Motor adaptation during a sound oriented task (NCM 2017, poster publication)

This poster will be presented in Dublin during the 2017 annual meeting of the Society for the Neural Control of Movement (NCM 2017).

Motor adaptation during a sound oriented task

Eric O. Boyer1,2, Frederic Bevilacqua2, Sylvain Hanneton3, Agnes Roby-Brami1,
1 ISIR – CNRS UMR 7222, UPMC, 2: IRCAM – STMS-CNRS, UPMC, 3: LPP – CNRS UMR 8242, University Paris Descartes, Paris, France

Introduction. Movement sonification systems appear promising for sensori-motor learning in providing users with auditory feedback of their own movements [1]. However, research on sonification for sensori-motor learning has been mainly directed toward “movement oriented tasks” where the instruction and the attention is put on the movement itself. In contrast, the aim of the present study was to test a situation were the instruction and attention is given to the sound, that we call a “sound-oriented task”.
The sonification mapping relied on the metaphor of friction sounds produced by drawing movements. We focused on the drawing of ellipses which is characterized by a well known invariant velocity-curvature relationship [2,3]. The replay of friction sounds (registered or synthesized) can evoke the shape of the drawings [4] and induce a sensori-motor perceptive bias on the reproduction of visual motion [5] .
Elaborating further on these bases, we tested the effect of on-line sonification with friction-like sounds on the kinematics and the shape of elliptical drawing movement. The mapping was implemented as a band pass filter whose center frequency varied linearly with the velocity of the movement [3,4]. We analyzed the motor adaptation of the drawing movements when the subject had the instruction to maintain a constant sonification pattern while the frequency of the filter changed without his/her knowledge.
Our hypothesis was that the alteration of the sonification mapping would induce temporal and/or spatial adaptation of the movement.

(see the poster for methods and results)

Discussion. The motor adaptation tended to compensate for the changes in sound feedback induced by the changes in the sound-movement mapping. This demonstrate that the participant could adapt their movement to the “sound oriented” task. The adaptation was manifested by modifications of the kinematics. There was a tendency for increase of frequency and decreased size of the drawing in the control situation. In addition, the movement was faster with larger movements when the gain of the mapping was increased and slower with smaller movements when it decreased. The global shape and orientation of the ellipse was not modified in 2D.
This demonstrates that the participant privileged the stability of the geometrical shape and adapted their velocity in order to satisfy the instruction to keep the sonification pattern constant. The increase in velocity was more due to a change in frequency while the decrease was more due to a shrinking of the shape, suggesting different movement regimen [e.g. alternative versus discrete, 7]. The modification of the angle of the ellipse during the experiment in 3D but not in 2D was probably due to greater inertial constraints as shown by Pfann et al [8]. Participants who were instructed to draw circles with shoulder-elbow movements made ellipses with increasing eccentricity when the velocity increased (order of magnitude: 1m/s). In addition, the elongation of the ellipse was in the direction of least inertia. A similar effect was also observed for handwriting-like movements (similar to our 2D task) by Dounskaia et al. [9] but for a much higher velocity regimen (instruction level “as fast as possible”, 0.34m/s), when the velocity we used corresponds to their “self paced” level).

Conclusion. This study demonstrates that movement sonification can be used i) to induce implicit motor adaptation in both planar and 3D movements and ii) to control the direction and magnitude of this adaptation through mapping parameters modification.