Judith L. Smith

Forms of Forward Quadrupedal Locomotion. II. A Comparison of Posture, Hindlimb Kinematics, and Motor Patterns for Upslope and Level Walking

To gain further insight into the neural mechanisms for different forms of quadrupedal walking, data on postural orientation, hindlimb kinematics, and motor patterns were assessed for four grades of downslope walking, from 25% (14 degrees slope) to 100% (45 degrees), and compared with data from level and downslope walking at five grades (5-25%) on the treadmill (0.6 m/s). Kinematic data were obtained by digitizing ciné film, and electromyograms (EMGs) synchronized with kinematic records were taken from 13 different hindlimb muscles. At grades from 25 to 75%, cycle periods were similar, but at the steepest grade the cycle was shorter because of a reduced stance phase. Paw-contact sequences at all grades were consistent with lateral-sequence walking, but pace walking often occurred at the steepest grades. The cats crouched at the steeper grades, and crouching was associated with changes in fore- and hindlimb orientation that were consistent with increasing braking forces and decreasing propulsive forces during stance. The average ranges of motion at the hindlimb joints, except at the hip, were often different at the two steepest slopes. During swing, the range of knee- and ankle-joint flexion decreased, and the range and duration of extension increased at the ankle joint to lower the paw downward for contact. During stance the range of flexion during yield increased at the ankle joint, and the range of extension decreased at the knee and metatarsophalangeal joints. Downslope walking was also associated with EMG changes for several muscles. The hip extensors were not active during stance; instead, hip flexors were active, presumably to slow the rate of hip extension. Although ankle extensors were active during stance, their burst durations were truncated and centered around paw contact. Ankle flexors were active after midstance at the steeper slopes before the need to initiate swing, whereas flexor and extensor digit muscles were coactive throughout stance. Overall the changes in posture, hindlimb kinematics, and activity patterns of hindlimb muscles during stance reflected a need to counteract external forces that would accelerate angular displacements at some joints. Implications of these changes are discussed by using current models for the neural control of walking.

Predictions for neural control based on limb dynamics

Abstract Knowledge of limb dynamics and joint torques can provide unique predictions about neural mechanisms responsible for the control of limb motions. During unrestrained limb movements, mechanical interactions between articulated limb segments can significantly influence the resultant limb trajectories. To appreciate this mechanical complexity, quantitative modelling is necessary to determine how joint torques are affected by both muscle contractions and interactive forces between limb segments. Examples from non-weight-bearing, stereotypic movements of the cat hindlimb show that muscle contractions can either cause movements or counteract torques arising from mechanical interactions between segments. When interactive torques are substantial, motion-dependent feedback can monitor changes in the limb dynamics. Supraspinal centers, therefore, do not necessarily have to preplan compensatory muscular actions, and feedback at spinal segments allows for neural adaptation of the on-going limb dynamics.

Locomotion in exercised and nonexercised cats cordotomized at two or twelve weeks of age

The influences of an animals age at the time of cord transection and the use of ambulatory training on the recovery of weight-supported locomotion were assessed with 21 cats spinalized (T12) at 2 or 12 weeks of age. The animals were assigned at random to an exercise (E) or to a nonexercise (NE) group, and exercise therapy consisted of assisting the cat to walk on a motorized treadmill for 30 min, five times per week. During a 4-month recovery period, spontaneous clonus and hyperactive cutaneous reflexes developed in all animals, but skeletoarticular disorders were more typical of the younger animals. Age at cord transection had an effect on the recovery of weight-supported walking, as locomotion in the 2-E and 2-NE animals rated significantly higher than in the 12-E or 12-NE groups. For the younger animals, training was not critical to the development of locomotion, but for the older ones it was important. EMG records from soleus and lateral gastrocnemius muscles of spinal cats that achieved weight-supported locomotion revealed that recruitment of these muscles was normal despite significant changes in fiber composition and muscle contractility. In contrast, kinematic analyses of the locomotor patterns revealed gait abnormalities, including an absence of the yield phase during stance and an uncoupling of knee and ankle actions. We conclude that although the spinal cord contains the basic pattern-generating circuitry essential for locomotion, the recovery of this behavior was influenced by the animals age at the time of cord transection and to a less obvious extent by subsequent training.

Neuromuscular patterns of stereotypic hindlimb behaviors in the first two postnatal months. I. Stepping in normal kittens

