Meeting Abstract

S1-3.11  Jan. 5  Testing Control Models In Rapid Running Insects Using Lateral Ground Translation REVZEN, Shai*; BISHOP-MOSER, Joshua; SPENCE, Andrew; FULL, Robert/J; Univ. of California, Berkeley; Univ. of California, Berkeley; Univ. of California, Berkeley; Univ. of California, Berkeley shrevz@berkeley.edu

Perturbation of simple passive, dynamic models of legged locomotion suggest the possibility of self-stabilization with minimal neural feedback. Rapid recovery from brief impulses to the body of fast, sprawled-posture runners and the absence of muscle activation pattern changes while traversing rough terrain support the hypothesis of recovery by mechanical feedback alone. Large and complex perturbations to rapid running insects imposed by a single, hip-high hurdle do produce significant leg phase and frequency changes showing that sensory feedback must play a role in recovery. To better determine the interrelationship between neural and mechanical feedback, we designed a trackway with a 10 x 25 cm platform insert that could translate laterally to a maximum acceleration of 10g in 50 msec. Cockroaches (Blaberus discoidalis; n=14) running at 30±8 cm/sec at a step frequency of 11.5±2.7 Hz onto a movable platform were accelerated laterally at 1g in a 100 msec interval providing a 56±3 cm/sec specific impulse. By automatically tracking body position and orientation and leg (tarsus) positions, we found no change in leg motion timing for at least 50 msec. Following this delay, animals decreased step frequency for one stride, and then partially recovered frequency thereafter. Results are consistent with previous research showing that the initial rapid recovery is accomplished by mechanical feedback promoting self-stabilization followed by neural feedback modulation of a central pattern generator at a slower rate occurring after a delay comparable to the duration of a step. Funded by NSF FIBR Grant.