We acquire knowledge by connecting events that are experienced at different times and places, forming cognitive maps that represent the commonalities among and differences between individual events and their elements. In this talk, I will highlight the role of hippocampus and prefrontal cortex in forming domain-general cognitive maps that exaggerate event commonalities and differences simultaneously, further showing how distorted maps bias decision making and inference. Using neurocomputational approaches, I will present additional evidence that hippocampal and prefrontal cognitive maps reflect abstract representational geometries that code the regular structure of the environment, promoting generalization through inference. I will end by showing that prefrontal cortex supports efficient formation of abstract knowledge through a process akin to goal-directed dimensionality reduction. Collectively, these data show how neural representations extend knowledge beyond direct experience to allow for adaptive decision making in new contexts.
If you would like to meet Alison on the day of the talk, please get in touch with Nima Khalighinejad or Nick Myers, Department of Experimental Psychology.