Related Experiment Video
Updated: May 21, 2026
05:43
Usability Evaluation of Augmented Reality: A Neuro-Information-Systems Study
Published on: November 30, 2022
An online brain-computer interface for detecting incongruity in augmented reality applications.
Michael Wimmer1,2, Neven Elsayed1, Bruce H Thomas3
1Know Center Research GmbH, Graz, Austria.
Journal of Neural Engineering
|May 19, 2026
Summary
We developed a hybrid brain-computer interface to detect when augmented reality information mismatches user expectations. This system accurately identifies incongruities, improving user experience and trust in AR systems.
Area of Science:
- Human-Computer Interaction
- Neuroscience
- Augmented Reality
Background:
- Augmented reality (AR) systems can present digital information overlaid on the real world.
- Discrepancies between AR information and user expectations, caused by data errors or biases, can degrade user experience and trust.
- Detecting these incongruities is crucial for reliable AR applications.
Purpose of the Study:
- To propose and evaluate a hybrid brain-computer interface (BCI) for detecting inconsistencies between physical objects and digital information in AR.
- To enhance user experience and system trustworthiness in AR environments.
Main Methods:
- Two experiments were conducted: an offline study integrating eye-tracking and electroencephalography (EEG) for incongruity detection, followed by an online study assessing real-time feedback.
- EEG signals were analyzed to identify neural correlates of incongruity.
Main Results:
- Consistent electroencephalographic responses, specifically a centroparietal N400 effect, were observed for incongruent AR augmentations across both studies.
- The system achieved an average balanced accuracy of 70% in distinguishing congruent from incongruent information during the online study.
Conclusions:
- The study demonstrates the feasibility of online incongruity detection using hybrid BCIs in AR.
- This capability enables autonomous system adaptation, such as adjusting information presentation or providing real-time contextual support to users.
