Impact of Movement Amplitude in Dynamic Visual Cues on Motor Imagery Ability and BCI Classification

Motor imagery-based brain-computer interface (MI-BCI) is an important research topic in the field of motor function rehabilitation. Dynamic visual cues can modulate motor imagery (MI) performance, yet the influence of movement amplitude remains underexplored. This study investigates the modulatory effects of movement amplitude in dynamic visual cues on motor imagery ability and MI-BCI classification performance. Participants performed right-hand MI tasks cued by dynamic visual stimuli of three amplitudes (large: 60°, medium: 40°, small: 20°). Subjective kinesthetic imagery vividness, beta-band Event-related desynchronization (ERD), and classification accuracy were measured. Results showed that large-amplitude visual cues significantly enhanced subjective vividness scores (KMI: 3.14 ± 0.85), elicited stronger MI-related cortical activation (beta-band ERD: -6.92 ± 9.97), and achieved higher MI-BCI classification accuracy (e.g., beta-band accuracy: 93.45% ± 5.37%) compared with medium and small amplitudes. These findings indicate that the amplitude of dynamic visual cues modulates both subjective and objective MI performance, providing practical insights for optimizing MI-BCI training paradigms and HCI-based rehabilitation interventions. During formal testing, the participants maintained physical stillness. Adopting the Graz training paradigm, each condition comprised 5 practice trials followed by 30 experimental trials (Yang et al., 2024). Before each MI block, participants practiced the tapping motion at the specified amplitude (20°, 40°, or 60°) guided by the experimenter. The experimenter demonstrated the correct angle using a goniometer and visually monitored the participant’s movements during practice to ensure compliance. A single trial lasted 13 seconds with four phases: Fixation (2 s): A black "+" at the center of the screen prompted attentional focus. Visual cue (1 s): A dynamic visual cue was presented on the screen. MI period (5 s): Participants repetitively imagined the right-hand palm-tapping sensation. Rest (5 s): A "relax" prompt appeared for recovery. After each block, participants completed the subjective KVIQ scale. Blocks lasted approximately 10 minutes with 2-minute inter-block rest periods. The ErgoLAB 3.0 software was used for paradigm presentation, EEG calibration and adjustment, and data acquisition. EEG signals were recorded using a 32-channel Bitbrain semi-dry electrode system (sampling rate: 256 Hz).