Mechanosensory inputs to the central pattern generators for locomotion in the lamprey spinal cord: resetting, entrainment, and computer modeling.

1. Mechanoreceptors in the lamprey spinal cord have inputs to the central pattern generator (CPG) for locomotion. In the present study, imposed sinusoidal and pulsed movements were applied to the end of the in vitro lamprey spinal cord to excite the mechanoreceptors so that the relationship between entrainment and resetting of the locomotor rhythm could be examined. 2. The range over which the locomotor rhythm could be entrained by sinusoidal movements was asymmetric and occurred mostly at movement cycle times below the resting cycle time. During entrainment at the shortest cycle times, the movement phases were relatively small. 3. The phase response curves (PRCs) displayed the greatest shortening of cycle times (phase advance) for movement pulses applied during the first half of the locomotor cycle, whereas movement pulses applied during the second half of the cycle were largely ineffective. The amplitude of phase shifts in the PRC correlated with the ranges of cycle times over which entrainment occurred. 4. During resetting from movement pulses applied early in the cycle, the burst and interburst parts of the cycle shortened by about the same percentage. In addition, resetting effects occurred simultaneously along the spinal cord, suggesting a rapid distribution of timing information. 5. A computer model of the CPGs, consisting of left and right oscillators and inputs from mechanosensory elements, produced entrainment ranges that were symmetric around the resting cycle time. The PRCs from the model showed phase advance for movement pulses applied during the first half of the cycle and phase delay for pulses applied during the second half of the cycle. 6. Because of the asymmetric experimental PRCs for the lamprey spinal cord, gating was incorporated into the cooffter model such that oscillators on one side of the model gated inputs from mechanosensory elements on the same side. With gating, the model produced entrainment ranges that were asymmetric and confined to cycle times below the resting cycle time. The PRCs still showed phase advance for pulses applied at the beginning of the cycle, and the amount of phase delay produced during the second half of the cycle was substantially reduced compared with the simulations without gating.