This study uses AI-generated biodigital architecture images and EEG signals to classify three emotional states: awe, disgust, and contentment. In a pre-experiment with 336 participants, 60 images were selected from an initial pool of 600. EEG recordings were then collected from 52 volunteers. Gamma and delta bands produced the strongest classification performance, with gamma reaching 77.07% ± 13.8% accuracy for awe. The study also associates greenery and non-uniform granularity with positive responses, while dampness was linked to negative reactions. The paper proposes EEG as an objective tool for evaluating architectural preferences.
No heat snapshots are available in the last 24 hours.