Keynotes | K&Ts | GACs | Talks | Posters | Search

Poster D in Poster Session D: Wednesday, August 5, 2:00 – 3:45 pm, Kimmel Center, Shorin & Rosenthal Rooms

Reconstructing subjective experiences across individuals from brain activity

Haibao Wang¹, Fan L. Cheng², Shuntaro C. Aoki^1,3, Misato Tanaka¹, Yoshihiro Nagano¹, Hideki Izumi¹, Yukiyasu Kamitani¹; ¹Kyoto University, ²Columbia University, ³ATR Computational Neuroscience Laboratories

Presenter: Haibao Wang

Subjective experiences are central to understanding how the brain represents mental contents and clinically relevant phenomena. However, decoding these mental contents remains challenging because it typically requires large amounts of subject-specific brain data. Functional alignment provides a potential solution by enabling decoding models to transfer across individuals through the alignment of brain activity patterns. Here, we investigated whether subjective contents can be decoded across individuals using functional alignment and image reconstruction. We trained a functional alignment model called the neural code converter on brain responses to natural images to establish inter-individual mappings, and then tested it on three subjective tasks: visual illusion, attention, and imagery. The model successfully generalized to these subjective tasks, enabling inter-individual image reconstructions that preserved subjective contents comparable to within-individual decoding. Our approach provides a framework for analyzing and externalizing subjective experiences across individuals.

Topic Area: Methods, Tools, Theory & Neural Coding