Multimodal Convergent Information Enhances Perception Accuracy of Human Movement Patterns

Previous neurophysiological and neurobiological research has revealed different areas of multisensory integration in the CNS. Behavioral consequences of multimodal convergent stimuli arrays (visual/auditory/ somatosensory) have been observed on different kinds of higher mammals (cats, monkeys). Multisensory convergence neurons output directly affects motor behavior: Depending on convergent or divergent input, extensive enhancement or depression in neural activity can be induced in multisensory neurons, resulting in varying forms of overt attentive and orientation behavior, as seen on cats (Stein/Meredith 1993). Areas of multisensory integration had also been detected within the human cortex (e.g. the heteromodal cortex in the left superior temporal sulcus), and crossmodal processing appears to have parallels in behavioral features as discussed by Stein (1998) and now shown with functional magnetic resonance imaging (fMRI) by Calcert et al. (2000) on bimodal speech perception processes. Obviously, mechanisms of multisensory integration are not limited to temporal spatial convergent stimuli and behavioral effects are not limited to determination of stimulus location: They play a role even on temporal spectral-dynamic convergent input and stimulus identification (Calvert 1998). The aim of this investigation was to determine whether mechanisms of multisensory integration even could be used to enhance perception accuracy of human gross motor patterns under real-world like conditions. Subjects were asked to judge the height of different counter-movement jumps of a person under three conditions. Visual treatment (VT) was implemented as a video-projection of the athlete performing jumps between 20 cm and 45 cm. Audio-visual treatment (AVT) consisted of VT with additional sonification of the vertical ground reaction force, so visual-auditory convergence was achieved for the entire kinematic pattern. Auditory treatment (AT) was equivalent to sonification alone. All subjects heard the continuous movement sonification for the first time. No feedback was given and order of treatments was randomized. Absolute error (AE) of VT and AT showed no significant difference, though perception accuracy was less under auditory condition (AE VT : 4.41 cm; AE AT : 4.70 cm; p = .146). Perception of bimodal convergent stimuli array results in higher precision: AE AVT decreased to 3.48 cm, differences to both single modal conditions were statistically significant (p = .000, both). Results demonstrate that additional convergent auditory information enhances perception accuracy of gross motor movements. An enhanced informational base should support processes of motor learning even as motor memory.