Designing adaptive feedback systems for managing cognitive load in augmented reality

Managing cognitive load is central to designing interactive systems, particularly within augmented reality (AR) environments that impose complex and immersive demands. This study investigates two complementary approaches in parts to managing cognitive load in AR: refining interaction modalities and integrating adaptive physiological feedback. In Part 1, eye-tracking and hand-based modalities are evaluated across tasks of varying difficulty, using skin conductance responses (SCRs) as a proxy for cognitive load. Results show that while hand gestures improved task performance in simple tasks, cognitive load levels are comparable across modalities. In Part 2, an adaptive feedback system based on a signal-derived metric, cumulative SCR (CSCR), is developed to trigger short rest interventions during sustained cognitive load. Statistical analyses illustrate that rest interventions significantly reduced cumulative cognitive load, though their effect on task performance was inconclusive. These findings emphasize the trade-offs between cognitive relief and performance continuity and highlight the potential of physiologically adaptive systems in supporting cognitive-aware interaction design.