Kiwa Tanaka¹, Chia-huei Tseng¹, Miao Cheng¹, Yuki Murai², Sai Sun¹; ¹Tohoku University, ²National Institute of Information and Communications Technology

Understanding how the human brain encodes body motion (BM) is essential for advancing theories in cognitive-social neuroscience. BM conveys rich, nonverbal information about inferred intentions (e.g., helping or harming), context-specific actions (e.g., greeting, expressing gratitude), or social emotions (e.g., anger directed toward others) even from a single individual’s movements that implicitly involve others. Using naturalistic video stimuli of professional performers with blurred faces, we focused on how emotion and sociality are conveyed purely through BM. In Experiment 1, we selected a set of 80 from 720 naturalistic videos in a performer-generated database (Tseng et al, 2025 VSS), based on 96 participants’ behavioral ratings along three dimensions: valence (positive vs. negative), arousal (high vs. low), and sociality (self-directed vs. other-directed). The selected stimuli were of equal arousal level with clear valence and sociality tendency. In Experiment 2, 28 participants viewed these videos during fMRI scanning, followed by a post-scan ratings task, allowing us to link brain responses to perceived emotional and social dimensions and validate the initial categorization. Neural results revealed that negatively valenced movements elicited stronger activation in the bilateral posterior insula and lingual gyrus, regions associated with interoception, empathy, and body-based semantic processing. In contrast, other-directed movements activated the bilateral superior temporal sulcus, a region involved in social perception and interpreting social interactions conveyed through BM. These findings provide novel neural evidence that the brain distinguishes between emotional, semantic, and social content conveyed through BM. By leveraging dynamic stimuli, our study advances understanding of how the brain encodes the multidimensional nature of body language. Future work will explore visual motion dynamics and individual differences in BM processing, with implications for early diagnosis and targeted interventions for individuals with social cognitive deficits.