Sensing wing deformation: it's about spike time

Abstract

Flying insects rely heavily on mechanoreception for flight control. True flies use halteres to provide feedback on body rotations. Recent behavioral and theoretical evidence suggest that wings, antecedents of the halteres, might also serve a similar sensory function. Both wings and halteres have mechanoreceptors called campaniform sensilla. In halteres those sensors are believed to be directionally sensitive, allowing them to detect exceedingly tiny out of plane deformations that arise from Coriolis forces. Similarly, sensilla could detect tiny deformations that are present as torsion in flapping wings subject to simultaneous body rotations. We combine theoretical and experimental analyses of flapping wings in the hawkmoth Manduca sexta . Like halteres, their sensilla respond to local strains which result from the combination of inertial, aerodynamic and gyroscopic forces acting on the wings. Their spike timing is exceedingly precise with sub-millisecond variability in response to mechanical stimuli. We ask whether the nervous system could determine body rotations based on spike arrival times, not requiring high directional sensitivity of individual sensilla. Combining structural simulation and electrophysiological measurements, we find that the resulting spike timing differences depend linearly on rotation rate. Moreover, rotation rates above 2130 ° s^(-1) result in spike timing differences larger than the median 160 microsecond spike time variability of single units. These data suggest an alternative hypothesis for the neural basis of gyroscopic sensing, based on spike timing rather than on extreme directional sensitivity of campaniform sensilla.

Additional details