Machine Learning Algorithm for Predicting Pre-Service Teachers' Readiness to Use Brain-Computer Interfaces in Inclusive Classrooms

We examine the factors influencing pre-service teachers' readiness to work with special needs students (SNSs) on Brain-Computer Interfaces (BCIs). While BCIs hold promise for enhancing learning experiences for SNSs, the successful collaborations between the teachers and the SNSs depends on the preparedness of the teachers. Existing research highlights the importance of attitudes, knowledge, and self-efficacy in technology adoption, but there is limited understanding of how these factors interact to influence readiness specifically among pre-service teachers to work with SNSs on BCIs. To address this gap, we adopted a machine learning decision tree analysis to identify key factors and their interactions among 320 pre-service teachers from various institutions. The results revealed that attitude (AT) is the most predictor of teachers' readiness, followed by subjective norm (SN), perceived self-efficacy (PSE), and basic knowledge of assistive technology (BKAt). These factors interact in complex ways, indicating that a multifaceted approach to training is essential. For instance, a high attitude combined with high self-efficacy and knowledge levels results in the highest readiness scores. The results suggest that teacher training programs should focus on fostering positive attitudes towards SNSs on BCIs, enhancing self-efficacy through hands-on experience, and providing comprehensive knowledge about assistive technologies. Creating a supportive social environment that encourages peer collaboration and mentorship is also crucial. We conclude that a holistic approach addressing multiple dimensions of readiness can better prepare pre-service teachers to work with SNSs on BCIs in inclusive classrooms, ultimately benefiting students with special needs. These insights can inform the design of targeted interventions and training programs, promoting effective technology adoption in education.