Saetbyeol LeeYouk, Sergio Mutis, Yuxiang Cheng, and Behnaz Farahi. 2026. Portable Somatic Wearable: AI-Assisted Reflection Through Gaze-Derived Somatic Markers. In Extended Abstracts of the 2026 CHI Conference on Human Factors in Computing Systems (CHI EA '26), April 13–17, 2026, Barcelona, Spain. ACM, New York, NY, USA. https://doi.org/10.1145/3772363.3799134

Abstract

AI-assisted reflection systems are increasingly common, yet most remain primarily linguistic, overlooking the embodied and pre-cognitive signals through which people experience and silently express affect. We present the Portable Somatic Wearable, an interactive demo that enables attendees to engage in a short AI-guided voice reflection while gaze-derived somatic markers shape the conversation in real time. The system captures nonverbal cues such as hesitation, fixation dynamics, and temporal pacing, feeding these markers back into the AI facilitator to adapt conversational pacing and supportive prompts. Addressing limitations observed in our prior Gaze to the Stars installation, the Portable Somatic Wearable enables rapid setup and repeatable deployment through a compact, accessible headphone-integrated sensing module. We contribute (1) a portable sensing platform for repeatable gaze-based somatic capture and (2) an integrated somatic-reactive pipeline that uses these markers to adapt AI interaction, offering a path toward more accessible, socially attuned and humane AI-mediated reflection systems.