Visual Cues Promote Concurrent Motor Learning from Exercise Instructional Videos: Evidence from Eye-Tracking and Motion Capture

This study aims to investigate the effects of visual cues in exercise instructional videos on learners' concurrent motor learning performance and attention allocation, and to examine the moderating role of prior motor experience. A 2 (visual cues: present vs. absent) × 2 (motor experience: high vs. low) between-subjects factorial design was employed. Fifty-two university students watched an exercise instructional video while simultaneously performing imitation tasks, wearing Dikablis Glasses 3 head-mounted eye-trackers and Xsens DOT inertial motion capture sensors. Five eye-tracking metrics—mean fixation duration, fixation count, mean saccade duration, saccade count, and mean saccade amplitude—were collected to assess attention allocation, while kinematic parameters including acceleration, angular velocity, and quaternions were extracted to quantify concurrent motor learning performance. Data were analyzed using partial least squares regression and two-way ANOVA. Results showed that visual cues significantly optimized learners' attention allocation strategies, as reflected by increased fixation counts and enhanced saccadic search efficiency. Two-way ANOVA further revealed a significant facilitative effect of visual cues on concurrent motor learning performance (p = .043, η² = .08). The main effect of motor experience and its interaction with visual cues did not reach statistical significance; however, high- and low-experience groups exhibited distinct eye-movement patterns. These findings indicate that visual cues significantly enhance concurrent motor learning efficiency by guiding the effective allocation of attentional resources, with a particularly pronounced compensatory effect for individuals with low motor experience. This study provides empirical evidence, grounded in multimodal physiological measurement data, for optimizing the design of online exercise instructional videos, and innovatively establishes an "eye-tracking + motion capture" synergistic assessment paradigm.