Neuroadaptive Uncertainty Displays: Closed-Loop Human–AI Decision Support Calibrated by EEG Markers of Expected vs. Unexpected Uncertainty

Prabhudesai et al. (2023) show that communicating model uncertainty slows users, reduces overreliance, and induces more analytical thinking. Chang et al. (2025) demonstrate EEG-based classification of cognitive states in expected vs. unexpected uncertainties within a decision support setting. We propose a neuroadaptive DSS that senses when users are experiencing unexpected uncertainty (elevated FRN to negative feedback and larger P3 to surprising outcomes) and dynamically shifts its uncertainty communication style—e.g., moving from numeric to graphical intervals, adding counterfactual exemplars, or adopting portfolio framing (Reeck & LaBar, 2024). Grounded in a sequential, feedback-sparse task (Cecchini et al., 2024), the system uses reinforcement learning to select communication strategies that best calibrate trust and performance for each user. Novelty lies in closing the loop: most DSSs are static, whereas we leverage online neural indicators of deviations from expectation to tailor uncertainty communication moment by moment. This could reduce automation bias, improve user calibration under environmental uncertainty, and scale insights into MAS contexts where humans arbitrate AI recommendations (Chang et al., 2025).

References:

1. Supervised Classification Model for Neural Correspondence to Environmental Uncertainty in Multiple Cue Judgment System with Decision Support. Yoo-Sang Chang, Younho Seong, Sun Yi (2025). Proceedings of the Human Factors and Ergonomics Society Annual Meeting.
2. Reining in regret: emotion regulation modulates regret in decision making. C. Reeck, K. LaBar (2024). Cognition & Emotion.
3. Cognitive mechanisms of learning in sequential decision-making under uncertainty: an experimental and theoretical approach. Gloria Cecchini, M. Depass, Emre Baspinar, M. Andujar, S. Ramawat, P. Pani, S. Ferraina, A. Destexhe, Rubén Moreno-Bote, Ignasi Cos (2024). Frontiers in Behavioral Neuroscience.
4. Understanding Uncertainty: How Lay Decision-makers Perceive and Interpret Uncertainty in Human-AI Decision Making. Snehal Prabhudesai, Leyao Yang, Sumit Asthana, Xun Huan, Q. Liao, Nikola Banovic (2023). International Conference on Intelligent User Interfaces.