A Map in the Mind: How the Brain Transfers Experience express
When moving to a new apartment, a person doesn't relearn how to open doors or turn on lights. Experience from the old home transfers automatically: a kitchen is a place with a stove and a sink, a bedroom is a place with a bed. The brain doesn't store an exact map but a flexible schema that it overlays onto unfamiliar space.
Scientists have built a computer model of this transfer. It has two levels. The upper level is the abstract task: find food, remember a safe corner. The lower level is a specific room with walls and turns. Between the levels works a translator: it links the goal "find food" with a point on the map.
When the model enters a new room, it doesn't panic. It applies the old schema: if food was in the northwest corner before, it looks there. If the corner is empty, it quickly relearns. Similarly, a person in a new apartment first goes to the place where the light switch was in the old one.
The upper level of the model turned out to resemble the prefrontal cortex: there are cells that respond to progress toward the goal regardless of the specific place. The lower level resembles the hippocampus: there are place cells, like those in rats in a maze. The model combined the two levels, offering a new perspective on how the brain transfers experience.
Original abstract
as written by the authors arXiv:2601.18946
Schemas -- abstract relational structures that capture the commonalities across experiences -- are thought to underlie humans' and animals' ability to rapidly generalize knowledge, rebind new experiences to existing structures, and flexibly adapt behavior across contexts. Despite their central role in cognition, the computational principles and neural mechanisms supporting schema formation and use remain elusive. Here, we introduce schema-based hierarchical active inference (S-HAI), a novel computational framework that combines predictive processing and active inference with schema-based mechanisms. In S-HAI, a higher-level generative model encodes abstract task structure, while a lower-level model encodes spatial navigation, with the two levels linked by a grounding likelihood that maps abstract goals to physical locations. Through a series of simulations, we show that S-HAI reproduces key behavioral signatures of rapid schema-based generalization in spatial navigation tasks, including the ability to flexibly remap abstract schemas onto novel contexts, resolve goal ambiguity, and balance reuse versus accommodation of novel mappings. Crucially, S-HAI also reproduces prominent neural codes reported in rodent medial prefrontal cortex during a schema-dependent navigation and decision task, including task-invariant goal-progress cells, goal-and-spatially conjunctive cells, and place-like codes at the lower level. Taken together, these results provide a mechanistic account of schema-based learning and inference that bridges behavior, neural data, and theory. More broadly, our findings suggest that schema formation and generalization may arise from predictive processing principles implemented hierarchically across cortical and hippocampal circuits, enabling the generalization of experience.
Rats have cells that fire when the animal approaches a goal — even if the goal itself is at the other end of the maze. These "progress cells" were also found in the model.
In the movie Inception, architects build dreams using general schemas that the dreamer fills with details, just as the brain overlays a familiar schema onto a new place.