290 Naturalistic Acute Pain States Decoded From Neural and Facial Dynamics

INTRODUCTION: Pain is a complex experience that remains largely unexplored in naturalistic contexts, hindering our understanding of its neurobehavioral representation in ecologically valid settings. METHODS: We enrolled 12 patients with bilateral invasive recording electrodes, implanted for clinical seizure monitoring, to study their pain experiences in an unconstrained fashion over multiple days. By combining intracranial electroencephalography (iEEG), pain self-reports, automated facial expression quantification, and machine learning techniques, we evaluated two forms of acute pain in this setting: 1) fluctuating acute pain measured by pain self-reports and 2) transient episodes of momentary pain identified by observational cues. Facial action units from video were continuously quantified using computer vision. Pain states were split at the median value to define binary pain states. iEEG and quantitative facial features were used for decoding analyses. RESULTS: High self-reported pain states were associated with elevated blood pressure, increased pain medication use, and distinct facial muscle activations. Using machine learning, we decoded individual participants' high versus low self-reported pain states from distributed neural activity patterns (mean AUC = 0.70), involving mesolimbic regions, striatum, and temporoparietal cortex. High self-reported pain states exhibited increased low-frequency activity in temporoparietal areas and decreased high-frequency activity in mesolimbic regions compared to low pain states. This neural pain representation remained stable for several hours and was modulated by pain onset or relief. Facial expression changes also classified self-reported pain states, with results concordant with electrophysiological predictions. Importantly, we identified periods of momentary pain as a distinct naturalistic acute pain measure, which could be reliably differentiated from affect-neutral periods using intracranial and facial features, albeit with neural and facial patterns distinct from self-reported pain. CONCLUSIONS: Here we find reliable neurobehavioral markers of naturalistic acute pain across contexts and timescales, underscoring the potential for developing personalized pain interventions in real-world settings.