EEGdash›NeMAR›NM000174

Iss. 174 · 15 subjects · 30 recordings · CC-BY-NC-ND-4.0

Dataset Brief · Imagined Speech EEG dataset comparing paradigm designs (Aguil…

NM000174: eeg dataset, 15 subjects#

Name: Imagined Speech EEG dataset comparing paradigm designs (Aguilera-Rodriguez et al. 2025)
License: CC-BY-NC-ND-4.0

Imagined Speech EEG dataset comparing paradigm designs (Aguilera-Rodriguez et al. 2025)

Access recordings and metadata through EEGDash.

Citation: Edgar Aguilera-Rodriguez, Alma Cuevas-Romero, Santiago Mendoza-Franco, Jonathan Wornovitzky-Green, Eduardo Rivera-Cerros, David Villanueva-Cazares, Luis Alberto Munoz-Ubando, David Ibarra-Zarate, Luz Maria Alonso-Valerdi (20). Imagined Speech EEG dataset comparing paradigm designs (Aguilera-Rodriguez et al. 2025). 10.82901/nemar.nm000174

Modality: eeg Subjects: 15 Recordings: 30 License: CC-BY-NC-ND-4.0 Source: nemar

Metadata: Complete (100%)

15-participant EEG dataset — Imagined Speech EEG dataset comparing paradigm designs (Aguilera-Rodriguez et al. 2025).

Data & curation Edgar Aguilera-Rodriguez · Alma Cuevas-Romero · Santiago Mendoza-Franco · Jonathan Wornovitzky-Green · Eduardo Rivera-Cerros · David Villanueva-Cazares · …
Senior author Luz Maria Alonso-Valerdi
Year 20 · Distributed via NeMAR

EEG · 24 ch500 HzBIDS 1.9.0Task · imagery2 sessions

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 NM000174

dataset = NM000174(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 = NM000174(cache_dir="./data", subject="01")

Advanced query

dataset = NM000174(
    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{nm000174,
  title = {Imagined Speech EEG dataset comparing paradigm designs (Aguilera-Rodriguez et al. 2025)},
  author = {Edgar Aguilera-Rodriguez and Alma Cuevas-Romero and Santiago Mendoza-Franco and Jonathan Wornovitzky-Green and Eduardo Rivera-Cerros and David Villanueva-Cazares and Luis Alberto Munoz-Ubando and David Ibarra-Zarate and Luz Maria Alonso-Valerdi},
  doi = {10.82901/nemar.nm000174},
  url = {https://doi.org/10.82901/nemar.nm000174},
}

§ 02Study · The README

About This Dataset#

An EEG dataset of imagined speech from 15 healthy participants comparing traditional cue-based and gamified (Pac-Man maze) paradigms for brain-computer interface applications. The dataset comprises 1,800 trials across four Spanish directional commands (avanzar, retroceder, derecha, izquierda) recorded at 500 Hz from 24 electrodes using the mBrainTrain Smarting system. This derivative dataset enables systematic evaluation of paradigm design effects on imagined speech decoding performance.

EEG data were acquired from 15 healthy participants (8 male, 7 female, ages 18-27) using a 24-channel mBrainTrain Smarting system at 500 Hz sampling rate with FCz reference and Fpz ground. Two experimental paradigms were compared: (1) traditional cue-based with visual and auditory cues (5 beeps at 1.4s rhythm) followed by 7 imagined speech repetitions, with the last 3 repetitions extracted for analysis, and (2) gamified paradigm using a Pac-Man maze interface with 1.4s periods for movement decision, imagined speech, and vocalized speech. Each participant completed 2 sessions with 120 trials per session (30 per class), yielding 450 trials per class across the dataset. The final dataset contains the last 3 imagined speech repetitions per trial from the traditional paradigm.

DOI

Imagined Speech EEG dataset comparing paradigm designs (Aguilera-Rodriguez et al. 2025)

Overview

How to Access via MOABB

Install MOABB and load this dataset directly:

