Hamburg Fluctuating Pain Database: Multimodal Biosignals from 45 Participants

Forty-five healthy participants completed twelve 3-minute trials each, resulting in 540 trials and 27 hours of recordings during fluctuating noxious heat stimulation. The multimodal dataset includes electroencephalography, electrodermal activity, heart rate, pupil diameter, facial expression features, and continuous pain ratings. It was created by Leonard Visser and last updated in April 2026.

Use Cases

• Predicting continuous pain ratings based on multimodal physiological signals like EEG and heart rate.
• Analyzing the temporal dynamics of pain perception using time-series data from fluctuating heat stimuli.
• Studying the correlation between autonomic nervous system activity (e.g., electrodermal activity) and subjective pain experience.
• Developing multimodal fusion models for pain state classification using features from EEG, video, and questionnaires.

Strengths

• Contains 540 trials from 45 participants, totaling 27 hours of recordings.
• Includes eight data modalities: EEG, electrodermal activity, heart rate, pupil diameter, facial features, pain ratings, calibration results, and questionnaires.
• Data is provided in a structured DuckDB database with 29 tables, including raw, causally preprocessed, and merged 10 Hz multimodal tables.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Data is from a controlled experiment with healthy participants, which may limit generalizability to clinical pain populations.

Provenance

Source

figshare

Collection Method

Data collected from laboratory experiments with healthy participants undergoing tonic noxious heat stimulation.

Freshness

Last updated 2026-04-30 11:51:03.

Geography

Hamburg (inferred from title).