The regulation of limb stiffness in the context of locomotor tasks

Abstract

Locomotion on ramped surfaces requires modulation of both pattern generating circuits and limb stiffness. In order to meet the mechanical demands of locomotion under these conditions, muscular activation patterns must correspond to the appropriate functions, whether the muscles are serving as force generators or brakes. Limb stiffness is a critical mechanical property that determines how the body interacts with the environment, and is regulated by both intrinsic and neural mechanisms. We have recently investigated how pattern generation, stiffness and proprioceptive feedback are modulated in a task specific way using the decerebrate cat preparation. Our results confirm previous research using intact animals that during level and upslope walking, hip and ankle extensors are recruited for propulsion during stance. During downslope walking, hip extensors are inhibited and hip flexors are recruited during stance to provide the needed braking action. Our new data further show that endpoint stiffness of the limb is correspondingly reduced for walking down a slope, and that the reduction in stiffness is likely due to an increase in inhibitory force feedback. Our results further suggest that a body orientation signal derived from vestibular and neck proprioceptive information is responsible for the required muscular activation patterns as well as a reduction in limb stiffness. This increased compliance is consistent with the function of the distal limb to cushion the impact during the braking action of the antigravity musculature.

Figure 1.

The left panel shows a conceptual model providing the origin of a body orientation signal. Physical interactions are shown in blue. The orientation of the head is detected by the otolith organs. Proprioceptors in the neck muscles detect neck angle, which is determined by the difference between head and body orientations. The neck angle and head orientation are recombined through neural pathways to provide an estimate of body orientation, which is distributed to pattern generating networks in the spinal cord as well as proprioceptive circuits. Movements of the head alone result in cancellation of the two signals. The right panel depicts the result of destruction of the vestibular apparatus. In this case, movements of the head alone lead to modulation of pattern generation and proprioceptive feedback. MLR: the mesencephalic locomotor region.

Figure 2.

Evidence that a body orientation signal regulates the pattern generating networks in the spinal cord. Electromyographic activity of three selected muscles in the step following tilting of the head up are shown, for level walking (upper traces) and for walking down a slope (lower traces). The anatomical sketch indicates the approximate locations of the posterior biceps and iliopsoas muscles. Note that the hamstrings muscle posterior biceps was not activated during stance in the downslope condition, and the hip flexor iliopsoas became active during stance. The change in pattern corresponded to a switch from propulsion to braking between the two tasks. Although an alteration in head position alone should evoke no change in muscular patterns since body orientation is constant, the change occurred because neck afferent input is more rapid than vestibular input. This pattern of activity reverted back to the level walking state within 2–3 steps as vestibular and neck afferent feedback finally cancelled. After destruction of the vestibular system, the change in pattern remained in effect as long as the head was tilted (data not shown).

Figure 3.

Demonstration that stiffness measured at the endpoint of the limb is reduced for the spinal state corresponding to walking down a slope (head tilt up following labyrinthectomy). Traces represent force trajectories in response to 2 cm perturbations of the limb imposed by a robotic arm along the long axis of the limb in the direction of limb compression. A large number of responses were collected over a range of background forces, and the means for the two conditions (head level and head up) are shown. The main effect of condition is statistically significant in the shaded region.

Figure 4.

Evidence that the change in limb stiffness after head tilt in a labyrinthectomized, decerebrate animal is due at least in part to modulation of inhibitory force feedback. The central panel represents the control condition with the head level. Closed circles represent force responses (obtained at the end of a ramp stretch (see inset, solid lines; vertical grey bar denotes time of measurement)) of the flexor hallucis longus muscle (FHL) obtained at different background forces (group 1). Force was modulated using a crossed-extension reflex, a response that naturally habituates with time, providing a range of background forces, and was obtained by electrical stimulation of the tibial nerve in the contralateral limb. Open diamonds represent the responses of FHL when the gastrocnemius muscles were stretched simultaneously (group 2, dashed responses in insets). The data points corresponding to the force trajectories shown in the insets are denoted by open circles and closed diamonds, respectively, for groups 1 and 2. The lines signify quadratic least squares fits to the data points. The extent of inhibition corresponds to the difference between the two fitted lines. The left panel depicts the condition in which the head was tilted down, corresponding to walking up a slope. Note that the magnitude of inhibition was reduced across the range of forces. The right panel depicts the condition for down slope walking. Note that inhibition was greater than in the other two conditions. In this experiment, the animal was not stepping, indicating that the modulation of force feedback can occur under steady postural conditions as well.