Project 3

In Vivo Recordings during Delayed non-match to Place Tasks
John White

Project 3 involves recording from the hippocampus and entorhinal cortex of rats that are performing a delayed non-match to place (DNMP) task. The project has two major goals. First, we will test a number of specific predictions from the computational models of Project 4:

  1. CA3 pyramidal cells generate queued representations of temporal context
  2. Separate populations in EC code for temporal context and and sequences of paths from the current location
  3. The place fields of context-sensitive neurons move toward the goal as the animal runs the task repeatedly
  4. Each cycle of the theta rhytm can be divided into encodig and retrieval phases

Our second major goal is to strengthen the case that hippocampal activity is causally related to behavior in the DNMP task. We will address this issue by examining activity in trials as the animal is learning the task and by comparing activity associated with correct vs. incorrect choices by the animal. Finally, we will attempt to manipulate the activity of context-sensitive neurons and observe associated changes in behavior. The last set of experiments are technically challenging and rely on disputed accounts that neurons coding for context are spatially clustered along the septotemporal axis of hippocampus. Still, if successful, these experiments will provide direct evidence that manipulation of hippocampal acitivity leads to alterations in episodic-like memory.

Participating Institutions:
Boston University Center for Memory and Brain Department of Biomedical Engineering Department of Mathematics