EEGdash›NeMAR›NM000249

Iss. 249 · 13 subjects · 13 recordings · CC-BY-4.0

Dataset Brief · BNCI 2022-001 EEG Correlates of Difficulty Level dataset

NM000249: eeg dataset, 13 subjects#

BNCI 2022-001 EEG Correlates of Difficulty Level dataset

Citation: Ping-Keng Jao, Ricardo Chavarriaga, Jose del R. Millan (2021). BNCI 2022-001 EEG Correlates of Difficulty Level dataset.

13-participant EEG dataset — BNCI 2022-001 EEG Correlates of Difficulty Level dataset.

Data & curation Ping-Keng Jao · Ricardo Chavarriaga · Jose del R. Millan
Year 2021 · Distributed via NeMAR
Funding Swiss National Centres of Competence in Research (NCCR) Robotics

EEG · 64 ch256 HzBIDS 1.9.0Task · imageryHealthyVisualAttention

Layer 01Study

What was asked

Hypothesis, independent & dependent variables, paradigm, cohort, and the editorial caveats around what the recordings can and cannot answer.

Layer 02Signal · BIDS

What was recorded

Sidecars, channels & electrodes, coordinate system, event semantics, and quality stats from the NEMAR pipeline when available.

Layer 03Training · ML

What you can train on

Recommended access modes — MNE Raw, braindecode windows, PyTorch DataLoader — plus the targets the metadata makes addressable.

§ 01Access · Get started

Quickstart#

Get Started

Install

pip install eegdash

Access the data

from eegdash.dataset import NM000249

dataset = NM000249(cache_dir="./data")
# Get the raw object of the first recording
raw = dataset.datasets[0].raw
print(raw.info)

Query & Filter

Filter by subject

dataset = NM000249(cache_dir="./data", subject="01")

Advanced query

dataset = NM000249(
    cache_dir="./data",
    query={"subject": {"$in": ["01", "02"]}},
)

Iterate recordings

for rec in dataset:
    print(rec.subject, rec.raw.info['sfreq'])

Cite This Dataset

If you use this dataset in your research, please cite the original authors.

BibTeX

@dataset{nm000249,
  title = {BNCI 2022-001 EEG Correlates of Difficulty Level dataset},
  author = {Ping-Keng Jao and Ricardo Chavarriaga and Jose del R. Millan},
}

§ 02Study · The README

About This Dataset#

BNCI 2022-001 EEG Correlates of Difficulty Level dataset.

Schema: HED 8.4.0 | Browse: https://www.hedtags.org/hed-schema-browser

BNCI 2022-001 EEG Correlates of Difficulty Level dataset

trajectory_start

├─ Experiment-structure
└─ Label/trajectory_start

View full README

BNCI 2022-001 EEG Correlates of Difficulty Level dataset

trajectory_start

     ├─ Experiment-structure
     └─ Label/trajectory_start

waypoint_miss

     ├─ Experiment-structure
     └─ Label/waypoint_miss

waypoint_hit

     ├─ Experiment-structure
     └─ Label/waypoint_hit

trajectory_end

├─ Experiment-structure
└─ Label/trajectory_end

Paradigm-Specific Parameters

Detected paradigm: motor_imagery
Imagery tasks: right_hand, left_hand, feet

Data Structure

Trials: {‘offline_session’: ‘32 trajectories of 32 waypoints each (~90 seconds per trajectory)’, ‘online_session_per_condition’: ‘12 trajectories of 33 waypoints each with 8 decision points’}
Blocks per session: 2
Trials context: Offline session: v-shape difficulty design (level 16→1→16). Online sessions: each condition had 12 trajectories, starting at level 1 for 1st trajectory, then 4 levels lower than final level of previous trajectory. Average 10.3 seconds per decision group (4 waypoints).

Preprocessing

Data state: preprocessed
Preprocessing applied: True
Steps: downsampling from 2048 Hz to 256 Hz, casual bandpass filtering between 1 and 40 Hz, SPHARA 20th order spatial low-pass filter for interpolation and artifact reduction, common-average re-referencing, ICA for EOG artifact removal, peripheral electrodes removed (25 central channels kept), artifact rejection: windows with peak value > 50 µV rejected
Highpass filter: 1.0 Hz
Lowpass filter: 40.0 Hz
Bandpass filter: [1.0, 40.0]
Filter type: Butterworth
Filter order: 14
Artifact methods: ICA, SPHARA, amplitude thresholding
Re-reference: car
Downsampled to: 256.0 Hz
Notes: Out of 39 recordings, P2 was removed twice from offline or online sessions due to short-circuit with the CMS or DRL electrode. On average, 15.8 ICA components were returned and 1.07 components were removed during construction of online decoders (correlation > 0.7 with EOG).

