Theses Doctoral

In vivo electrophysiology in humans reveals neural codes for space and memory

Qasim, Salman Ehtesham

Memory serves an integral function in every aspect of human life. Losing that function can be adevastating consequence of disease, dementia, and trauma. In order to develop treatments or prophylactics for memory disorders we must identify the neural basis of memory. Animal research has made prominent strides studying the neural correlates of memory by examining the more easily observable and manipulable neural correlates of spatial context, since the brain regions necessary for declarative memory intersect profoundly with those needed for spatial navigation. My research has two main goals. My first two studies, in Chapters 2 and 3, translate animal research relating the neural correlates of space to memory processes, and go beyond animal work to explore how internal features of experience such as goal states influence these conjunctive representations of space and memory. In Chapter 4, I expand my scope to examine how another internal feature, emotional context, affects the same brain regions on a network level to influence memory representations in the human brain. To perform these studies I recorded directly from the human brain in epilepsy patients performing a variety of memory tasks.

First, I measured single-neuron activity as subjects navigated a virtual environment, encountering various objects at unique locations. As subjects moved through the environments, they were instructed to recall the locations of specific objects they encountered—I identified neurons in the human entorhinal cortex, called “memory-trace cells”, which selectively activated near the object-location that people were instructed to retrieve from memory. This is the first evidence that neurons in the brain can be tuned to the spatial context of an event for memory, and demonstrated a direct link between memory retrieval and the spatial tuning properties of neurons. For my second study, I examined whether spatially-tuned neurons in the MTL discharge at intervals organized by theta (2–10 Hz) oscillations (which represent network level brain-activity). I identified a particular pattern that is prominent in rodents, called “phase precession”, during which spatially-tuned neurons spike slightly faster than the network oscillation, and which is theorized to hold great value throughout the brain for learning and memory. In addition to discovering this pattern for spatial sequences, I discovered that phase precession was also present during more abstract features of experience, like the specific goal a person was seeking. These findings suggest that principles of network-level brain activity for organizing spatial navigation may extend to humans, and to broader forms of cognition and memory. Finally, I examined the role of the amygdala in memory encoding during a verbal episodic memory task, finding that the emotional context of a word influenced the probability of its subsequent recall. By measuring the prevalence and coordination of brain oscillations in the amygdala-hippocampal circuit, I found that gamma oscillations (30–120 Hz) increased in both regions as a function of word arousal and encoding success, and connectivity within the amygdala-hippocampal circuit also showed significant theta-gamma coupling as a function of memory and high arousal. Furthermore, direct 50 Hz stimulation impaired memory for high arousal words. These findings suggest a causal relationship between gamma oscillations in the amygdala-hippocampal circuit for memory as a function of emotional context during encoding.

My work generalizes important neuronal principles from animal studies to humans (such as spatially-tuned neurons and phase precession), but also extends those findings more deeply to memory, and to internal/subjective aspects of memory that are difficult to directly measure in animals. Overall this work represents an important step towards understanding how the human brain enables declarative memory.


  • thumnail for Qasim_columbia_0054D_16299.pdf Qasim_columbia_0054D_16299.pdf application/pdf 15.7 MB Download File

More About This Work

Academic Units
Biomedical Engineering
Thesis Advisors
Jacobs, Joshua
Ph.D., Columbia University
Published Here
December 14, 2020