Design Emotion Dataset

25 participants underwent structured art healing sessions while their brain activity and psychological state were recorded. The dataset contains synchronized 64-channel EEG recordings and anxiety scale scores from university students, supporting research on art-based anxiety interventions. It was created by Chenyi Chen and last updated in May 2026.

Use Cases

• Predicting anxiety relief labels based on occipital alpha and parietal theta power features.
• Analyzing synchronized EEG and behavioral video data using the provided video timestamps.
• Training machine learning models for mental state classification using preprocessed 64-channel EEG signals.
• Validating BIDS-standard neuroimaging data pipelines for multimodal research.

Strengths

• Data quality is validated with an XGBoost classifier achieving an AUC of 0.877.
• Contains synchronized multimodal data: 64-channel EEG, video, and psychological assessments.
• Follows the BIDS (v1.8.0) international standard for organization and anonymization.
• EEG data is preprocessed with a 1-40 Hz filter and a 200Hz sampling rate.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Data may reflect bias inherent to the specific university student population studied.

Provenance

Source

figshare

Collection Method

Collected from 25 participants during a structured art healing session consisting of mindfulness and painting phases.

Freshness

Last updated 2026-05-01 05:05:34; freshness should be verified.