Signal Processing

Classifiers: LDA, Generalized Linear Model with elastic net regularization
Feature extraction: PSD, ICA, log-PSD
Frequency bands: analyzed=[2.0, 28.0] Hz; theta=[4.0, 8.0] Hz; alpha=[10.5, 13.0] Hz
Spatial filters: SPHARA, common-average reference

Cross-Validation

Method: leave-one-pair-out cross-validation (4x or 64x depending on class balance)
Folds: 4
Evaluation type: within_subject, cross_session

Performance (Original Study)

Accuracy: 76.7%
Offline Validation Accuracy Mean: 76.7
Offline Validation Accuracy Std: 5.1
Online Session 2 Accuracy Mean: 56.2
Online Session 2 Accuracy Std: 8.6
Online Session 3 Accuracy Mean: 54.7
Online Session 3 Accuracy Std: 11.0
Online Above Chance Recordings: 16 out of 26 (~62%)

BCI Application

Applications: drone control, adaptive learning, difficulty regulation, visuomotor learning
Environment: indoor laboratory
Online feedback: True

Tags

Pathology: Healthy
Modality: EEG
Type: Experimental/Research

Documentation

DOI: 10.1109/TAFFC.2021.3059688
Associated paper DOI: 10.1109/THMS.2020.3038339
License: CC-BY-4.0
Investigators: Ping-Keng Jao, Ricardo Chavarriaga, Jose del R. Millan
Senior author: Jose del R. Millan
Contact: ping-keng.jao@alumni.epfl.ch; ricardo.chavarriaga@zhaw.ch; jose.millan@austin.utexas.edu
Institution: Ecole Polytechnique Federale de Lausanne
Address: 1015 Geneva, Switzerland
Country: Switzerland
Repository: BNCI Horizon
Publication year: 2021
Funding: Swiss National Centres of Competence in Research (NCCR) Robotics
Acknowledgements: The authors would like to thank Alexander Cherpillod for his help in the implementation of the simulator and Ruslan Aydarkhanov for his suggestions in designing the protocol. Some figures were drawn with the Gramm MATLAB toolbox.
Keywords: EEG, real-time decoding of difficulty, closed-loop adaptation, (simulated) flying, workload, challenge point, brain-machine interface

Abstract

Adaptively increasing the difficulty level in learning was shown to be beneficial than increasing the level after some fixed time intervals. To efficiently adapt the level, we aimed at decoding the subjective difficulty level based on Electroencephalography (EEG) signals. We designed a visuomotor learning task that one needed to pilot a simulated drone through a series of waypoints of different sizes, to investigate the effectiveness of the EEG decoder. The EEG decoder was compared with another condition that the subjects decided when to increase the difficulty level. We examined the decoding performance together with behavioral outcomes. The online accuracies were higher than the chance level for 16 out of 26 cases, and the behavioral results, such as task scores, skill curves, and learning patterns, of EEG condition were similar to the condition based on manual regulation of difficulty.

Methodology

The study compared two conditions for difficulty regulation during a simulated drone piloting task: (1) EEG-based automatic difficulty adjustment using real-time decoding of perceived difficulty, and (2) Manual self-paced adjustment where subjects pressed a button when they found the level easy. Each subject participated in one offline session (for building subject-specific decoders) and two online sessions (each containing both EEG and Manual conditions in counterbalanced order). The task involved piloting a drone through circular waypoints with 16 difficulty levels defined by waypoint radius. Features were extracted using Thomson’s multitaper algorithm with 2-second sliding windows, and classification used generalized linear models with elastic net regularization followed by LDA. The study evaluated both decoding accuracy and behavioral outcomes (task scores, skill curves, learning patterns).

References

Jao, P.-K., Chavarriaga, R., & Millan, J. d. R. (2021). EEG Correlates of Difficulty Levels in Dynamical Transitions of Simulated Flying and Mapping Tasks. IEEE Transactions on Human-Machine Systems, 51(2), 99-108. https://doi.org/10.1109/THMS.2020.3038339 Notes .. versionadded:: 1.3.0 This dataset is designed for cognitive workload assessment and difficulty level detection. Unlike motor imagery datasets, the task involves actual motor control while the cognitive state (perceived difficulty) varies.

