Neuromuscular reflex control of limb movement - validating models of the locusts hind leg control system using physiological input signals

Greater understanding of neuromuscular control of limb movements is vital for optimizing the treatment of patients with neuromuscular dysfunction. Locusts provide an opportunity to develop new investigative techniques and gain physiological insight into a related, but simpler and more accessible control system. Imposed movements of a stretch receptor in the locusts hind leg generate reflex responses in the motor neurons which control movements of the tibia. Previous work by Newland and Kondoh used a Wiener/Volterra model to represent the nonlinear responses of elements of this system to Gaussian white noise excitation. The interpretation of these models, however, is challenging both in terms of physiological relevance and statistical robustness. We have therefore extended the previous work to investigate the response of these models to physiologically realistic inputs. Our results show that the performance of the Wiener/Volterra models at predicting the response of the system to physiologically realistic inputs is poor. Investigation into this failure has allowed a simpler and more accurate model to be proposed.