Doctoral Dissertation Defense: Nicole Massarelli
Advisor: Dr. Kathleen Hoffman
Location
Mathematics/Psychology : 401
Date & Time
June 8, 2016, 2:00 pm – 4:00 pm
Description
Title: Analysis of Sensory Feedback in the Lamprey Central Pattern Generator for Locomotion
Abstract
Sensory feedback is an integral part of the complex closed-loop system of locomotion. Lampreys are model organisms for vertebrate locomotion. Lamprey locomotion is driven by a central pattern generator (CPG), a circuit of neurons located in the spinal cord that produces a traveling wave of neural activity that innervates muscles for swimming. The CPG is capable of producing this activity independently of descending control or sensory feedback, however, sensory feedback is essential in responding to perturbations and adjusting the CPG output. Our goal is to model sensory feedback from edge cells, proprioceptor organs on the margin of the spinal cord, to the lamprey CPG in order to complete a closed-loop model of lamprey swimming.
Entrainment is a property of the CPG where the rhythmic output of the CPG will tend to approach the same frequency as a periodic stimulus. The lamprey CPG is modeled using a chain of coupled oscillators, where each oscillator corresponds to an anatomical segment of the spinal cord. We model entrainment experiments in two CPG models: a neural model where unit oscillators are represented by several classes of cells and a derived phase model where unit oscillators are represented by a single variable, its phase. In both models we included the effects of bending by including input from edge cells at the location of the forced oscillator. Both models required Asymmetry in the ascending and descending connection strengths in order to qualitatively match experimental entrainment data. Additionally, we showed that the model agreed with experimental results that the CPG is highly robust in response to large levels of noise added to the bending signal.
From a different perspective, we experimentally record edge cell activity, and analytically characterized edge cells responses to ramp bending experiments. While some cells respond to static stretch, the strongest responses are seen during the active periods of stretch when movement was occurring. We further used frequency domain techniques, for both the experimental data and model simulations, to compute a map, called the harmonic transfer function, from perturbations of the input signal to changes in the output signal. Results reveal that an under-damped harmonic oscillator with phase dependent forcing that depends on the sinusoidal bending captures key features of the experimental data, and thus, represents the mapping from bending to edge cell output.
Abstract
Sensory feedback is an integral part of the complex closed-loop system of locomotion. Lampreys are model organisms for vertebrate locomotion. Lamprey locomotion is driven by a central pattern generator (CPG), a circuit of neurons located in the spinal cord that produces a traveling wave of neural activity that innervates muscles for swimming. The CPG is capable of producing this activity independently of descending control or sensory feedback, however, sensory feedback is essential in responding to perturbations and adjusting the CPG output. Our goal is to model sensory feedback from edge cells, proprioceptor organs on the margin of the spinal cord, to the lamprey CPG in order to complete a closed-loop model of lamprey swimming.
Entrainment is a property of the CPG where the rhythmic output of the CPG will tend to approach the same frequency as a periodic stimulus. The lamprey CPG is modeled using a chain of coupled oscillators, where each oscillator corresponds to an anatomical segment of the spinal cord. We model entrainment experiments in two CPG models: a neural model where unit oscillators are represented by several classes of cells and a derived phase model where unit oscillators are represented by a single variable, its phase. In both models we included the effects of bending by including input from edge cells at the location of the forced oscillator. Both models required Asymmetry in the ascending and descending connection strengths in order to qualitatively match experimental entrainment data. Additionally, we showed that the model agreed with experimental results that the CPG is highly robust in response to large levels of noise added to the bending signal.
From a different perspective, we experimentally record edge cell activity, and analytically characterized edge cells responses to ramp bending experiments. While some cells respond to static stretch, the strongest responses are seen during the active periods of stretch when movement was occurring. We further used frequency domain techniques, for both the experimental data and model simulations, to compute a map, called the harmonic transfer function, from perturbations of the input signal to changes in the output signal. Results reveal that an under-damped harmonic oscillator with phase dependent forcing that depends on the sinusoidal bending captures key features of the experimental data, and thus, represents the mapping from bending to edge cell output.
Tags:
