Qualia Structurology — Bridging Subjective Conscious Experience and Objectivity

This project aims to bring subjective conscious experiences (qualia) — such as the redness of red or the painfulness of pain — onto the scientific stage as structures. The research area builds methodologies that measure, describe, and compare the “structure” of qualia at the intersection of psychophysics, neuroscience, information theory, and machine learning.

We participate in the planned research project “Correspondence between Qualia Structures and Information Structures” (PI: Masafumi Oizumi), and use deep generative models and representation learning to model the correspondence between the structures of human subjective experience and the information structures inherent in neural activity and behavioral data. In particular, we quantitatively verify the consistency between latent representations obtained via self-supervised learning and similarity structures derived from psychophysics, taking on the challenge of a constructive understanding of “the structure of inner experience” in robots and AI.