Contextual Influence on Learning Styles: A VARK and Tensor-Based Approach for Adaptive Pervasive Learning

This paper explores the contextual influences on learners' learning styles using the VARK (Visual, Auditory, Reading/Writing, Kinesthetic) model and a tensor-based approach to generate contextual learning styles. Each learner's baseline learning style is represented as a VARK vector, with weights assigned to each feature. We expand the analysis by introducing engagement scores as vectors corresponding to VARK features, reflecting learners' engagement levels in various contexts. We randomly generate contextual dimensions for each learner-such as location, activity, device, and noise-each with associated VARK vectors. A tensor-based approach calculates the mean of these contextual VARK vectors, representing the learners' contextual learning styles. The study further calculates the Eu-clidean distance between contextual learning styles and baseline learning styles to assess the impact of contextual changes on engagement. By analyzing the distances and correlating them with the learners' engagement vectors, we aim to provide insights into how deviations from baseline learning styles can affect learning engagement and performance outcomes. This research contributes to the development of adaptive pervasive learning systems that tailor educational experiences to the diverse needs of learners based on their contextual influences.