Integration of visual and proprioceptive afferents in kinesthesia

Abstract

Proprioceptive signals are of prime importance in kinesthesia. However, in conditions of visuo-proprioceptive conflicts, strong visual-evoked biases can be observed. In three experiments, we parsed the interaction between visual and proprioceptive afferents using the 'mirror box' paradigm. Participants' left arm, the image of which was reflected in a mirror, was passively moved into flexion/extension or remained static. In Experiment 1 proprioceptive afferents of the unseen static right arm were masked with diffuse arm vibration. In Experiments 2 and 3, afferent signals were enhanced by muscle vibration of biceps or triceps stretch receptors. Illusory arm movements were evaluated with subjective reports and matching adjustments. Results revealed that participants did not experience kinesthetic illusions when the mirror reflected the image of a static arm while proprioceptive afferents conveyed signals of a moving arm (Experiment 2). In this specific case, vision apparently contributed much more strongly to the final percept than proprioceptive signals. However, in most circumstances, the percept reflected integration of both afferent signals (Experiments 1-3). For instance, when both sensory channels conveyed signals of arm displacement but in the opposite direction, kinesthetic illusions occurred but were either proprioceptively (vibration illusion) or visually driven (mirror illusion), according to individual sensorial preferences (Experiments 2 and 3). These results indicate that kinesthesia is the product of cooperative integration processes in which the final percept strongly depends on the experimental conditions as well as sensorial preferences. The observed changes in the relative contribution of each input across experimental conditions likely reflect reliability-dependent weights.