Electronic Thesis and Dissertation Repository

Thesis Format

Integrated Article

Degree

Master of Science

Program

Applied Mathematics

Supervisor

Muller, Lyle

2nd Supervisor

Mináč, Ján

Abstract

The possibility of temporal coding in neural data through patterns of precise spike times has long been of interest in neuroscience. Recent and rapid advancements in experimental neuroscience make it not only possible, but also routine, to record the spikes of hundreds to thousands of cells simultaneously. These increasingly common large-scale data sets provide new opportunities to discover temporally precise and behaviourally relevant patterns of spiking activity across large populations of cells. At the same time, the exponential growth in size and complexity of new data sets presents its own methodological challenges. Specifically, it remains unclear how best to (1) discover precise spike-time coordination in data sets that challenge existing analysis techniques, and (2) determine whether detected coordination is relevant to behaviour. Here, we introduce a new approach for analyzing the structure of spike-time coordination, in which patterns of spikes are represented as complex-valued vectors. This approach discovers clusters of similar spike patterns, makes effective links between spike timing and behaviour, and provides insight into the structure of putative spike-time codes.

Summary for Lay Audience

Neurons in the brain represent and transmit information by firing “spikes” at specific times and at measurable rates. While firing rates are known to play a role in behaviour and cognition, it is still debated whether the precise times of spikes are meaningful. Due to recent and rapid advancements in experimental neuroscience, it is now possible to record the spikes of hundreds to thousands of cells simultaneously. These increasingly common large-scale data sets provide new opportunities to discover patterns of spikes with specific relationships to behaviour. However, they also introduce methodological challenges related to analyzing such large amounts of data. First, it is unclear how best to discover precisely repeating patterns of spikes in data sets that challenge existing analysis techniques. Second, once such patterns are found, it is not clear how to determine whether they are relevant for behaviour. Here, we introduce a new approach for analyzing the structure of spike-time patterns. This approach discovers instances of similar spike-time patterns in neural data, provides insights into the structure of these patterns, and makes meaningful links between spike timing and behaviour.

In Chapter 1, this approach is applied to spiking data from macaque monkeys performing a sophisticated working memory task that takes place in a virtual environment. Neural activity was recorded from the lateral prefrontal cortex, a brain region known to be important for working memory, while the subjects performed a memory guided navigation task. This approach discovered sequences of spiking activity that represent task-relevant information held in working memory.

In Chapter 2, this approach is used to study the phenomenon of “phase precession” in the rodent hippocampus. In this phenomenon, specific sequences of spiking activity are thought to function like a “GPS” that represents the rodent’s location. Here, we develop a mathematical description of these sequences that allows us to study how they transform over time and across cells. The mathematical properties of these transformations show that the sequences represent more than just current position. Instead, they represent trajectories between past positions and possible future positions, linking the roles of phase precession in navigation and memory formation.

Download

Share

COinS