About
This data collection was conducted at the Science Museum by researchers from Imperial College London.
The MyBrainTunes Experiment
Members of the public volunteered to have their brainwaves recorded with a wearable EEG device while listening to a selection of popular songs.
The dataset includes a total of 761 sessions with 721 subjects (387 female, 334 male) aged 12 and above (mean=29, std=11).
For each subject, 30s of EEG baseline was recorded, then 30s segments of 30 songs were played in a random order. Subjects were asked to sit still during recording, with their eyes open and focused on a dot in the middle of a screen.
Song segments were pre-defined before the start of the experiment. Songs were selected from the top of the UK charts, covering several decades and genres. Just before their EEG recording, participants were asked to indicate their familiarity with each song, and rate how much they liked each song. A rating of 0 was given for disliked songs, and ratings from 1 to 5 for the degree of liking.
Participants also completed a questionnaire with basic information including: age, sex, handedness, education level, musical experience, and coffee intake within the last few hours.
EEG Data
EEG was recorded using Emotiv EPOC+ wet electrode headsets
- 14 channels: AF3, F7, F3, FC5, T7, P7, O1, O2, P8, T8, FC6, F4, F8, AF4
- Reference: left/right mastoids
- Sampling frequency: 256Hz
- Resolution: 14 bits with 1 LSB = 0.51μV
No further pre-processing has been applied, beyond the steps performed by the Emotiv recording setup.
Citation
- Siegfried Ludwig, Stylianos Bakas, Dimitrios A. Adamos, Nikolaos Laskaris, Yannis Panagakis, and Stefanos Zafeiriou. (2024). EEGminer: Discovering Interpretable Features of Brain Activity with Learnable Filters.
Download
To download the full MyBrainTunes dataset please email a signed copy of the EULA to mybraintunes@imperial.ac.uk with the subject "MyBrainTunes Request Access"