Entorhinal Cortex Signals Dimensions of Past Experience That Can Be Generalized in a Novel Environment

Area of Science:

• Neuroscience
• Cognitive Science
• Decision Making

Background:

• Generalization is crucial for adapting to new challenges by leveraging past experiences.
• Distinguishing transferable aspects from novel ones is a key challenge in generalization.
• The entorhinal cortex (EC) is hypothesized to facilitate generalization through factorized neural representations.

Purpose of the Study:

• To test the hypothesis that the entorhinal cortex (EC) forms factorized representations to support generalization.
• To investigate how the brain distinguishes generalizable features from novel ones during early exposure to new environments.
• To examine the neural mechanisms underlying selective knowledge transfer.

Main Methods:

• Functional magnetic resonance imaging (fMRI) was used to record brain activity in 40 participants.
• Participants trained on behavioral sequences in one environment and then navigated a new environment with partial overlap in transitions.
• Behavioral accuracy and neural pattern similarity in the EC were analyzed based on the generalizability of transitions.

Main Results:

• Behavioral accuracy was higher for generalizable sequence transitions compared to novel ones, especially during initial exposure.
• Neural patterns in the entorhinal cortex (EC) differentiated generalizable aspects early in novel environments.
• Increased neural similarity in the EC for familiar starting locations correlated with improved performance when prior knowledge was reusable.

Conclusions:

• The entorhinal cortex (EC) plays a role in distinguishing generalizable features during early exposure to novel environments.
• Factorized representations in the EC may support selective transfer of past experiences to new situations.
• These findings elucidate neural mechanisms supporting adaptive generalization behavior.