Pupils and Heartbeats: Using Biometrics for Real-time Measurement of Student Cognitive Load in Virtual Reality Learning Experiences

There are separate poster presentation times for odd and even posters.
Odd poster #s – first hour
Even poster #s – second hour

Co-authors:

Tod Clapp - Colorado State University; Chad Eitel - Colorado State University; Heather Hall - Colorado State University; Kenneth Ivie - Colorado State University; Brian Kelly - Colorado State University; Samantha McGrath - Colorado State University; Carolyn Meyer - Colorado State University; Becky Wiltgen - Colorado State University

Abstract Body : Integrating virtual reality (VR) into education can transform teaching and learning allowing users to interact with three-dimensional data similar to how it exists in nature. Instructors across disciplines use VR to innovate and overcome traditional education barriers. There is a significant need for evidence-based resources for VR-based instructional design to ensure effective implementation. This project seeks to establish instructional design practices for VR that optimize cognitive load, enhance learning, and improve the learner's experience. A mixed-factorial research design was used to assess differences and associations among undergraduate students between (n = 91) and within (n = 43) groups. Our objectives include evaluating cognitive load differences based on data type, assessing the impact of content sequencing on cognitive load, and exploring the interaction between student attitude/behavior and cognitive load. Combining pupillometry, eye gaze, and heart rate biometric data, we utilized biometrics to define these real-time measurements as an indicator of cognitive load. Analyses included independent and paired t-tests, single and multi-factor ANOVA, and linear regression. Results indicate statistically significantly lower mean cognitive load for students viewing 3D content compared to 2D content in between-group, t(41) = 4.9, p < 0.001, and within-group comparisons, t(90) = 10.83, p < 0.001. Attitude toward virtual reality did not statistically significantly impact mean cognitive load (F(2, 87) = 0.521, p = 0.596). Conversely, content sequencing significantly influenced cognitive load (F(1, 84) = 4.430, p = 0.038). The data demonstrates that virtual learning experience design and data visualization can manipulate cognitive load, while student attitude has limited impact on instructional design effects.