A neuroscience dataset combining electroencephalography (EEG) and near-infrared spectroscopy (NIRS) signals to measure cognitive workload. The data was collected from subjects performing repetitions of three difficulty conditions of the N-back task and separated into five-second windows. It was created by Emily B. J. Coffey of the University of Amsterdam.
Use Cases
- Binary classification of workload condition based on EEG bandpower-derived features.
- Binary classification of workload condition based on NIRS average hemoglobin levels.
- Evaluating the combined utility of EEG and NIRS for real-time workload monitoring.
- Benchmarking physiological signal processing methods for human-machine interaction research.
Strengths
- Data includes eight EEG channels (Cz, Pz, FCz, Fz, C3, C4, F3, F4) and three NIRS channels over the left forehead.
- Signals were acquired simultaneously during controlled N-back task repetitions with three difficulty conditions.
- The study reports that EEG could reliably classify workload condition for most subjects.
Limitations
- Row count and dataset size are unknown, which may limit suitability assessment.
- Column-level documentation is absent; field semantics must be inferred after download.
- The description notes the NIRS signal was less helpful and did not contribute to classification accuracies when combined with EEG.
Provenance
- Source
- University of Amsterdam
- Collection Method
- Simultaneous acquisition of EEG and NIRS signals during repetitions of the N-back task.