Abstract
Converging evidence suggest that actions do not obey Weber's law. Indeed, in grasping movements, the variability of the maximum grip aperture is unaffected by variations in object size, in striking contrast with the variability of perceptual estimates which linearly increases with stimulus magnitude. The origins of this perception-action dissociation are still a matter of discussion and various explanations have been put forward, such as the different coding of visual size information for perception and action, the use of positional information to control the movement of the digits in space, the impact of biomechanical factors or the influence of sensorimotor calibration. However, the absence of Weber’s law has been found exclusively in grasping tasks, and it still is unclear whether this violation extends to the whole action domain. To address this contentious debate, we examined Weber's law in two experiments, by means of a two-finger pointing task. Participants reached and simultaneously touched with their thumb and index finger two targets positioned at different distances from each other. In Experiment 1, participants performed the movements with full vision (closed-loop), whereas in Experiment 2, visual information was removed just after the beginning of the movement (open-loop). Closed-loop movements displayed the typical overall reduction in variability with respect to open-loop movements. However, in accordance with Weber's law, we found that the variability of the inter-finger separation at the moment the fingers contacted the targets increased with larger inter-target distances in both closed-loop and open-loop conditions. These results suggest that Weber's law is not confined to the perceptual domain, but it is regularly at play when we consider measures that are directly related to the goal of the action.