Dr. William Hockeimer received his PhD from Johns Hopkins University, working in the lab of Dr. James Knierim where he was an NSF GRFP Fellow. There, he performed simultaneous recordings of multiple hippocampal CA1 neurons in rodents using tetrode electrophysiology. He studied a phenomenon known as 'place field repetition' wherein place cells, which typically have a single spatial firing field, fire at multiple similar locations of an environment in a regular pattern. He analyzed the spatial and non-spatial firing properties of repeating neurons to better understand how CA1 instantiates a cognitive map. Before this, he worked at the Lieber Institute for Brain Development in the lab of Dr. Hao Yang Tan studying executive control in schizophrenia using fMRI. He received his undergraduate degree from the University of Michigan in 2013 with High Honors in Neuroscience and a minor in Philosophy and completed his honors thesis project in the lab of Dr. Michael Sutton studying LTP in dissociated hippocampal neurons.
Research Interest Summary
Brain computer interfaces; Neuroprosthetics; Motor Control; High-dimensional state spaces; Population doctrine approach to neuroscience
Research Interests
Dr. Hockeimer wants to understand cognition through the lens of dynamical systems in high-dimensional state spaces (Saxena and Cunningham, 2019; Vyas et al., 2020; Ebitz and Hayden, 2021). From this perspective, the activity of a population of neurons defines a point in high-dimensional space which updates according to dynamical rules defined by the network architecture and the inputs to the system. These constraints define a neural manifold along which the activity moves over time. Brain-computer interfaces (BCIs) provide a unique and powerful way to explore this idea by mapping neural activity, here intracortical motor cortex spiking, to control signals driving an effector, here a computer cursor. His postdoc at RNEL will explore how the neural manifold that drives cursor activity changes over time and learning. He and his colleagues hypothesize that neural activity will change in specific ways as participants learn how to use the BCI, and that these changes will lead to more expert use of the cursor. Further, perturbing the relationship between neural activity and cursor movement could help train participants to use the BCI more effectively. In a related effort, he will study how to overcome temporal nonstationarities in the neural signal that currently require the BCI to be frequently recalibrated – an issue that hinders in-home use. Animal research suggests that, despite changes in the population over time, a stable neural manifold exists that can be leveraged in a way that does not require as much time-consuming recalibration (Degenhart et al., 2020; Gallego et al., 2020). Dr. Hockeimer will pursue this idea in human participants. Together, these projects hope to create more robust BCIs, bringing this transformative technology closer to more widespread adoption.