Neural arithmetic of motion vision

ended

Signal processing by neurons involves arithmetic operations, many of which are nonlinear. These nonlinearities are thought to account for much of the brain’s computational power, yet, little is known about the molecular mechanics that underlie even simple operations like multiplication and division in individual neurons. I will address this problem in the visual system of Drosophila, where the detection of motion represents a canonical example of nonlinear information processing: To perceive visual motion, the signals of adjacent photoreceptors are differentially delayed to coincide at the dendrites of the bushy T-cells, T4 and T5, where multiplication and/or division of these inputs are thought to give rise to direction selectivity. The proposed work will address the following questions: Are the dendritic transformations multiplicative, divisive, or both? Which synaptic inputs constitute the numerator and which the denominator? What receptors and ion channels account for the dendritic nonlinearities and how do they influence the visual perception of motion? The fruit fly enables me to bridge molecular biophysics and optomotor behaviour by granting genetic, electrical, optical, and molecular access to virtually all relevant neurons in a circuit of well-documented synaptic connectivity and transcriptional activity. I will use whole-cell patch clamp recordings in vivo, patch-seq, optogenetics, optomotor behaviour, RNA interference, and computational modelling to identify the biophysical basis of multiplicative and divisive operations in the dendrites of bushy T-cells. The answers will likely hold information that extends beyond the fly’s sense of sight and might uncover previously unknown ways of signal processing by single neurons at the molecular scale.