The public release contains only the first session (offline) data. Additional behavioral data and online sessions with closed-loop difficulty adaptation are not included. The paradigm “imagery” is used for compatibility, though the actual task involves motor execution with cognitive load variations. See Also BNCI2015_004 : Multi-class mental task dataset with imagery and cognitive tasks BNCI2014_001 : 4-class motor imagery dataset Examples

>> from moabb.datasets import BNCI2022_001 >>> dataset = BNCI2022_001() >>> dataset.subject_list [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13]

Appelhoff, S., Sanderson, M., Brooks, T., Vliet, M., Quentin, R., Holdgraf, C., Chaumon, M., Mikulan, E., Tavabi, K., Hochenberger, R., Welke, D., Brunner, C., Rockhill, A., Larson, E., Gramfort, A. and Jas, M. (2019). MNE-BIDS: Organizing electrophysiological data into the BIDS format and facilitating their analysis. Journal of Open Source Software 4: (1896). https://doi.org/10.21105/joss.01896 Pernet, C. R., Appelhoff, S., Gorgolewski, K. J., Flandin, G., Phillips, C., Delorme, A., Oostenveld, R. (2019). EEG-BIDS, an extension to the brain imaging data structure for electroencephalography. Scientific Data, 6, 103. https://doi.org/10.1038/s41597-019-0104-8 Generated by MOABB 1.5.0 (Mother of All BCI Benchmarks) NeuroTechX/moabb

§ 03Cohort · Participants

Cohort#

Dataset Statistics#

Age distribution by gender (n=13, range 23–23 yr, mean 22.0 yr)

20

Other · 13

Channel counts: 64 ch (n=13 recordings)

Sampling frequencies: 256.0 Hz (n=13 recordings)

Total recording duration: 16 h 11 min

§ 04Signal · Electrodes & trace

Signal · Electrodes & live trace#

Fig. 01 Signal & montage 64 ch · EEG · 256 Hz · 13 subjects, 13 recordings

Live trace viewer — sub-13 · ses-0task · task-imagery · run-0

Showing one representative recording out of 13 subjects and 13 recordings in this dataset. Browse the full set on OpenNeuro; drop any other _eeg.{set,edf,bdf,vhdr} file onto the viewer (or pass ?eeg=<url>) to inspect it.

Electrode layout — EEG · 64 sensors — 64 channels

NEMAR Processing Statistics#

The plots below are generated by NEMAR’s automated EEG pipeline. The histogram shows pipeline success for data cleaning and ICA decomposition, the percentage of data frames and EEG channels retained after artefact removal, line noise per channel (RMS, dB), and the age/gender distribution of participants.

HED event descriptors word cloud

§ 05Manifest · BIDS tree

Manifest#

File Explorer#

Browse the BIDS file structure of this dataset. Records are fetched on demand from the EEGDash catalog the first time you open the explorer.

Recordings—

Files—

Subjects—

Modalities—

Click to load file structure…

§ 06API · Programmatic access

API Reference#

Signature

eegdash.dataset

class

eegdash.dataset.NM000249(cache_dir, query=None, s3_bucket=None, **kwargs)

Bases: EEGDashDataset

Author (year)Jao2022

Canonical—

Importable asNM000249 · Jao2022

Sourceeegdash/dataset/registry.py · [source ↗]

class eegdash.dataset.NM000249(cache_dir: str, query: dict | None = None, s3_bucket: str | None = None, **kwargs)[source]#

BNCI 2022-001 EEG Correlates of Difficulty Level dataset

Study:: nm000249 (NeMAR)
Author (year):: Jao2022
Canonical:: —

Also importable as: NM000249, Jao2022.

Modality: eeg; Experiment type: Attention; Subject type: Healthy. Subjects: 13; recordings: 13; tasks: 1.

Parameters:

cache_dir (str | Path) – Directory where data are cached locally.
query (dict | None) – Additional MongoDB-style filters to AND with the dataset selection. Must not contain the key dataset.
s3_bucket (str | None) – Base S3 bucket used to locate the data.
**kwargs (dict) – Additional keyword arguments forwarded to EEGDashDataset.

data_dir#

Local dataset cache directory (cache_dir / dataset_id).

Type:: Path

query#

Merged query with the dataset filter applied.

Type:: dict

records#

Metadata records used to build the dataset, if pre-fetched.

Type:: list[dict] | None

Notes

Each item is a recording; recording-level metadata are available via dataset.description. query supports MongoDB-style filters on fields in ALLOWED_QUERY_FIELDS and is combined with the dataset filter. Dataset-specific caveats are not provided in the summary metadata.

