Perceptual learning of bimanual coordinated rhythmic movements: Information matters more than movements
Jack Leech & Andrew D. Wilson
Prior to training, only two coordinated rhythmic movements are stable: 0° and 180°. Other coordinations (e.g. 90°) must be learned. This pattern emerges from a task dynamic in which relative phase is perceived as the relative direction of motion, modified by the relative speed (Bingham, 2004). People can learn how to move at 90° but this entails learning to use a different information variable (relative position; Wilson & Bingham, 2008). Learning a novel coordination requires feedback; typically this feedback is presented in the form of a transformed display such as a Lissajous plot. This display removes relative direction as a source of information about relative phase, and as a result allows people to move at any required coordination with a minimum of practice – all coordinations become equally easy. Wilson et al (2010) developed a second form of feedback, coordination feedback, which also drives learning but without altering the information available for relative phase. This study directly compared the two feedback methods. 12 subjects (aged 18-27, M=21) learned to move bimanually at 90° using either Lissajous (N=6) or coordination (N=6) feedback. We tested coordination stability with both displays in baseline, post training and retention sessions, with baseline and post training separated by 5 training sessions. Results indicated that both feedback methods are valid methods that facilitate learning. However there was no transfer of this learning between the feedback methods, confirming that the feedback methods provide different perceptual information and that perceptual learning underpins the improvements in movement stability. The two feedback methods create fundamentally different task dynamics that are informationally distinct from one another, and must be treated as such.