Integrating Adaptive Cueing in a Physiology-sensitive Learning Platform: Exploring Design Needs

Abstract

Adaptive cueing in currently-existing technology-enhanced learning (TEL) often prioritizes performance metrics over the cognitive load experienced by a learner, thereby losing comprehensive view to realizing learning outcomes. Cues e.g., visual, auditory, tactile cues have been shown to be effective in guiding the learner's attention and managing cognitive load. However, offering cues to promote one's skill learning based only on individual performance indicators can be misleading, since added to the performance scores, cognitive aspects and learning ability of the learner are also crucial for effective learning. Though the cognitive aspects are subtle in nature, yet these can be harnessed with the use of neurophysiological tools. Tools like Eye-gaze tracking and Electroencephalography (EEG) offer avenues to infer attention, memory load, and decision-making. Accessing such neurophysiological signals need one to deal with technical barriers, e.g., multi-modal synchronization, latency constraints, and real-time signal processing, hindering their adoption in dynamic learning environments. Here, we present the architecture that can be used to realize and overcome the technological challenges faced while integrating the cueing paradigm with synchronized multi-modal neurophysiological signal acquisition. This platform (i) estimates cognitive load through synchronized eye-tracking and EEG data during task execution, (ii) correlates it with performance outcomes, and (iii) generates adaptive cues tailored to individual cognitive profiles to optimize learning efficiency. Further, by integrating with existing platforms like LAReflect, our approach provides actionable feedback for both learners and trainers. The broad aim is to enable implementation of cognition-aware skill learning platforms with adaptive, individualized cueing to foster effective learning.