Manipulating an Equivocal World - How Force Control and Weight Perception use Probabilistic Information
Choosing an adequate motor performance can be considered as a decision-making process based on probabilities. The present cumulative dissertation aimed at investigating how probabilistic information about an object’s weight is used to modify expectations about object properties and how they impact the predictive scaling of fingertip forces. Healthy volunteers and subjects with Parkinson’s disease (PD) were studied using a novel paradigm in the context of two tasks. Subjects saw four different probabilistic visual cues, each of which announced the likely weight of an object. In the grip-lift (GL) task an object was grasped and lifted with a pinch grip after the probabilistic cue was presented. Effects of probabilistic information on predictive force scaling were analyzed by comparing the trials where the same actual weight was lifted, yet pre-ceded by four different probabilistic cues. In the weight perception (WP) task participants gradually adjusted the object’s weight until its heaviness matched the expected weight for a given probabilistic cue. Findings of healthy subjects reported in this thesis revealed that probabilistic cues influenced weight adjustments in the WP task and peak force rates in the GL task in a similar manner. The interpretation and utilization of the probabilistic information was critically influenced by the initial task. PD subjects on the other side, used the same probabilistic information to evaluation the perceptual weight, but their force prediction remained unaffected by the cues. Findings reported in this thesis are in line with aspects of the ‘dual-stream theory of the visual system’ and the opposing ‘common-coding approach’ but they do not favor one of them. Rather they indicate that specific requirements of a task critically influence the interpretation and use of probabilistic information in perception and action.