Changes in the Representations of Hippocampal Neurons During Learning (8/00-7/01)
Advisor: Howard Eichenbaum, Boston University Context: Research Assistantship
In Dr. Eichenbaum’s lab I headed an on-going project using in vivo electrophysiology to investigate how neuronal representations in the hippocampus change as an animal learns a new task. Since the 1970s research groups have recorded from "place-cells" in the hippocampus, which are pyramidal cells that fire selectively when the animal is in a particular location in a given environment. Previous work in Dr. Eichenbaum’s lab found that as an animal performs a spatial alternation task, a majority of the hippocampal pyramidal cells fire differently on right-turn and left-turn trials. Using a paradigm that controls for the animal’s overt behavior and sensory input, I began a project investigating how neuronal representations in the hippocampus changes from place-related firing to turn-selective firing as an animal learns a spatial alternation task. Specifically, I looked at whether the same cells that represented spatial locations in an initial task become turn-selective during learning of the spatial alternation task or whether previously silent cells in the hippocampus begin firing in a turn-selective manner. This work will be extended to determine the role of NMDA receptors in mediating these changes in neuronal representations by using local injections of an NMDA receptor antagonist. Since the NMDA receptor is thought to play a vital role in synaptic plasticity and in hippocampal function, I hypothesize that representations of neurons in the local area around the injection site won’t change, whereas neuronal representations in the contralateral hippocampus will change as normal. In the process of getting this project off the ground, I faced and overcame a number of technical difficulties. For example, I adapted the lab’s electrode driver so that it could drive six electrodes independently to maximize the number of simultaneously recordable units. This was important for my experiment because all the units from an animal must be recorded while the animal learns the task and the animal can only learn the task once. To improve the reliability of our data analysis, I also wrote a computer program to utilize an automatic spike sorting algorithm that has been shown to be more reliable than manual spike sorting.
Back