Aesthetic preference recognition of 3D shapes using EEG.

Area of Science:

• Neuroscience
• Industrial Design
• Computer Science

Background:

• Aesthetic preferences significantly influence consumer purchasing decisions in industrial design.
• Neuromarketing employs physiological measurements to understand consumer responses to stimuli.
• Previous research explored the connection between humans, art, and aesthetics.

Purpose of the Study:

• To present a novel method for measuring user aesthetic preference using electroencephalogram (EEG) signals.
• To analyze user responses to virtual 3D shapes with motion, specifically bracelet designs generated by the Gielis superformula.
• To classify user preferences (like/dislike) based on EEG signal patterns.

Main Methods:

• Collected EEG signals using a medical-grade B-Alert X10 device at a 256 Hz sampling rate.
• Decomposed EEG signals into alpha, beta, theta, gamma, and delta rhythms via time-frequency analysis.
• Classified preferences using Support Vector Machines and K-Nearest Neighbors (KNN) classifiers with extracted frontal channel features (Fz, F3, F4).

Main Results:

• Achieved up to 80% classification accuracy in distinguishing between 'like' and 'dislike' responses.
• Identified alpha, theta, and delta rhythms from frontal EEG channels as key features for preference classification.
• Demonstrated the efficacy of KNN classifier for this specific neuromarketing application.

Conclusions:

• The proposed EEG-based method offers a reliable approach to quantifying aesthetic preferences for 3D virtual objects.
• This research contributes to the field of neuromarketing by providing objective measures of user aesthetic judgment.
• The findings have potential applications in optimizing product design and marketing strategies based on user-centric aesthetic evaluations.