Adaptive control for backward quadrupedal walking VI. metatarsophalangeal joint dynamics and motor patterns of digit muscles.

1. We compared the dynamics of the metatarsophalangeal (MTP) joint of the cat's hind paw and the motor patterns of two short and four long muscles of the digits for two walking forms, forward (FWD) and backward (BWD). Kinematic (angular displacements) data digitized from high-speed ciné film and electromyographic (EMG) data were synchronized and assessed for bouts of treadmill walking. Kinetic data (joint forces) were calculated from kinematic and anthropometric data with the use of inverse-dynamic calculations in which the MTP joint net torque was divided into gravitational, motion-dependent, ground contact (absent for swing), and muscle torque components. Swing-phase kinetics were calculated from treadmill steps and stance-phase kinetics from overground steps in which one hind paw contacted a miniature force platform embedded in the walkway. 2. The plantar angle at the intersection of the metatarsal and phalangeal segmental lines was used to measure MTP angular displacements. During swing for both walking forms, the MTP joint flexed (F) and then extended (E); however, the F-E transition occurred at the onset of FWD swing and at the end of BWD swing. For FWD walking, the MTP joint extended at a constant velocity during most of stance as the cat's weight rotated forward over the paw. During the unweighting phase at the end of stance, the MTP joint flexed rapidly before paw lift off. For BWD walking, the MTP joint extended briefly at stance onset (similar to a yield) and then flexed at a constant velocity as the cat's weight rotated backward over the paw. At the end of stance, the MTP joint extended and then flexed slightly as the paw was unweighted before paw lift off. 3. For both forms of walking, three of the six muscles tested were recruited just before paw contact and remained active for most (75-80%) of stance for both walking forms: plantaris (PLT), flexor hallucis longus (FHL), and flexor digitorum brevis (FDB). Their recruitment contributed to the flexor muscle torque at the MTP joint during most of FWD and BWD stance and was responsible for the absorption of mechanical power at the MTP joint for FWD stance and generation of mechanical power at the MTP joint during BWD stance. Also, the FHL and PLT, along with the soleus (SOL; also recorded in this study), contributed to an extensor muscle torque (described in paper IV of this series) and the generation of mechanical power at the ankle joint during stance of FWD and BWD walking. 4. The timing of activity for three muscles recruited during FWD swing was distinct for the two walking forms. The hallmark burst of the flexor digitorum longus (FDL)--a single burst, brief in duration and high in amplitude--occurred at the end of FWD swing (as the toes flexed rapidly) but shifted to the onset of BWD stance (as the claws protruded and toes extended) during paw weighting. The extensor digitorum longus (EDL) was recruited after paw off and was active for most of FWD swing; its activity contributed to an extensor muscle torque at the MTP joint and a flexor muscle torque at the ankle joint. For BWD walking, EDL recruitment shifted to an earlier phase in the step cycle and coincided with toe extension, which occurred at the end of stance before paw lift off. This pre-lift off activity continued into the first part of swing and contributed to an extensor muscle torque at the MTP joint and a flexor muscle torque at the ankle.(ABSTRACT TRUNCATED AT 250 WORDS)