Accepted for/Published in: JMIR Serious Games
Date Submitted: Jul 28, 2022
Date Accepted: Nov 2, 2022
An Automated Virtual Reality Training System for Teacher-Student Interaction: A Randomized Controlled Trial
ABSTRACT
Background:
Shortages in qualified supervision and other resources prevent education personnel from rehearsing effective practices. Interactive simulations, though increasingly employed in education, frequently require instructor management. Automated simulations rarely engage trainees in skills related to practice (e.g., speech).
Objective:
We evaluated the capability of delivering behavioral skills training through a virtual reality (VR) simulation utilizing artificial intelligence (AI) to improve the use of a nondirective mathematical questioning strategy.
Methods:
We recruited and randomly assigned 30 college-aged participants into treatment (i.e., lecture, modeling, VR) and control groups (i.e., lecture and modeling only). Lessons concerned the use of a nondirective mathematical questioning strategy in instances where a simulated student provided correct or incorrect answers to word problems. Measures included observed and automated assessments of participant performance and subjective assessments of participant confidence. Participants completed pre-test, post-test, and maintenance probes each week over the course of three weeks.
Results:
A mixed ANOVA revealed significant main effects of time (F[2,27]=124.154, P < .001, ηp2 = 0.816) and treatment (F[1,28]=19.281, P < .001, ηp2 = 0.408), as well as an interaction effect (F[2,28]=8.429, P < 0.001, ηp2 = 0.231) for the average percentage of steps in the questioning procedure. Post-test scores for the intervention group (M = 88%, SD = 22.62) exceeded control group performance (M = 63.33%, SD = 22.64), with t[28] = 3.653, P < .001, d = 1.334. Maintenance scores indicated a positive effect of intervention (M = 83.33%, SD = 24.40) relative to control (M = 54.67%, SD = 15.98), t[28] = 3.807, P < .001, d = 1.39. A Mann-Whitney U test indicated the treatment groups’ self-ratings of confidence (M = 2.41, SD =.51) were higher than those of the control group (M = 2.04, SD =.52), U = 64, P = .043, r = .137).
Conclusions:
Results demonstrate the potential of AI-augmented VR to deliver effective, evidence-based training with limited instructor management. Additional work is needed to demonstrate the cascading effect of training on authentic practice and encompass a wider range of skills.
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.