The effect of externally generated loading on predictive grip force modulation

Abstract

A characteristic of skilled movement is the ability of the CNS to predict the consequences of motor commands. When we lift an object there is an anticipatory increase in grip force that prevents a grasped object from slipping. When an object is pulled from our grasp by an external force, a reflexive modulation in grip force prevents slippage. Here we examine how external perturbations to a grasped object influence anticipatory grip force during object manipulation using a bimanual task, with each hand holding a computer-controlled object. Subjects were instructed to maintain the position of the object held in the right hand. Loading was applied to this restrained object: either self-generated by the action of their left hand or externally generated by a motor. The magnitude of the grip force response to self-generated loading increased after the object was loaded, and the latency of this response remained predictive of load force. This implies that external and self-generated loading increase the anticipatory grip force response. Unlinked trials, where the subject's moved their left hand but no loading was generated on the right-hand object were used to assess the presence of purely predictive control of grip force. External loading soon after self-generated loading maintained an existing predictive response once the linkage between the subject's action and object loading had been removed. However, external loading had no influence as the existing prediction decays. Therefore, the predictive grip force response during object manipulation can be significantly modified by object loading from an external source.

Fig. 1

Schematic diagram of the apparatus used to create objects with novel dynamic relations. Each hand held a computer-controlled object with a force-transducer embedded and attached to a torque motor. The subject was required to pull on the fixed left object and to maintain the position of the object held in the right hand. The objects could be either “linked”, so they acted together, or “unlinked”, so that they acted as two independent objects. On external trials, the motors pulled on the right-hand object.

Fig. 2

Average across the subjects’ mean grip force (N) for trials with a significant influence of trial history, where _LU means the modulation of unlinked trials which follows linkage (and similarly for other trial types) (N = 6). (a and b) Condition Linked–Unlinked; (c and d) condition Linked–External; (e and f) condition External–Null. The vertical dashed line is the time of peak load force (or for the unlinked trials time of expected peak load force). Subjects’ mean grip force modulation (N) for each of the three conditions dependent on the previously experienced trial with all trial types shown (a and b) condition Linked–Unlinked c–d) condition Linked–External. (e and f) Condition Null–External. Error bars show S.E. individual means (N = 6) ((*) significant at p < 0.05; (***) significant at p < 0.005).

Fig. 3

Average across subject's mean grip force on successive unlinked trials in a series where external loading was after (a) first unlinked trial (N = 6); (b) fifth unlinked trial (N = 5) c.f. previous work . (c) Average across subject's mean grip force modulation (same data series as 3a and 3b) External after first unlinked (dark), after fifth unlinked (pale). (d) Average across subject's mean change in (d) grip force modulation (N). External trial present between the pair (dashed line). No external trial present between the pair (solid line). Error bars show S.E. across subject means.