Abstract
Postural adjustments to imposed (passive) and voluntary (active) unloading conditions of the forearm were studied in normal subjects and a deafferented patient. The latency of the postural behaviour (deactivation of the biceps supporting the weight) was linearly related to the displacement amplitude of the unloaded forearm, independent of the unloading conditions. The postural behaviour consisted of an anticipatory postural adjustment (APA) occurring prior to active unloading (in both normals and the patient) and conversely in an unloading reflex response following passive unloading (only in the normals). In both the deafferented and the normal subjects, the amplitudes of the displacement during active unloadings were much smaller (3×) than in the passive conditions and an APA was present in both the deafferented and the normal subjects. The APA could not be triggered by some types of active movement and was absent when the movement was not directly producing the unloading. The EMG latencies of the APA and of the contralateral muscles used to unload were tightly coupled. However, the latency would sometimes be decoupled, particularly when a temporal delay was introduced between the active movement and the unloading in normal subjects. In contrast to the normal subjects, who were able to adapt quickly to an unusual unloading condition (produced by voluntary knee flexion), the deafferented patient did not show an APA in this task. It was concluded that, although the APA is of central origin, it cannot be generated only on the basis of internal timing cues and must rely on afferent information for its generation during unfamiliar unloading conditions.Key words: posture, bimanual coordination, deafferentation, open-loop control.