Sensory and pharmacologic perturbations to explore cognitive and movement correlates in motor cortical population activity
Abstract/Contents
- Abstract
- My dissertation research is centered on probing cognitive signals in motor cortex. It encompasses two primary projects studying the correlates of movement selection, planning, and execution in the activity of large populations of motor cortical neurons in awake, behaving monkeys. Each project follows the fundamental approach of applying non-specific perturbations with established effects on behaviors for which there exist known neural correlates in motor cortical population activity, enabling clear and testable predictions about their neural effects. In the first project (Chapter 2, in collaboration with Diogo Peixoto et al.), to probe the mechanisms of choice formation, I employed closed-loop, neurally triggered visual stimulus perturbations. Decision formation is a dynamic but inherently covert process. We sought to track and perturb individual decisions as they unfolded. We were able to decode a time-varying motor cortical neural signal, which we term the "Decision Variable (DV), " tracking an animals' evolving internal decision states as they performed a visuomotor motion discrimination task. Moment-to-moment fluctuations in our decoded DV were nearly instantaneously (statistically) predictive of monkeys' choices. I then tackled the question of whether the state of the DV influences the neural and behavioral response to additional sensory evidence, hypothesizing that a subject would be less sensitive to additional evidence if it was presented when they were more strongly committed to a choice. During individual choices, I triggered weak pulses of additional sensory information at distinct DV levels. As predicted, I found that neural and behavioral responses to these pulses were weaker when the pulses were delivered at strong DVs reflective of stronger commitment to a choice. The system was also more resistant to pulses presented longer after stimulus onset. Additionally, we found that DV variability decreased with prolonged stimulus presentation. These results together provide evidence of time-varying, absorbing decision boundaries in motor cortical population activity, allowing us to rule out particular mechanistic models of decision-making (e.g., those that assume static decision boundaries or lack boundaries altogether). Intriguingly, we also found that many of the moment-to-moment fluctuations in the decoded decision state were not tied to the randomly varying motion information in the visual stimulus, raising exciting new questions for future work to explore regarding other potential sources of this variability. Variants of our general approach may also ultimately prove useful for studying other covert cognitive or affective processes. This work has been published in a co-first-authored article in Nature. The second project, in collaboration with Saurabh Vyas, is an investigation of the effects of systemic methylphenidate (MPH), a commonly used stimulant, on cued reaching behavior (Chapter 3) and motor cortical neural population activity related to movement planning and execution (Chapter 4). The neural mechanisms underlying the behavioral effects of MPH are not well understood. Given its documented effects on movements across species, we sought to characterize the effects of MPH on the neural computations underlying reaching behavior. We embraced the "computation through dynamics" approach that has recently advanced motor neuroscience by understanding motor cortical activity as a dynamical system that generates movements, focusing on known population-level signals ("dynamical motifs") in hopes of furthering both our understanding of the drug's mechanism of action and the relationship between motor cortical activity and behavior. We expected that MPH would speed reaches, and therefore anticipated that the drug might somehow optimize these neural motifs (e.g., by making them faster or less noisy). We administered clinically relevant doses of MPH or placebo to monkeys performing a delayed reaching task while we recorded from motor cortex. We observed behavioral effects including decreased reaction times, increased peak reaching speeds, and decreased trial-by-trial reach variability. We also found effects of MPH on motor cortical dynamical motifs: at a low, clinically relevant dose, MPH reduced the variability of tuned preparatory activity, decreased the latency of the condition-invariant trigger signal, and increased the frequency and signal-to-noise ratio (SNR) of rotational dynamics associated with reach execution -- despite on average decreasing firing rates compared to placebo. The finding of increased rotational frequency contrasts with prior findings in instructed reach speed control and natural speed variability, suggesting that MPH-driven increases in speed may derive from a different mechanism. Overall, a low, clinically relevant dose of MPH appears to boost the gain and SNR of motor cortical population dynamics previously described during reaching behaviors. This project is an important step toward understanding the mechanism of action of a common psychotropic drug from a systems/computational neuroscience perspective, joining a recent body of related work spanning rodents, monkeys, and humans. It also raises intriguing new questions, including about how modulation of other brain regions by MPH may contribute to its effects on motor cortex. This work is currently in preparation as a co-first-authored manuscript; preliminary results have been presented at several conferences. In addition to using perturbations with established behavioral effects to learn something new about motor cortical dynamics, each of these studies also has the potential to inform the evolving field of human systems neuroscience as well as clinical and translational research relating to neuropsychiatric disorders. Both tracking covert mental states, as in the choice decoder project, and understanding the effects of psychoactive drugs on neural activity patterns may prove promising tools for identifying neural dynamics underlying particular mental and behavioral states; therefore, each of these projects could ultimately inform the development of more targeted treatments for motor and/or cognitive dysfunction, including work toward cognitive prosthetics.
Description
| Type of resource | text |
|---|---|
| Form | electronic resource; remote; computer; online resource |
| Extent | 1 online resource. |
| Place | California |
| Place | [Stanford, California] |
| Publisher | [Stanford University] |
| Copyright date | 2023; ©2023 |
| Publication date | 2023; 2023 |
| Issuance | monographic |
| Language | English |
Creators/Contributors
| Author | Verhein, Jessica Rose |
|---|---|
| Degree supervisor | Newsome, William T |
| Thesis advisor | Newsome, William T |
| Thesis advisor | Longo, Frank J, 1939- |
| Thesis advisor | Moore, Tirin, 1969- |
| Thesis advisor | Nuyujukian, Paul Herag |
| Degree committee member | Longo, Frank J, 1939- |
| Degree committee member | Moore, Tirin, 1969- |
| Degree committee member | Nuyujukian, Paul Herag |
| Associated with | Stanford University, School of Medicine |
| Associated with | Stanford University, Neurosciences Program |
Subjects
| Genre | Theses |
|---|---|
| Genre | Text |
Bibliographic information
| Statement of responsibility | Jessica Rose Verhein. |
|---|---|
| Note | Submitted to the Neurosciences Program. |
| Thesis | Thesis Ph.D. Stanford University 2023. |
| Location | https://purl.stanford.edu/vd672wh0314 |
Access conditions
- Copyright
- © 2023 by Jessica Rose Verhein
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Also listed in
Loading usage metrics...