Ongoing Tracking of Engagement in Motor Learning

TL;DR

This work tackles real-time tracking of learner engagement in motor learning using noninvasive, privacy-preserving wearables. It introduces a two-stage approach: first, a behavioral model anchors eng as a latent construct from self-report scales within Flow theory; second, a sensor-based ML pipeline (with windowed features and interpolation) predicts eng in near real time, with a supplementary discomfort predictor to enhance accuracy. The best-engagement classifier (XGBoost) achieves a robust $F_1$ score around $0.82$ when discomfort is included, and movement-based features from IMU/GYR drive the strongest performance, suggesting practical utility for teacher feedback and human-robot interaction control. The study demonstrates a pathway to continuously monitor engagement during motor tasks and to extend the framework to other activities like handwriting and drum play in educational settings.

Abstract

Teaching motor skills such as playing music, handwriting, and driving, can greatly benefit from recently developed technologies such as wearable gloves for haptic feedback or robotic sensorimotor exoskeletons for the mediation of effective human-human and robot-human physical interactions. At the heart of such teacher-learner interactions still stands the critical role of the ongoing feedback a teacher can get about the student's engagement state during the learning and practice sessions. Particularly for motor learning, such feedback is an essential functionality in a system that is developed to guide a teacher on how to control the intensity of the physical interaction, and to best adapt it to the gradually evolving performance of the learner. In this paper, our focus is on the development of a near real-time machine-learning model that can acquire its input from a set of readily available, noninvasive, privacy-preserving, body-worn sensors, for the benefit of tracking the engagement of the learner in the motor task. We used the specific case of violin playing as a target domain in which data were empirically acquired, the latent construct of engagement in motor learning was carefully developed for data labeling, and a machine-learning model was rigorously trained and validated.

Figures (8)

• Figure 1: The Flow theory: 'flow' is attained when the level of skill matches the perceived complexity of the challenge.
• Figure 2: Experimental protocol
• Figure 3: Wearable sensors
• Figure 4: Behavioral-model for the manipulation of eng
• Figure 5: Box-plot of eng ('engagement') levels for each discomfort level