The information architecture of the naturalistic human brain
Year: 2026
Type: preprint
Abstract: How does the brain orchestrate its central currency, information, for complex adaptive cognition in the real world? Here we provide a comprehensive investigation into the spatiotemporal structure of qualitatively distinct modes of information dynamics in the human brain under ecological conditions. Applying a novel framework for information decomposition to fMRI data from individuals watching full-length feature films, we reveal a multidimensional large-scale information architecture, wherein different schemes for storing, transferring, and transforming representations occupy disjoint topographic and network profiles. Further, this architecture reveals a computational division of labour, with certain regions acting as generalists that support multiple information-processing schemes and others specialising in a few. We show how these interactions reconfigure over time, coalescing into a structured cycle synchronised by sensory input, whose elements are tuned to semantic kinematics. Finally, we replicate these findings in an independent, large-scale dataset. Together, we provide a formal and empirical foundation for understanding the principles that govern distributed computation in the human brain during everyday life.
Language: English
Subfield: Cognitive Neuroscience
Field: Neuroscience
Domain: Life Sciences
Open Access status: green
OpenAlex ID: W7236561992
