mne_rt.tools.simulate_nf_session#

mne_rt.tools.simulate_nf_session(duration: float = 120.0, sfreq: float = 256.0, n_channels: int = 64, alpha_frange: tuple = (8.0, 13.0), alpha_amplitude: float = 1.5e-05, background_amplitude: float = 5e-06, n_blinks: int = 15, blink_amplitude: float = 0.00015, muscle_rate: float = 0.05, muscle_amplitude: float = 3e-05, drift_amplitude: float = 2e-05, alpha_reactivity: bool = True, nf_epoch_fraction: float = 0.5, rng_seed: int | None = None, verbose: bool | str | None = None) → tuple[source]#

Generate a realistic multi-artifact EEG simulation for NF testing.

Produces synthetic EEG with:

  • 1/f background noise (pink noise) on all channels

  • Alpha-band oscillations with realistic spatial topography (occipital channels have strongest alpha)

  • Alpha reactivity: alpha power is reduced during simulated NF “active” epochs (mimics ERD during cognitive effort)

  • Eye-blink artefacts at realistic rate (≈15/min) with frontal topography

  • Muscle noise bursts as short high-frequency transients

  • Slow electrode drift (low-frequency random walk)

Parameters:
durationfloat, default 120.0

Total recording duration in seconds.

sfreqfloat, default 256.0

Sampling rate in Hz.

n_channelsint, default 64

Number of EEG channels (must be 32, 64, 128, or 256 for standard biosemi/easycap montages).

alpha_frangetuple, default (8.0, 13.0)

Alpha-band frequency range (Hz).

alpha_amplitudefloat, default 15e-6

Peak alpha oscillation amplitude at occipital electrodes (V).

background_amplitudefloat, default 5e-6

Broadband 1/f noise amplitude (V).

n_blinksint, default 15

Number of blink artefacts to inject.

blink_amplitudefloat, default 150e-6

Peak blink amplitude (V) at Fp1/Fp2.

muscle_ratefloat, default 0.05

Fraction of windows contaminated with muscle bursts (0–1).

muscle_amplitudefloat, default 30e-6

Peak muscle burst amplitude (V).

drift_amplitudefloat, default 20e-6

Peak slow-drift amplitude (V).

alpha_reactivitybool, default True

If True, alpha power is reduced by ~50% during NF epochs.

nf_epoch_fractionfloat, default 0.5

Fraction of total duration where NF-state (reduced alpha) applies.

rng_seedint | None, default None

Random seed for reproducibility.

verbosebool | str | None, default None

MNE verbosity.

Returns:
rawmne.io.RawArray

Simulated raw EEG.

nf_statendarray, shape (n_samples,)

Boolean mask: True where simulated NF-state (reduced alpha) is active.

Notes

The function uses a standard biosemi64 montage channel layout. Channels are sorted into occipital (O1, Oz, O2, POz, PO3, PO4, PO7, PO8), frontal (Fp1, Fp2, AF7, AF8), and remaining groups to assign spatially realistic source weights.

References

Niedermeyer, E., & da Silva, F. L. (2005). Electroencephalography: Basic Principles, Clinical Applications, and Related Fields. LWW.

Pfurtscheller, G., & Aranibar, A. (1977). Event-related cortical desynchronisation detected by power measurements of scalp EEG. Electroencephalography and clinical Neurophysiology, 42(6), 817–826.

On this page