| Summary: | <p>Memory is a cognitive function that is fundamental to our being. From specific facts that we learn about the world to the rich moments that we live through, our brains have a remarkable ability to store immense amounts of information and access it efficiently when needed. So, how do our brains achieve this? Psychology, animal electrophysiology, lesion, and non- invasive neuroimaging studies have provided robust evidence for underlying neural substrates of memory. However, to thoroughly understand the neural code for human memory, we must study the activity of individual neurons. Here, I leverage the unprecedented opportunity to record single neuron data from the anterior temporal lobe of patients with drug-resistant epilepsy to investigate how the brain encodes knowledge and experience.</p>
<p>Single neuron recordings provide high spatial and temporal resolution data, which can be difficult to decipher. Prior to conducting any experiments, I evaluated existing methods used to decode this data for neural insights and explored alternative approaches. This allowed me to establish a set of analyses procedures that I consistently used throughout this thesis.</p>
<p>I first endeavored to study the neural code for individual events in memory. Spiking sequences observed in the anterior temporal lobe has been shown to represent semantic information and also play a role in memory. Like any neural code, the reliability of their representations can fluctuate. I present results suggesting that these fluctuations depend on upstream stimulus factors and whether the subject actively engages in memory encoding. The reliability of these representations in turn determines whether the encoded stimuli are successfully recalled, reflected in the replay of spiking sequences during memory retrieval.</p>
<p>Unlike replay of single neuron activity in rodent spatial navigation which recapitulates entire trajectories, single neurons in the anterior temporal lobe replay to recapitulate individual stimuli. Given that most of our naturalistic experiences occur in sequence over time, I next investigated the neural code for sequential events in memory. Taking inspiration from the rodent replay literature and applying techniques used to detect replay of non-physical experiences from spontaneous neural activity in humans, I present preliminary evidence that cortical spiking sequences replay hierarchically to represent episodic memories.</p>
<p>Altogether, this thesis demonstrates the unique insight that can be gained from studying single neuron activity and makes suggestions for their role in semantic and episodic memory.</p>
|
|---|