ERI: Rewiring Student Attention: Adaptive and Intelligent VR STEM Labs

Laboratory education plays a vital role in preparing the future science, technology, engineering, and mathematics workforce by providing opportunities to develop practical skills that cannot be acquired through classroom instruction alone. Virtual reality technology offers unique opportunities for laboratory learning by enabling students to safely practice hands-on procedures that may be hazardous, expensive, or difficult to conduct in traditional laboratories. However, maintaining attention during immersive laboratory activities remains a significant challenge, particularly when learners must sustain concentration while performing precise manual tasks. Attentional lapses can reduce learning effectiveness, increase errors, and diminish the overall learning experience. This project seeks to advance the scientific understanding of how adaptive instructional support can help learners maintain attention, improve motor skill development, and enhance learning in immersive virtual laboratories. The resulting knowledge will inform the design of intelligent learning technologies that strengthen science, technology, engineering, and mathematics education, broaden access to high-quality laboratory experiences, and contribute to the preparation of a highly skilled workforce. This Engineering Research Initiation (ERI) project will investigate how adaptive empathic interventions influence attention regulation, motor performance, and learning during immersive virtual laboratory training. The project will develop a closed-loop framework that continuously estimates learners' attentional states using multi modal behavioral and physiological measures, including eye-gaze behavior, pupil dynamics, galvanic skin response, electromyography, and task performance. Based on individualized attention models, an empathic virtual instructor, presented either as a virtual human or a voice-only agent, will dynamically adapt its verbal and nonverbal behaviors to sustain attention and facilitate recovery from attentional lapses. Controlled studies using virtual chemistry and computer science laboratories will evaluate how adaptive empathic interventions compare with conventional perceptual cues in improving attention maintenance, motor skill acquisition, learning outcomes, and user experience. The project is expected to contribute to a longstanding knowledge gap of whether learning in virtual environemts supported by intelligent agents can produce skills, particularly motor skills, that can effectively transfer to real-world settings. Further, the projects can potentially produce validated adaptive intervention strategies, personalized multi modal attention models, and generalizable design principles for intelligent immersive learning systems. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.