Electrophysiological Frequency Band Ratio Measures Conflate Periodic and Aperiodic Neural Activity

Abstract

Acknowledgements 32 33We would like to thank members of the Voytek Lab for insightful comments and suggestions 34 throughout this project. We would also like to express gratitude to the many people involved in 35 generating the open-access datasets and developing the open-source tools that made this project 36 possible. Abstract 39 40A common analysis measure for neuro-electrophysiological recordings is to compute the 41 power ratio between two frequency bands. Applications of band ratio measures include 42 investigations of cognitive processes as well as biomarkers for conditions such as attention-deficit 43 hyperactivity disorder. Band ratio measures are typically interpreted as reflecting quantitative 44 measures of periodic, or oscillatory, activity, which implicitly assumes that a ratio is measuring the 45 relative powers of two distinct periodic components that are well captured by predefined frequency 46 ranges. However, electrophysiological signals contain periodic components and a 1/f-like aperiodic 47 component, which contributes power across all frequencies. In this work, we investigate whether 48 band ratio measures reflect power differences between two oscillations, as intended. We examine 49 to what extent ratios may instead reflect other periodic changes-such as in center frequency or 50 bandwidth-and/or aperiodic activity. We test this first in simulation, exploring how band ratio 51 measures relate to changes in multiple spectral features. In simulation, we show how multiple 52 periodic and aperiodic features affect band ratio measures. We then validate these findings in a 53 large electroencephalography (EEG) dataset, comparing band ratio measures to parameterizations 54 of power spectral features. In EEG, we find that multiple disparate features influence ratio measures. 55For example, the commonly applied theta / beta ratio is most reflective of differences in aperiodic 56 activity, and not oscillatory theta or beta power. Collectively, we show how periodic and aperiodic 57 features can drive the same observed changes in band ratio measures. Our results demonstrate how 58 ratio measures reflect different features in different contexts, inconsistent with their typical 59 interpretations. We conclude that band ratio measures are non-specific, conflating multiple possible 60 underlying spectral changes. Explicit parameterization of neural power spectra is better able to 61 provide measurement specificity, elucidating which components of the data change in what ways, 62 allowing for more appropriate physiological interpretations. 63 64 Keywords 65 66 neural oscillations, frequency band ratios, spectral power ratios, theta / beta ratio, theta / alpha 67 ratio, alpha / beta ratio, electroencephalography, 1/f activity, aperiodic neural activity 68 69 Abbreviations 70 71 EEG: electroencephalography; MEG: magnetoencephalography; ECoG: electrocorticography; LFP: 72 local field potential; TBR: theta / beta ratio; TAR: theta / alpha ratio; ABR: alpha / beta ratio; CF: 73 center frequency; PW: power; BW: b...

Decomposing the role of alpha oscillations during brain maturation

“…The decrease of the aperiodic intercept replicates previous observations (Donoghue, Dominguez, & Voytek, 2020a;He, 2014). While on study (Donoghue et al, 2020a) investigated a correlation across the age range from 6 to 44, which is beyond the here investigated sample, the other study (He et al, 2019) found a decrease between the young and the adult group and also within the young subsample. The present analysis further indicates that this decrease is evident within the process of brain maturation during childhood and adolescence.…”

Standard multiscale entropy reflects neural dynamics at mismatched temporal scales: What’s signal irregularity got to do with it?

“…Smaller exponents (indicating shallower PSD slopes) characterize signals with relatively strong high-frequency contributions (i.e., reduced temporal autocorrelations, and less predictability) compared to larger exponents that indicate steeper slopes. This conceptual link between PSD slopes (or high-to-low frequency power ratios that may have strong broadband slope contributions [39]) and sample entropy has been empirically observed across subjects, wakefulness and task states [14,17,40]. However, the sensitivity of fine-scale entropy to PSD slopes-a multi-scale characteristic-highlights that the contribution of slow-to-fast signal content to fine-scale entropy is unclear.…”

How can noise alter neurophysiology in order to improve human behaviour? A combined tRNS and EEG study

“…The relationship between TBR and behavioural outputs may instead be explained in differenced in the underlying aperiodic signal rather than theta and beta band specific changes 29,57 . To explore this, we extracted and analysed the aperiodic exponent from the resting and online power spectra.…”

“…Therefore, TBR may predict the effect of tRNS by improving sustained attention during tRNS. Alternatively, TBR might reflect the underlying background noise (the aperiodic signal) of the power spectrum 29 . High aperiodic exponents are thought to reflect lower excitation/inhibition ratios (E/I ratio) 30,31 .…”