Separating Fractal and Oscillatory Components in the Power Spectrum of Neurophysiological Signal

By Haiguang Wen, Zhongming Liu. 2015

Subfields: Cognitive Neuroscience, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Cognitive Neuroscience

In Signal Processing

Citations in the corpus

The development of aperiodic neural activity in the human brain (2024, unknown): “The irregular-resampling auto-spectral analysis method 9 (IRASA) was used to estimate the 1/ƒ power-law exponent.”
Non-Neural Sources Systematically Impact Aperiodic EEG Activity (2026, unknown): “…by a previous study 36 , oscillatory peaks near the lower or upper bound of the fitted frequency ranges can bias the estimation of the aperiodic parameters in the algorithm of specparam 1 , but also in an alternative estimation method such as irregular-resampling autospectral analysis (IRASA 37 ).”
A meta-analysis of periodic and aperiodic M/EEG components in Parkinson’s disease (2025, unknown): “Electrophysiological signals reflect a combination of periodic (oscillatory) and aperiodic components, which are postulated to arise from distinct neural mechanisms (Donoghue et al. 2020; Hughes et al. 2012; Wen and Liu 2016).”
Model selection for spectral parameterization (2024, unknown): “In the power spectrum, periodic components can be modelled as Gaussian-shaped peaks emerging from an arhythmic (aperiodic) background (Wen & Liu, 2016; Donoghue et al., 2020; Wilson et al., 2022).”
Age-Related Characteristics of Resting-State Electroencephalographic Signals and the Corresponding Analytic Approaches: A Review (2024, methodology): “It was developed to extract the fractal power by computing the auto-power spectra of a bundle of resampled time-series signals modified from the original signals in terms of Hurst exponent and Fourier transform [74].”
Aperiodic and Hurst EEG exponents across early human brain development: A systematic review (2024, unknown): “…prior to combining into a representative model (Donoghue et al., 2020); or (3) non-integer resampling prior to Fourier-based spectral decomposition, followed by taking the median of the auto-spectral distribution (Irreg-ular Resampling Auto-Spectral Analysis [IRASA]) (Wen and Liu, 2016).”
Evaluating early EEG correlates of restricted and repetitive behaviors for toddlers with or without autism (2024, methodology): “Given concerns about power measures, novel approaches have been developed to decompose aperiodic and periodic components of the EEG power spectrum, avoiding the constraints associated with prede�ned canonical frequency bands of interest [29, 30].”
The development of oscillatory and aperiodic resting state activity is linked to a sensitive period in humans (2023, background): “Recently, it has been recognized that periodic and aperiodic components of the power spectrum carry partially distinct information and that changes in aperiodic aspects might look like changes in the periodic activity (He, 2014; Wen and Liu, 2016; Donoghue et al., 2020).”
Time-resolved parameterization of aperiodic and periodic brain activity (2022, unknown): “Recent methodological advances have contributed practical tools to decompose and parameterize the neural power spectrum (periodogram) into aperiodic and periodic components (Donoghue et al., 2020; Wen and Liu, 2016; He, 2014).”
Globally elevated excitation–inhibition ratio in children with autism spectrum disorder and below-average intelligence (2021, background): “Theoretically, aperiodic part of the spectrum is self-similar, or fractal, across a wide range of frequencies, such that the estimation of the spectral slope should be independent of the chosen fitting range [50, 84, 85].”
Separating Neural Oscillations from Aperiodic 1/f Activity: Challenges and Recommendations (2021, methodology): “For that reason, estimating the 1/f component before determining the power of periodic activity has recently been suggested (Donoghue et al., 2021; Wen & Liu, 2016).”
Decomposing the role of alpha oscillations during brain maturation (2020, background): “These new approaches decompose measured power into periodic and aperiodic signal components (see Figure 4; Donoghue et al., 2020b; Hughes et al., 2012a; Wen and Liu, 2016).”
Aperiodic and oscillatory systems underpinning human domain-general cognition (2024, unknown): “IRASA achieves this by resampling the original signal at slightly shifted rates (e.g., 1.1× to 1.9×), averaging the resulting power spectra to minimize periodic signals, andisolating thebroadband aperiodiccomponent.”
Spatiotemporal patterns of theta-band activity during rapid-eye movement sleep: a magnetoencephalography analysis (2024, methodology): “found to be greater in phasic than tonic REM (Simor et al., 2020) Rosenblum et al., 2024; Wen and Liu, 2016).”
Fronto-medial theta coordinates posterior maintenance of working memory content (2021, methodology): “To obviate the possibility that any difference in these values reflects a shift in the slope of the 1/f component of the EEG signal (Donoghue et al., 2020), the IRASA method (Irregular Resampling Auto-Spectral Analysis; (Wen & Liu, 2016)) was employed to remove the 1/f component from the signal.”