""" .. _ex-read-xdf: ==================== Reading XDF EEG data ==================== Here we read some sample XDF data. Although we do not analyze it here, this recording is of a short parallel auditory response (pABR) experiment :footcite:`PolonenkoMaddox2019` and was provided by the `Maddox Lab `__. """ # Authors: Clemens Brunner # Eric Larson # # License: BSD-3-Clause # Copyright the MNE-Python contributors. # %% import pyxdf import mne from mne.datasets import misc fname = misc.data_path() / "xdf" / "sub-P001_ses-S004_task-Default_run-001_eeg_a2.xdf" streams, header = pyxdf.load_xdf(fname) data = streams[0]["time_series"].T assert data.shape[0] == 5 # four raw EEG plus one stim channel data[:4:2] -= data[1:4:2] # subtract (rereference) to get two bipolar EEG data = data[::2] # subselect data[:2] *= 1e-6 / 50 / 2 # uV -> V and preamp gain sfreq = float(streams[0]["info"]["nominal_srate"][0]) info = mne.create_info(3, sfreq, ["eeg", "eeg", "stim"]) raw = mne.io.RawArray(data, info) raw.plot(scalings=dict(eeg=100e-6), duration=1, start=14) # %% # References # ---------- # .. footbibliography::