Work

Multisensory integration and decision-making in Drosophila larvae

Year: 2016
Type: dissertation
Abstract: Numerous studies have shown that a wide range of behaviors from sensory processing to motor control involve near-optimal probabilistic inference. Most of these studies have focused on vertebrates, suggesting that the ability to perform probabilistic inference requires large nervous systems. Yet, neural theories of probabilistic inference can be implemented with the most basic neural networks. To explore the possibility that organisms with small nervous systems perform near-optimal probabilistic inference, I tested the ability of Drosophila larvae to integrate information from unisensory and multisensory cues. Larvae were placed in a circular behavioral arena, where their positions were monitored during exposure to single or combined sensory gradients. Combined gradients consisted either of two odor gradients or a thermosensory and an odor gradient, to test within and cross-modal integration respectively. In collaboration with theorists, I predicted the optimal behavior for the combined gradients given the behavior in the single sensory conditions, with a Bayesian model. The behavior of the larvae matched the predictions of the Bayesian model closely for both, the within and cross-modal, scenarios of integration. Another suboptimal model with fixed weights failed to predict the combined behavior. This work sets the stage for a systematic analysis of the neural computations underlying probabilistic inference in an insect brain amenable to genetic manipulations and physiological inspections.
Language: English
Field: Psychology
Open Access status: green
OpenAlex ID: W2806330388
Cites: 172
Cited by:
Related to: 10