Humans can organize their thoughts and actions to achieve specific goals by building “task sets” or “mental programs” describing which are the operations, facts, goals and relations relevant in a situation. However, how the human brain represents the different and multiple types of information involved in complex task sets is not fully understood.
In a series of fMRI experiments we explored the neural representation of different task features: simple stimulus-response rules, priority information, rule hierarchies, logical relations, and subtask assignment to different people. Furthermore, we explored the discovery and representation of features possibly relevant for new, yet to be implemented, rule sets.
Overall, we found that the neural representation of complex rules is “compositional”, i.e. built on the neural representation of their constituent rules or features. Complex task sets are “decomposed” in their elementary features. The constituent features are then represented in different brain structures, depending on the type of information to be stored. We argue that compositionality, “decomposition” and information-guided representation are general features of prefrontal cortex functional organization.
The presentation will focus specifically on three recent experiments, on the representation of the person to whom a subtask was assigned, on the representation of logical relations, and on the representation of yet-to-be used task features.