Sensorimotor memory and grip force control: does grip force anticipate a self-produced weight change when drinking with a straw from a cup?

摘要

We examined whether self-generated weight changes are anticipated by adequate grip force adjustments when repeatedly lifting an instrumented manipulandum. Subjects lifted a cup filled with 500 mL of water prior to and following drinking two portions of water with a straw without touching it. One half of the subjects drank from and lifted an uncovered cup receiving constant visual information about its filling level and the other half of the subjects drank from a covered cup without such visual feedback. During the lifts immediately following the drinking procedures, grip force scaling was erroneously programmed for the heavier weight of the preceding lift as was obvious from an inadequately high rate of grip force development. Vision had only a minor influence on the rate of grip force increase. The influence of vision on the scaling of peak grip force was more pronounced. More accurate force scaling was obtained with an increasing number of lifts performed under each weight condition, indicating an ongoing force adjustment process probably based on sensory feedback. We conclude that self-generation of a change in the weight of an object to be lifted is not, in itself, sufficient to elicit a predictive grip force output. Rather, accurate feedback information associated with the self-generated weight change is essential to update internal models related to the mechanical object properties. This assumption was confirmed in pilot experiments; when subjects lifted the cup after having poured water from it, they accurately scaled their fingertip force to the self-produced weight change. Here, direct sensory feedback from the grasping fingers could signal the weight change and update internal models while pouring water from the cup. Our data support the hypothesis that the sensorimotor system planning and processing predictive fingertip force can operate independently of higher-level cognitive and perceptual systems.