The Impact of Visual and Kinesthetic Motor Imagery on Mental Fatigue and Classification Performance in Untrained Participants

Abstract

Motor imagery (MI) is a critical component of Brain-Computer Interface (BCI) technology, but the effective use of MI-BCI requires significant user training. In order to improve the training effect, it is necessary to choose a stable and easy to use paradigm. To explore the effects of different paradigms on training effectiveness, 18 untrained participants were recruited to perform kinesthetic motor imagery (KMI) and visual motor imagery (VMI) experiments. The participants’ subjective vividness and mental fatigue levels were measured using a subjective scale and (θ + α)/β value in the experiments, respectively. Additionally, the Common Spatial Pattern (CSP) algorithm and the Support Vector Machine (SVM) algorithm were used to extract EEG signal features and classify them. The results showed that the majority of participants demonstrated a greater aptitude for excelling in the VMI paradigm, where the average accuracy rate was 84.7%, compared to the KMI paradigm, where the average accuracy rate was 79.6%. In addition to this, participants with higher levels of sports proficiency showed better adaptability in the KMI paradigm. Compared to KMI, VMI can enable most inexperienced participants to achieve a better user experience with less fatigue and improve training performance.