Neuromuscular patterns associated with the development of hindlimb stepping behaviors were studied from birth to postnatal day 60 in normal kittens. Hindlimb muscles were chronically implanted with EMG electrodes at birth to characterize interlimb coordination and intralimb synergies during development of overground and treadmill stepping. Airstepping was also examined but seldom occurred after the second postnatal week. All kittens performed stepping under each condition, including weight-supported stepping, by postnatal day 3. The number of sequential steps on the treadmill and overground increased with age and cycle periods decreased. At onset, stepping behaviors were characterized by adult-like EMG patterns. Interlimb coordination was typified by alternating extensor bursts of similar duration. Extensors at the knee and ankle were coactive during the stance phase, and extensor burst durations were strongly correlated with the cycle periods over a wide range of stepping frequency. Ankle flexor and extensor muscles were reciprocally active during postural tremor, bouts of airstepping, and weight-supported steps on the treadmill and overground. The duration of the reciprocal flexor bust did not vary with cycle period or age. Observations of stepping behaviors and adult-like EMG patterns during initial postnatal development were contingent on optimal testing conditions. Taken together, the data suggest that pattern-generating circuits for regulating interlimb coordination and intralimb muscle synergies are potentially functional prior to the normal ontogenetic onset of locomotion. Perhaps the prolonged postnatal development of locomotion reflects the time required to establish adaptive mechanisms, such as postural control and agility, rather than spinal pattern-generating circuits for locomotion.

Gait-related motor patterns and hindlimb kinetics for the cat trot and gallop

To assess speed- and gait-related changes in semitendinosus (ST) activity, EMG was recorded from three cats during treadmill locomotion. Selected step cycles were filmed, and hip and knee joint kinematics were synchronized with EMG records. Swing-phase kinetics for trot and gallop steps at 2.25 m/s were compared for gait-related differences. Also, swing kinetics for different gallop forms were compared. With few exceptions, ST-EMG was characterized by two bursts for each step cycle; the first preceded paw off (STpo), and the second preceded paw contact (STpc). The two-burst pattern for the walk was defined by a high-amplitude STpo burst and a brief, low-amplitude STpc burst; at the slowest walk speeds, the STpc burst was occasionally absent. For the trot, the STpo burst was biphasic, with a brief pause just after paw off. With increasing walk-trot speeds, the duration of both bursts (STpo, STpc) remained relatively constant, but recruitment increased. Also, the onset latency of the STpo burst shifted, and a greater proportion of the burst was coincident with knee flexion during early swing. At the trot-gallop transition, there was an abrupt change in the two-burst pattern, and galloping was characterized by a high-amplitude STpc burst and a brief, low-amplitude STpo burst. At the fastest gallop speeds, the STpo burst was often absent, and the reduction in or elimination of the burst was associated with a unique pattern of swing phase kinetics at the knee. Knee flexion during the gallop swing was sustained by two inertial torques related to hip linear acceleration (HLA) and leg angular acceleration (LAA); correspondingly, muscle contraction was unnecessary. Conversely, knee flexion at the onset of the trot swing relied on a flexor muscle torque at the knee acting with an inertial flexor torque (LAA). Rotatory and transverse gallops at 4.0 m/s had similar swing phase kinetics and ST-EMG. Gait-related changes in ST-EMG, particularly at the trot-gallop transition, are not congruent with neural models assuming that details of the ST motor pattern are produced by a spinal CPG. We suggest that motor patterns programed by the spinal CPG are modulated by input from supraspinal centers and/or motion-related feedback from the hindlimbs to provide appropriate gait-specific activation of the ST.

Postcontraction changes in sensitivity of muscle afferents to static and dynamic stretch

Abstract The effects on dynamic and static components of the afferent stretch response from a muscle were examined in the period after a brief tetanus, which previous study has shown is marked by a persisting elevation in sensory discharge. Afferent activity of the cats triceps surae or medial gastrocnemius muscle was monitored as integrated multiunit discharge in a whole dorsal root, and the ventral roots were stimulated to produced a tetanus. The levels of activity at a static muscle length were found to be elevated in decreasing degree over several millimeters of lengthening beyond that at which the muscle was held during tetanus. During a ramp stretch, the velocity-related overshoot in dorsal root activity was also increased. Such changes were still detected after the elevated discharge had been made to disappear by momentary limited extension of the muscle. They are consistent with the hypothesis of residual effect on the intrafusal fibers, perhaps in the form of persisting actin-myosin crossbridges. Under sinusoidal stretch, the multiunit responses were generally decreased in the first few cycles following tetanus. Observations on single units indicated that the decrease was due chiefly to diminished responses of Ib afferents, which had incompletely recovered from adaptation brought on by excitation during the tetanus.

Speed-related changes in hindlimb intersegmental dynamics during the swing phase of cat locomotion

