiety often focuses on clinically relevant behaviors and physiological responses, particularly emphasizing recruitment of amygdala, insula, and cingulate cortex. Whether these same circuits instantiate subjective experience of anxiety remains unclear, a vital hurdle for clinical neuroscience. Methods: In this preregistered study, we used a semi-naturalistic, anxiogenic stimulus (animated movie) to evoke anxiety during fMRI in a pediatric sample with and without anxiety disorders (N = 84, before exclusion). After, participants provided verbal responses to a set of interview questions about the stimulus. We quantified semantic content and valence of responses via natural language processing algorithms. Results: Preregistered analyses found wide-spread brain activity during the movie—including in anterior insula cortex—related to participants’ descriptions of the movie’s narrative. Secondary, non-preregistered analyses also revealed interactions between anxiety, age, and verbal appraisals. Conclusions: Our primary analyses link movie-evoked brain activity and subsequent interview content in a transdiagnostic pediatric sample. The language participants used when recalling anxiogenic movie content robustly indexed brain activity. Secondary analyses suggested older but not younger patients with anxiety demonstrate more negative appraisals of the anxiogenic experience. Our findings underscore a promising avenue for clinical neuroscience research on subjective emotional experiences in anxiety.