Wang2021Combined — Raw EEG data for combined SSVEP BCI with single stimulus location (repackaged)

  Repackaged raw EEG data from Wang et al. (2021), originally hosted on Figshare (https://figshare.com/ndownloader/files/25000886). The original nested RAR archives have been converted to
  per-subject ZIP files for easier programmatic access.

  This dataset accompanies the paper: "Single stimulus location for two inputs: A combined brain-computer interface based on Steady-State Visual Evoked Potential (SSVEP)" published in the
  European Journal of Neuroscience (DOI: 10.1111/ejn.15030).

  Paradigm

  A "one-to-two" SSVEP paradigm where spatially overlapping stimuli are presented at the same screen location. The core idea is to simplify the BCI keyboard layout by encoding two
  independent inputs (e.g. character selection and confirmation) into a single fixation point, reducing attention shifting and making the interface more accessible for users with eye
  dyskinesia.

  Subjects attend to one of four arrow directions (up, down, left, right), each flickering at a distinct frequency derived from the 85 Hz CRT refresh rate:
  - Up: 14.17 Hz (85/6)
  - Down: 12.14 Hz (85/7)
  - Right: 9.44 Hz (85/9)
  - Left: 7.73 Hz (85/11)

  Each trial: 500 ms fixation + 500 ms cue + 5000 ms stimulation.

  Channels

  33 channels per .cnt file (30 EEG + 3 auxiliary):

  EEG (30): FP1, FP2, F7, F3, FZ, F4, F8, FC5, FC1, FPZ, FC2, FC6, T7, C3, CZ, C4, T8, CP5, CP1, POZ, CP2, CP6, P7, P3, PZ, P4, P8, O1, OZ, O2

  Auxiliary (3): HEOG, VEOG, M2 (right mastoid, used as online reference)

  Recording setup

  - System: eego mylab (ANT Neuro)
  - Channels: 33 (30 EEG + HEOG + VEOG + M2)
  - Reference: M2 (right mastoid)
  - Sampling rate: 1000 Hz
  - File format: .cnt (ANT Neuro native)
  - Trigger codes: 11 (up), 21 (down), 31 (right), 41 (left)
  - Additional triggers: 12/22/32/42 (cue markers, 10 per class per run); 1/3/4 (block/session markers)
  - Run duration: ~960–1170 seconds each

  Trials

  200 trials per run, balanced across 4 directions (50 trials per class per run). This is consistent across all 22 .cnt files in the dataset.

  Subjects

  8 subjects across two experiments (the original Figshare archive was labeled "not all" and contained 8 of 20 subjects):
  - Experiment 1 (S01-S05): Scheme 1 — moving dot-formed arrows, 2 runs per subject
  - Experiment 2 (S06-S08): Scheme 2 (space-based attention) + Scheme 3 (object-based attention), 4 runs per subject

  Files

  8 per-subject ZIP files (S01.zip–S08.zip), each containing the .cnt files for that subject.

  ┌─────────┬────────────┬────────────┬──────────────┐
  │ Subject │ Experiment │ .cnt files │ Total trials │
  ├─────────┼────────────┼────────────┼──────────────┤
  │ S01     │ Exp 1      │ 2          │ 400          │
  ├─────────┼────────────┼────────────┼──────────────┤
  │ S02     │ Exp 1      │ 2          │ 400          │
  ├─────────┼────────────┼────────────┼──────────────┤
  │ S03     │ Exp 1      │ 2          │ 400          │
  ├─────────┼────────────┼────────────┼──────────────┤
  │ S04     │ Exp 1      │ 2          │ 400          │
  ├─────────┼────────────┼────────────┼──────────────┤
  │ S05     │ Exp 1      │ 2          │ 400          │
  ├─────────┼────────────┼────────────┼──────────────┤
  │ S06     │ Exp 2      │ 4          │ 800          │
  ├─────────┼────────────┼────────────┼──────────────┤
  │ S07     │ Exp 2      │ 4          │ 800          │
  ├─────────┼────────────┼────────────┼──────────────┤
  │ S08     │ Exp 2      │ 4          │ 800          │
  └─────────┴────────────┴────────────┴──────────────┘

  Repackaging

  This archive was repackaged from the original Figshare deposit for integration with the MOABB benchmarking framework (https://github.com/NeuroTechX/moabb). No data was modified — only the archive format was changed from nested RAR to per-subject ZIPs.