References

OpenNeuro dataset: https://openneuro.org/datasets/nm000249 NeMAR dataset: https://nemar.org/dataexplorer/detail?dataset_id=nm000249

Examples

>>> from eegdash.dataset import NM000249
>>> dataset = NM000249(cache_dir="./data")
>>> recording = dataset[0]
>>> raw = recording.load()

__init__(cache_dir: str, query: dict | None = None, s3_bucket: str | None = None, **kwargs)[source]#

save(path: str, overwrite: bool = False, offset: int = 0)[source]#

Save datasets to files by creating one subdirectory for each dataset:

path/
    0/
        0-raw.fif | 0-epo.fif
        description.json
        raw_preproc_kwargs.json (if raws were preprocessed)
        window_kwargs.json (if this is a windowed dataset)
        window_preproc_kwargs.json  (if windows were preprocessed)
        target_name.json (if target_name is not None and dataset is raw)
    1/
        1-raw.fif | 1-epo.fif
        description.json
        raw_preproc_kwargs.json (if raws were preprocessed)
        window_kwargs.json (if this is a windowed dataset)
        window_preproc_kwargs.json  (if windows were preprocessed)
        target_name.json (if target_name is not None and dataset is raw)

Parameters:

path (str) –

Directory in which subdirectories are created to store
-raw.fif | -epo.fif and .json files to.
overwrite (bool) – Whether to delete old subdirectories that will be saved to in this call.
offset (int) – If provided, the integer is added to the id of the dataset in the concat. This is useful in the setting of very large datasets, where one dataset has to be processed and saved at a time to account for its original position.

Access modesMNE → braindecode → PyTorch → ML

.rawMNE Raw object — standard tools (filter, epoch, ICA, plot_psd).mne

BaseConcatDatasetEach record is a lazy BaseDataset from braindecode — windowed via create_windows_from_events.braindecode

DataLoaderWraps the windowed dataset into a PyTorch DataLoader; supports parallel workers and on-the-fly augmentations.pytorch

Zarr cacheOptional braindecode Zarr mirror for fast resume; persisted to cache_dir.zarr

Hugging FaceNo per-dataset mirror published yet — browse the EEGDash org listing for sibling datasets. See the datasets loader API.huggingface

Croissant 1.0Machine-readable JSON-LD descriptor — NM000249.croissant.json (MLCommons schema, ingestible by PyTorch / TensorFlow / JAX).mlcommons

Examples using EEGDashcurated · start here

Find datasets with the EEGDash APIQuery the catalogue, filter by task or modality, list candidates.

Load one EEG recordingResolve a single record to an MNE Raw with channels and events.

EEG recording to PyTorch DataLoaderWrap braindecode windows in a DataLoader for model training.

Preprocess EEG and create windowsFilter, resample, epoch — and persist the windowed dataset.

Save and reload prepared dataCache a windowed dataset to disk and reattach it without recompute.

Download a dataset locallyPrefetch BIDS files to a local cache and validate the layout.

Swap any load_dataset(...) call for nm000249 to reproduce the tutorial on this dataset.

Citation

Ping-Keng Jao, Ricardo Chavarriaga, Jose del R. Millan (2021). BNCI 2022-001 EEG Correlates of Difficulty Level dataset.

Provenance

¹Contributed to nemar in BIDS format.

²Curated & ingested by the EEGDash catalog; see CITATION.cff for canonical reference.

Related & sibling datasets

NM000198EEG · 13 subj NM000194EEG · 12 subj NM000219EEG · 18 subj NM000188EEG · 10 subj NM000259EEG · 10 subj

+ 1 more — see See Also below →

BIDS

BIDS 1.9.0

Sidecars

events · events.json · channels · eeg.json

Provenance

CC-BY-4.0 · DOI not on file

Machine-readable

schema.org/Dataset · Croissant

Mirrors

OpenNeuro · NEMAR · HF org

Dataset ID	`NM000249`
Title	BNCI 2022-001 EEG Correlates of Difficulty Level dataset
Author (year)	`Jao2022`
Canonical	—
Importable as	`NM000249`, `Jao2022`
Year	2021
Authors	Ping-Keng Jao, Ricardo Chavarriaga, Jose del R. Millan
License	CC-BY-4.0
Citation / DOI	Unknown
Source links	OpenNeuro \| NeMAR \| Source URL

NM000249: eeg dataset, 13 subjects#

Quickstart#

About This Dataset#

Cohort#

Dataset Statistics#

Signal · Electrodes & live trace#

NEMAR Processing Statistics#

Manifest#

File Explorer#

API Reference#

Citation

Provenance

Related & sibling datasets

See Also#