Skip to main content
eScholarship
Open Access Publications from the University of California

Algorithmic representations in the human brain that underlie schema generalisation

Creative Commons 'BY' version 4.0 license
Abstract

The human brain represents structure in the external world as "cognitive maps". However, it remains unknown how it represents structure in one's own behaviour. Recent findings show rodent medial-prefrontal cortex (mPFC) does this with a structure-sensitive representation, where all future actions are represented simultaneously. Here, using 7T fMRI, we test for this representation and its properties in humans. Using RSA and a computational model, we show this representation in mPFC and orbitofrontal cortex, while entorhinal and orbitofrontal cortices contain a pure abstraction of task structure. Preceding the future actions representation, action plans are ‘loaded' to mPFC once subjects were given all action-relevant information, suggestively through replay. iEEG data of patients solving the same task shows that sharp-wave ripple rate is increased during this planning time. Together, our findings suggest an algorithm of how human mPFC encodes future actions, and provide evidence for a replay-based mechanism of loading that representation.