Goal-directed animals must optimize their behavior to address five fundamental questions: ‘What do I need?’, ‘Why do I need it?’, and ‘Where, when, and how do I get it?’. The brain solves this problem by creating internal representations of the animal’s world which are refined as the animal continues to explore its surroundings. These internal representations are thought to form the foundation of episodic memories, personal experiences occurring in a specific spaciotemporal context. The crucial role of the hippocampus in episodic memory formation and spatial navigation has been well documented, as well as the importance of hippocampal place cells in memory and navigation. However, it is still unknown exactly how this system works. In this thesis, we test the hypothesis that spatial decoding of an environment can be explained from cell-pair coactivity relationships and relies crucially on the most anti-coactive cells in the hippocampal CA1 region. In this experiment mice were virally infected so that CA1 principal cells expressed GCaMP6f, a fluorescent calcium sensor used to infer firing activity. A GRIN lens and UCLA miniscope were implanted to record fluorescent CA1 imaging data. Mice were placed in two distinct environments, a circle enclosure and a square enclosure, and were allowed to freely explore. Kendall tau correlations were calculated to rank cells by their participation in anti-cofiring cell pairs. IsoMap dimensionality reduction was computed for each enclosure and 2D manifolds were constructed. The proportion overlap between the manifolds of the circle and the square enclosure was used as a measure of the ability of the mouse to distinguish between each environment. Increasing percentages of anti-cofiring cells were then removed from the model to determine what would happen to the proportion overlap between the manifolds. Data was analyzed using the Kruskal-Wallis equality-of-populations rank test to reveal the following conclusions in support of the hypothesis: the removal of increasing percentages of anti-cofiring cells from the model was correlated with a significant increase in the proportion overlap between manifolds of the circle and square enclosures; anti-cofiring cells became more meaningful with time, indicating that these cells may be uniquely involved in spatial learning; removing the same number of cells at random was not correlated with a significant increase in the proportion overlap between manifolds.
| Identifer | oai:union.ndltd.org:bu.edu/oai:open.bu.edu:2144/45430 |
| Date | 10 January 2023 |
| Creators | Farb, Ariel |
| Contributors | Symes, Karen, Fenton, Andre |
| Source Sets | Boston University |
| Language | en_US |
| Detected Language | English |
| Type | Thesis/Dissertation |
Page generated in 0.0162 seconds