Sensor Fusion for Feedback Control of Gaze Stabilisation in the Fly
Final rept. 30 Mar 2009-30 Mar 2010
CAMBRIDGE UNIV (UNITED KINGDOM) DEPT OF ZOOLOGY
Pagination or Media Count:
Our results show that the fly gaze stabilisation reflex in general has the properties of a low-pass filter - the shape of the gain and phase plots illustrate this perfectly. However, two parameters of the filter - corner frequency, cut-off - are not defined by the motor system, but change according to the available sensory information. Low-frequency performance is maintained through sensory input from the compound eyes, while the addition of sensory input from the halteres smoothly extends the range of the reflex to much higher frequencies. Interestingly, sensory input from the ocelli gives a small advantage across the entire frequency range whereas, given the phasic nature of ocellar inputs to VS neurons, we would have expected little or no difference at frequencies where the period of the waveform approaches 500ms. In general, size of the reduction in latency of the headroll reflex with ocellar input matches what we would expect from previous electrophysiology experiments. In earlier studies, the authors characterised the headroll reflex in terms of sensitivity at different angular velocities of motion. Here we describe the reflex in terms of frequency response - these data are essential so that we can begin to model the flys control system with grasp of real physical parameters. Now we have a more complete picture of the filtering properties of the entire reflex, as well as the individual sensors - the compound eyes, ocelli and halteres - the next step in this process will involve measuring some of the physical properties of the motor system, while further probing the integration of sensory signals at the motor neuron level. While continuing to expand our description of the headroll reflex, we also intend to extend our study to the flys body stabilisation reflex. The dynamics of whole-body stabilisation are completely different to those of head stabilisation, so we would expect this to be reflected in the frequency response of the flight stabilisation reflex.
- Anatomy and Physiology