SummaryTo determine speed-related changes in hindlimb motion that might account for the mutability of bifunctional (hip extensor/knee flexor) muscle activity during the E1 phase of swing, we studied hip and knee joint kinematics and kinetics during swing over a ten-fold increase in locomotor speed (0.35 to 3.5 m/s). Three cats were filmed (100 frames/s) while locomoting on a motorized treadmill; kinematics were analyzed for the entire step cycle and kinetics for the swing phase. During swing, angular excursions at the hip and knee joints were similar for walking and trotting, but hip flexion and extension were significantly less after the transition from trot to gallop, while knee-angle range of motion increased during gallop phases E1, E2, and E3. During swing, knee-extension velocity peaked early in E1 and increased linearly with speed, while hip-flexion velocity peaked late in the flexion (F) phase and also increased linearly, but decreased precipitously at the trotgallop transition and remained constant as speed of galloping increased. Muscle torque directions during E1, flexor at the knee and extensor at the hip, were consistent with the proposed role of bifunctional posterior thigh muscles to decelerate thigh and leg segments for paw contact. At the knee joint, muscle torque during E1 counteracted a large interactive torque due to leg angular acceleration; the magnitudes of both torques were speed related with maximal values at the fastest speed tested (3.5 m/s). At the hip joint, muscle torque during E1 also counteracted a large interactive torque due to leg angular acceleration; the magnitudes of these two torques were speed related during the walk and trot, and like hip flexion velocity, decreased at the trot-gallop transition. Our data on speed-related changes in hindlimb dynamics suggest that the E1 burst amplitude (and perhaps duration) of posterior thigh muscles will be speed related during the walk and trot. After the trot-gallop transition at about 2.5 m/s, the recruitment of these bifunctional muscles may decline due to the changes in hindlimb dynamics. Because activity of these muscles counteracts interactive torques primarily related to leg angular acceleration, we suggest that motion-related feedback decoding this action may be important for regulating recruitment during E1.

Age-dependent capacity for somatosensory cortex reorganization in chronic spinal cats

Primary somatosensory cortex was mapped in chronic spinal cats that were spinalized (T12) at two weeks and 6 weeks of age. The magnitude of cortical reorganization is age-dependent. In cats spinalized at two weeks, extensive reorganization of the deafferented hindlimb region resulted in a second complete map of intact tactile input from the trunk and forelimb, while in cats transected at 6 weeks of age, trunk afferent input only partially activated the deafferented hindlimb region.

Responses of elbow extensors to landing forces during jump downs in cats

SummaryForces and displacements at the elbow joint have been related to EMG responses of flexor and extensor muscles during landing from jump downs at heights of 1.2 m to 0.6 m in five cats. Prelanding EMG activity consisted of two prelanding extensor bursts. Onset of both bursts was constant across all jump heights with reference to landing and not to take-off, occurring on average 73 ± 12 ms and 17 ± 8 ms prior to ground contact for the lateral triceps. Post-landing EMG activity was less than prelanding activity and was often packaged in three bursts, occurring on average at 18 ± 6, 34 ± 8 and 50 ± 9 ms after touchdown. Other measurements from extensor EMG including burst duration and integrated activity pre- and post-landing were also invariant. Across jump heights, maximum flexion angular velocity and elbow displacement were reached on average 28 ± 5 and 85 ± 7 ms postlanding, respectively. Although vertical (y) and horizontal (x) ground reaction forces increased with jump height, torque values at the elbow joint were not significantly different and were small in magnitude. At landing an animal typically experienced a 20 ms flexor torque (0.3 Nm/kg b.wt.) followed by an extensor torque (0.4 Nm/kg b.wt.) that continued for the major portion of elbow flexion.The temporal constancy of the kinematic and kinetic data and EMG activity across jump heights suggests that a generalized motor program can be used to activate extensor muscles at the elbow joint during the prelanding phase of self-initiated jumps. Since the onset of extensor activity is related to landing rather than to take-off, it is hypothesized that this activity is triggered by visual cues rather than by vestibular reflexes. After impact, adjustments for slight differences in post-landing torque about the elbow may be accomplished by intrinsic properties of the activated muscle as well as through segmental reflexes.

Visual and vestibular contributions to prelanding EMG during jump-downs in cats.

SummaryPrelanding EMG responses in elbow flexors and extensors were assessed during landing from jump-downs (0.6, 0.8, and 1.0 m) in normal blind-folded cats and labyrinthectomized cats with and without vision occluded. Jump-down conditions determined the strategy of response elicited in the normal cat. When the height could be anticipated by the blindfolded animals, a response typical of a jump in the presence of visual cues occurred; extensor activity began an average of 73±12 ms before landing, while flexor activity was minimal. When the animal was ‘tricked’ by an unexpected change in jump height, it displayed a pattern of EMG activity appropriate for the jump height just previously experienced, not for the actual height. If the jump height was uncertain, the cat commonly exhibited continuous extensor activity that began soon (100–150 ms) after both forepaws left the platform.In the presence of visual cues, labyrinthectomized cats were able to execute jump-downs at 0.6 m. Onset of extensor EMG activity was normal during the first postoperative jumps, although the typical two-burst plattern was absent, and average prelanding extensor IEMG was less than that of the control jumps. In addition, landing was usually awkward, as the forelimbs collapsed and the ventral surface of the trunk contacted the landing pad. During subsequent sessions, the two-burst pattern reappeared, average prelanding, extensor IEMG increased, and flight position improved, so that landing occurred without the forelimbs collapsing. Without visual cues, the labyrinthectomized cats were unable to execute a jump-down.These results suggest that visual input may normally regulate timing of the extensor prelanding motor program; however, without visual input, prelanding responses can be preprogrammed. When visual input is present, vestibular input during the flight phase may be primarily manifested as gain in extensor activity rather than in the temporal sequencing. When visual input is absent and jump height is uncertain, vestibular input may become more influential in determining the pattern of prelanding activity.