A central challenge in neuroscience is to connect an empirical description of a neural circuit's activity to its role in guiding behavior or supporting perception. In the auditory pathway, our models of neural computations stem from characterizing receptive field properties of neurons along the pathway from periphery to brainstem and cortex. However, neural responses can contain more information about the stimulus than our models predict. Indeed, without precise control over neural activity, testing how well a model predicts a circuit's functional importance can be difficult. With a relatively small, tractable auditory system, Drosophila melanogaster is an excellent model for studying the neural encoding of auditory stimuli. The powerful genetic tools available for Drosophila can allow us to investigate how specific neural computations underlie hearing and communication. Drosophila detects and responds to natural sounds, and, like humans, use sound to communicate, specifically during courtship. Here I propose to investigate auditory encoding in class of cells called B1 neurons. B1 neurons are responsive to sound and have been shown to be critical for eliciting behaviors that are dependent on courtship song, but little is known about their physiology or responses to auditory stimuli. I will perform experiments to address two specific questions: 1) what features of sound stimuli do B1 neurons encode, and 2) how do molecular and network mechanisms shape B1 neuron sound responses? I will first target genetically labeled B1 neurons for electrophysiological recordings to characterize their responses to sound. I will then genetically restrict auditory function to distinct populations of primary sensory neurons in order to determine the source of B1 input. Finally, I will test how cell-intrinsic properties contribute to B1 sound responses by genetically manipulating their intrinsic excitability. Together, these aims seek to test whether the population of B1 neurons exhibit diverse but complementary functional properties that permit computation of abstract features of sound.
Hearing loss is the most prevalent sensory deficit in developed countries, and efforts to improve interventions and therapies, such as encoding algorithms in cochlear implants, will benefit from a detailed understanding of neural computations at successive levels of auditory processing. With a compact auditory system and powerful genetic tools allowing superior control over neural circuit activity, Drosophila melanogaster has proved an important model for sensory transduction and coding in the past, and because flies use sound to accomplish similar tasks as other organisms, the fly promises to be a useful model system for auditory research and for investigating the influence of genes on audition dependent behavior. A quantitative description of the compact Drosophila auditory system may also lead to improvements in sound compression algorithms, voice recognition, machine learning, and the construction of novel microphones and neural prostheses.
| Azevedo, Anthony W; Wilson, Rachel I (2017) Active Mechanisms of Vibration Encoding and Frequency Filtering in Central Mechanosensory Neurons. Neuron 96:446-460.e9 |
