Leo Wood, QBioS Thesis Proposal

Leo Wood
Quantitative Biosciences Thesis Proposal
School of Physics
Advisor: Dr. Simon Sponberg (School of Physics)
Open to the Community

Integration of sensory streams into a coordinated, precise, and robust hawkmoth flight motor program
Friday, October 14, 2022, at 1:00 pm
In Person Location: Howey Physics, Room N201/202

Committee Members:
Dr. Hannah Choi (School of Mathematics)
Dr. Jeffrey Markowitz (School of Biomedical Engineering)
Dr. Young-Hui Chang (School of Applied Physiology)

Abstract:
Biological systems are unparalleled at controlling highly coupled, nonlinear systems in a robust and generalized manner, integrating often noisy and conflicting sensory information on-the-fly into a precise and coordinated motor program. Vision, olfaction, proprioception, and many other senses are somehow continuously transformed into a patterned set of action potentials to each muscle. While a great deal is understood about how steady-state behaviors are controlled and produced, it is less clear how behaviors such as visual target tracking or navigation, where centralized visual information must continuously alter the motor program, are generated and controlled. Even less is understood about how visual information modulates the motor program of species such as the hawkmoth Manduca sexta whose motor program is both millisecond-scale precise in spike timing and highly coordinated. How does descending visual information integrate with local mechanosensation and proprioception to generate a coordinated motor program?

I aim to answer this question by studying the flight motor program of the hawkmoth Manduca sexta with a range of techniques such as electrode array recordings, electromyography of a comprehensive spike-resolution motor program, and data-driven modeling. I hypothesize that wing reafference drives within-cycle spike timing precision of the motor program, but coordination patterns and relative timing of muscles are primarily modulated by descending visual information. I propose testing this hypothesis through 5 specific aims. 

First, I will determine whether the hawkmoth motor program is truly functionally coordinated by causally manipulating spike timing of a single muscle and observing if consistent behavioral outcomes result. Second, I will test whether this coordination is more likely performed through spike timing or spike phase, as well as determine the phylogenetic context of this thesis, by performing a comparative study of the motor program of several species of moth. Third, to test a major alternative hypothesis, that descending visual information is responsible for spike timing precision in the motor program, I will perform simultaneous recordings of the comprehensive motor program with population recordings in the ventral nerve cord, testing for precision using information-theoretic methods. Fourth, to study the dynamics of sensory integration and generate testable predictions, I will use data from previous aims to train dynamical systems models of motor program generation. Finally, I will test what wing reafference modifies in the motor program by performing controlled alterations such as removal of wing reafference or injection of noise into the wing strain field.