Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.806148
Title: Modelling the neuromechanics of exploration and taxis in larval Drosophiila
Author: Loveless, Jane
Awarding Body: University of Edinburgh
Current Institution: University of Edinburgh
Date of Award: 2020
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
Abstract:
The Drosophila larva is emerging as a useful tool in the study of complex behaviours, due to its relatively small size, its genetic tractability, and its varied behavioural repertoire. The larva executes a stereotypical exploratory routine that appears to consist of stochastic alternation between straight peristaltic crawling and reorientation events through lateral bending. The larva performs taxis by biasing this behavioural pattern, allowing it to move up or down attractive and aversive stimulus gradients. Existing explanations of exploration and taxis behaviour often neglect the larva's embodiment, focusing on central pattern generation and decision making circuits within the nervous system. In Chapter 1 of this thesis, I review the current state of knowledge regarding larval peristalsis, exploration, and taxis behaviours, as well as existing theories of their generation. I argue that an understanding of the animal's embodiment should lead to a deeper understanding of its behaviour. In Chapter 2, I present a model of the axial mechanics of the larva, and demonstrate how the animal's body physics can be exploited to produce peristalsis by using segmentally localised, positive feedback of strain rate. The mechanical model includes viscoelastic tissue mechanics, muscular inputs, and substrate interaction while sensory feedback is modelled as a linear feedback control law. In Chapter 3, I extend the mechanical model to study motion in the plane, including both axial and transverse deformations of the body. The feedback law is replaced by a simple model of the larval nervous system. The model includes both a segmentally localised reflex arc as well as long-range, mutual inhibition between segments. The complete model is capable of generating both peristalsis and spontaneous reorientation, leading to emergent exploration behaviour in the form of a deterministic superdiffusion process grounded in the chaotic mechanics of the larva's body. In Chapter 4, I consider taxis behaviour. I introduce a transverse reflex capable of modulating the effective transverse viscosity of the larval body. When the larva is experiencing an increasing attractive (aversive) stimulus, the reflex acts to increase (decrease) the effective transverse viscosity, causing bending to occur less (more) easily. As a result, the model larvae approach attractive stimuli and avoid aversive stumuli. On a population level, I show that the transverse reflex can be thought of as biasing the model animals towards sub- or super-diffusion. I compare the statistics of this behaviour to those of the real larva. In Chapter 5, I shift focus to engineered soft systems. Having successfully deployed an energy-based modelling approach in Chapters 2--4, I argue for the adoption of an energy-focused (specifically, port-Hamiltonian) approach within the field of soft robotics. In Chapter 6, I present some initial theoretical extensions to the models presented in chapter 2--4. I first focus on the mechanics of self-righting and rolling behaviours, before modelling the ventral nerve cord of the larva using a ring attractor architecture. Finally, in Chapter 7, I summarise the results of the previous chapters and discuss directions for future research.
Supervisor: Webb, Barbara ; Stokes, Adam Sponsor: Engineering and Physical Sciences Research Council (EPSRC)
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.806148  DOI:
Keywords: drosophila ; larval peristalsis ; larval behaviour ; spontaneous reorientation ; superdiffusion ; port-Hamiltonian approach ; energy-based modelling ; self-righting ; modelling
Share: