HypypData

Sample datasets for the HyPyP tutorials — the Hyperscanning Python Pipeline.

These files are automatically fetched and cached by hypyp.datasets using pooch. You should not need to clone or download this repository manually.

from hypyp import datasets

# EEG (MNE .fif)
epo1_path = datasets.eeg_epochs(participant=1)

# fNIRS (SNIRF)
snirf_paths = datasets.fnirs_samples()

# XDF (Lab Streaming Layer)
xdf_path = datasets.xdf_dyad_noise()

---

Repository Structure

HypypData/
├── eeglab/           EEG data in EEGLAB format (.set / .fdt)
├── fnirs/
│   ├── samples/      fNIRS demo files (SNIRF, mono-subject proxy)
│   ├── DCARE/        DCARE study — real hyperscanning dyad
│   ├── FCS01/        Raw fNIRS recordings (child + parent)
│   └── lionirs/      Channel grouping reference files
├── matlab/           Legacy MATLAB EEG recordings
├── mne/
│   ├── epochs/       Preprocessed EEG epochs (.fif)
│   └── raw/          Raw EEG recordings (.fif)
└── xdf/              Multi-stream LSL recordings (.xdf)

---

Datasets

EEG — MNE Epochs (mne/epochs/)

File	Size	Description
participant1-epo.fif	~16 MB	EEG epochs, participant 1
participant2-epo.fif	~2 MB	EEG epochs, participant 2

Origin: Demo data originally distributed with HyPyP. These are not original hyperscanning recordings — they are single-subject EEG epochs repurposed to demonstrate the HyPyP dual-participant pipeline.

How to load:

import mne
from hypyp import datasets

epo1 = mne.read_epochs(datasets.eeg_epochs(participant=1), preload=True)
epo2 = mne.read_epochs(datasets.eeg_epochs(participant=2), preload=True)

---

EEG — MNE Raw (mne/raw/)

File	Size	Description
sub-110_session-1_pre.fif	~1 MB	Raw EEG recording, subject 110, session 1 (pre-condition)

Origin: HyPyP demo data. Usage: not currently referenced in any tutorial notebook.

How to load:

import mne
from hypyp import datasets

raw = mne.io.read_raw_fif(datasets.eeg_raw(), preload=True)

---

EEG — EEGLAB Format (eeglab/)

File	Description
eeglab_data.set + .fdt	Continuous EEG (32 channels)
eeglab_data_epochs_ica.set + .fdt	Epoched, ICA-cleaned EEG
eeglab_chan32.locs	Channel location file (32-ch montage)

Origin: Standard EEGLAB sample dataset, distributed with EEGLAB for demonstration and testing purposes.

Reference: Delorme A & Makeig S (2004). EEGLAB: an open source toolbox for analysis of single-trial EEG dynamics. Journal of Neuroscience Methods, 134, 9–21.

How to load:

import mne
from hypyp import datasets

# Loads epoched, ICA-cleaned data
epo = mne.io.read_epochs_eeglab(datasets.eeglab_epochs())

---

fNIRS — Sample Files (fnirs/samples/)

File	Size	Description
sample_1.snirf	~0.7 MB	Demo fNIRS recording — "child" dyad A
sample_2.snirf	~0.7 MB	Demo fNIRS recording — "parent" dyad A
sample_3.snirf	~0.7 MB	Demo fNIRS recording — "child" dyad B
sample_4.snirf	~0.7 MB	Demo fNIRS recording — "parent" dyad B
slow_breathing.snirf	~0.5 MB	fNIRS recording with slow breathing protocol

Origin: Recorded by Patrice Fortin (PPSP Lab, CHU Sainte-Justine) during an online biosecurity training session.

Important note: sample_1 through sample_4 are mono-subject recordings segmented into four windows to simulate two hyperscanning dyads. They are used exclusively as proxy data for tutorial demonstrations. Connectivity results computed from these files have no hyperscanning interpretation. A real hyperscanning fNIRS dataset is planned as a future replacement.

