Modeling the spinal cord neural circuitry controlling cat hindlimb movement during locomotion

We have developed a computational model of the spinal cord neural circuitry that controls locomotor movements of simulated cat hindlimbs. The neural circuitry includes two central pattern generators integrated with reflex circuits. All neurons were modeled in the Hodgkin-Huxley style. The musculoskeletal system includes two three-joint hindlimbs and the trunk. Each hindlimb is actuated by nine one- and two-joint muscles (a Hill-type model). Our simulations allow us to suggest a specific network architecture in the spinal cord and a pattern of feedback connectivities (from Ia and Ib fibers and touch sensors) that provide stable locomotion and realistic patterns of muscle activation and kinematics of limb movements. (C) 2002 Elsevier Science B.V. All rights reserved.