NEMAR Metadata#

[![DOI](https://img.shields.io/badge/DOI-10.82901%2Fnemar.nm000174-blue)](https://doi.org/10.82901/nemar.nm000174) # Imagined Speech EEG dataset comparing paradigm designs (Aguilera-Rodriguez et al. 2025) ## Overview An EEG dataset of imagined speech from 15 healthy participants comparing traditional cue-based and gamified (Pac-Man maze) paradigms for brain-computer interface applications. The dataset comprises 1,800 trials across four Spanish directional commands (avanzar, retroceder, derecha, izquierda) recorded at 500 Hz from 24 electrodes using the mBrainTrain Smarting system. This derivative dataset enables systematic evaluation of paradigm design effects on imagined speech decoding performance. ## Dataset Summary | Property | Value | |---|—| | Subjects | 15 | | Channels | 24 | | Classes | 4 | | Trial length | 4 s | | Sampling frequency | 500 Hz | | Sessions | 1 | | Total trials | 1800 | | Paradigm | MotorImagery | ## Data Collection Methods EEG data were acquired from 15 healthy participants (8 male, 7 female, ages 18-27) using a 24-channel mBrainTrain Smarting system at 500 Hz sampling rate with FCz reference and Fpz ground. Two experimental paradigms were compared: (1) traditional cue-based with visual and auditory cues (5 beeps at 1.4s rhythm) followed by 7 imagined speech repetitions, with the last 3 repetitions extracted for analysis, and (2) gamified paradigm using a Pac-Man maze interface with 1.4s periods for movement decision, imagined speech, and vocalized speech. Each participant completed 2 sessions with 120 trials per session (30 per class), yielding 450 trials per class across the dataset. The final dataset contains the last 3 imagined speech repetitions per trial from the traditional paradigm. ## How to Access via MOABB Install MOABB and load this dataset directly: `python from moabb.datasets import AguileraRodriguez2025 from moabb.paradigms import MotorImagery paradigm = MotorImagery() dataset = AguileraRodriguez2025() X, y, metadata = paradigm.get_data(dataset) ` For more details see the [MOABB documentation](https://moabb.neurotechx.com/) and the [MOABB dataset page](https://moabb.neurotechx.com/docs/generated/moabb.datasets.AguileraRodriguez2025.html). ## Citation If you use this dataset please cite the primary publication: > DOI: [10.1038/s41597-025-05926-5](https://doi.org/10.1038/s41597-025-05926-5) ## NEMAR / MOABB Benchmark Collection This BIDS-formatted dataset was converted from the original data using the [MOABB](https://moabb.neurotechx.com/) pipeline and re-hosted on [NEMAR](https://nemar.org/) as part of the MOABB benchmark collection. The original data and license terms apply — see dataset_description.json for details.

License: CC-BY-NC-ND-4.0

Authors:

Edgar Aguilera-Rodriguez
Alma Cuevas-Romero
Santiago Mendoza-Franco
Jonathan Wornovitzky-Green
Eduardo Rivera-Cerros
… and 4 more

Versions:

Version	DOI	Released
`current`	10.82901/nemar.nm000174

§ 03Cohort · Participants

Cohort#

Dataset Statistics#

Channel counts: 24 ch (n=30 recordings)

Sampling frequencies: 500.0 Hz (n=30 recordings)

Total recording duration: 5 h 1 min

§ 04Signal · Electrodes & trace

Signal · Electrodes & live trace#

Fig. 01 Signal & montage 24 ch · EEG · 500 Hz · 15 subjects, 30 recordings

Live trace viewer — sub-14 · ses-1 · task-imagery · run-0

Showing one representative recording out of 15 subjects and 30 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 · 24 sensors — 24 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.NM000174(cache_dir, query=None, s3_bucket=None, **kwargs)

Bases: EEGDashDataset

Author (year)—

Canonical—

Importable asNM000174

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

class eegdash.dataset.NM000174(cache_dir: str, query: dict | None = None, s3_bucket: str | None = None, **kwargs)[source]#

Imagined Speech EEG dataset comparing paradigm designs (Aguilera-Rodriguez et al. 2025)

Study:: nm000174 (NeMAR)
Author (year):: —
Canonical:: —

Also importable as: NM000174.

Modality: eeg; Subject type: Unknown. Subjects: 15; recordings: 30; 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/nm000174 NeMAR dataset: https://nemar.org/dataexplorer/detail?dataset_id=nm000174 DOI: https://doi.org/10.82901/nemar.nm000174

Examples

>>> from eegdash.dataset import NM000174
>>> dataset = NM000174(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 — NM000174.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 nm000174 to reproduce the tutorial on this dataset.

Citation

Edgar Aguilera-Rodriguez, Alma Cuevas-Romero, Santiago Mendoza-Franco, Jonathan Wornovitzky-Green, Eduardo Rivera-Cerros, … (20). Imagined Speech EEG dataset comparing paradigm designs (Aguilera-Rodriguez et al. 2025). 10.82901/nemar.nm000174

Provenance

¹Contributed to nemar in BIDS format.

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

³Persistent identifier: 10.82901/nemar.nm000174.

Related & sibling datasets

ON004785EEG · 17 subj ON006563EEG · 12 subj ON006695EEG · 19 subj ON003626EEG · 10 subj ON003343EEG · 20 subj

+ 1 more — see See Also below →

BIDS

BIDS 1.9.0

Sidecars

events · events.json · channels · eeg.json

Provenance

CC-BY-NC-ND-4.0 · 10.82901/nemar.nm000174

Machine-readable

schema.org/Dataset · Croissant

Mirrors

OpenNeuro · NEMAR · HF org

Dataset ID	`NM000174`
Title	Imagined Speech EEG dataset comparing paradigm designs (Aguilera-Rodriguez et al. 2025)
Author (year)	—
Canonical	—
Importable as	`NM000174`
Year	20
Authors	Edgar Aguilera-Rodriguez, Alma Cuevas-Romero, Santiago Mendoza-Franco, Jonathan Wornovitzky-Green, Eduardo Rivera-Cerros, David Villanueva-Cazares, Luis Alberto Munoz-Ubando, David Ibarra-Zarate, Luz Maria Alonso-Valerdi
License	CC-BY-NC-ND-4.0
Citation / DOI	10.82901/nemar.nm000174
Source links	OpenNeuro \| NeMAR \| Source URL

NM000174: eeg dataset, 15 subjects#

Quickstart#

About This Dataset#

NEMAR Metadata#

Cohort#

Dataset Statistics#

Signal · Electrodes & live trace#

NEMAR Processing Statistics#

Manifest#

File Explorer#

API Reference#

Citation

Provenance

Related & sibling datasets

See Also#