Montage: Same probe layout as the SYNCHRO project (Gallagher Lab, CHU Sainte-Justine).

Format: SNIRF v1.0 (Shared Near Infrared Spectroscopy Format).

How to load:

import hypyp.fnirs as fnirs
from hypyp import datasets

paths = datasets.fnirs_samples()          # returns [sample_1, sample_2, sample_3, sample_4]
slow = datasets.fnirs_slow_breathing()   # returns slow_breathing.snirf

rec = fnirs.Recording().load_file(paths[0])

---

fNIRS — DCARE Study (fnirs/DCARE/)

File	Size	Description
DCARE_02_sub1.snirf	~4.6 MB	DCARE dyad 02, participant 1
DCARE_02_sub2.snirf	~4.6 MB	DCARE dyad 02, participant 2
MCARE_01_probeInfo.mat	~3.8 MB	Probe geometry and channel info (MATLAB)

Origin: DCARE project, PPSP Lab, CHU Sainte-Justine. Real hyperscanning fNIRS dyad.

---

fNIRS — FCS01 Raw Recordings (fnirs/FCS01/)

Raw fNIRS recordings for one dyad (child + parent), in the Homer2/NIRS multi-file format.

Path	Description
FCS01/child/NIRS-2019-09-28_002.*	Child participant, 12 files
FCS01/parent/NIRS-2019-09-28_002.*	Parent participant, 12 files

File extensions: .nirs, .dat, .hdr, .wl1, .wl2, .tpl, .set, .evt, .avg, .inf, _probeInfo.mat, _config.txt

Origin: FCS01 study, PPSP Lab, CHU Sainte-Justine.

---

fNIRS — LIONirs Reference (fnirs/lionirs/)

File	Description
channel_grouping_7ROI.mat	7-ROI channel grouping for standard montage
Standard_Channels.txt	Standard channel list for LIONirs

Origin: Reference files for the LIONirs pipeline. Used for channel-to-ROI mapping in fNIRS preprocessing.

---

XDF — Lab Streaming Layer (xdf/)

File	Size	Description
dyad-example-noise.xdf	~1.4 MB	Two Starstim-32 headsets recording pure noise (no participants)
dyad-example-with-markers.xdf	~0.2 MB	Synthetic noise with event markers
data_with_clock_resets.xdf	~1.2 MB	Recording with LSL clock reset events

Origin: Synthetic recordings generated using LabRecorder with two Starstim-32 amplifiers. No participants were wearing the devices — the signal is pure electronic noise. Used to demonstrate XDF import and LSL stream handling.

Format: XDF (Extensible Data Format), the standard output format of Lab Streaming Layer (LSL).

How to load:

from hypyp.xdf import XDFImport
from hypyp import datasets

xdf = XDFImport(datasets.xdf_dyad_noise(), convert_to_mne=True)
print(xdf.mne_raws_dict)

---

MATLAB — Legacy EEG (matlab/)

File	Size	Description
pce01230807/eeg_recordings/pce01_P1_Rest1.mat	~85 MB	EEG rest recording, participant PCE01, session P1

Origin: Legacy MATLAB recording from the PCE study, PPSP Lab. Not currently referenced in any tutorial notebook.

---

License

Data files retain their original licenses. EEGLAB sample data is distributed under the EEGLAB license. All other files are © PPSP Lab, CHU Sainte-Justine, and are provided for educational use with HyPyP.

Citation

If you use HyPyP in your research, please cite:

Ayrolles A, Brun F, Chen P, Djalovski A, Beauxis Y, Delorme R, Bourgeron T, Dikker S, Dumas G. (2021). HyPyP: a Hyperscanning Python Pipeline for inter-brain connectivity analysis. Social Cognitive and Affective Neuroscience, 16(1-2), 72–83. https://doi.org/10.1093/scan/nsaa141