Frequency and power of human alpha oscillations drift systematically with time-on-task

Abstract

Oscillatory neural activity is a fundamental characteristic of the mammalian brain spanning multiple levels of spatial and temporal scales. Current theories of neural oscillations and analysis techniques employed to investigate their functional significance are based on an often implicit assumption: In the absence of experimental manipulation, the spectral content of any given EEG-or MEG-recorded neural oscillator remains approximately stationary over the course of a typical experimental session (~1 hour). Here, we show that this assumption is inaccurate for ongoing neural oscillations in one of the most prominent frequency bands of the EEG/MEG signal, the alpha band (8:13 Hz). We found that alpha peak frequency systematically decreased over time, while alpha power increased. Intriguingly, these systematic trends occurred independently of each other suggesting the existence of two nonstationary endogenous processes inherent in macroscopic alpha band activity. Our findings provide novel insight into the time-varying intrinsic properties of large-scale neural networks.They have important implications for the analysis and interpretation of studies that aim at identifying functionally relevant oscillatory networks or at driving them through external stimulation. Significance statementMounting evidence suggests that neural oscillations underlie cognitive processes in the brain.Hence, mapping endogenous characteristics of oscillatory networks represents an important venture for modern neuroscience. Here, we reveal two endogenous non-stationary processes occurring in the alpha band -a prominent rhythm of the brain. Alpha peak frequency systematically decreased and alpha power systematically increased over ~1 hour of . CC-BY 4.0 International license not peer-reviewed) is the author/funder. It is made available under a The copyright holder for this preprint (which was . http://dx.doi.org/10.1101/263103 doi: bioRxiv preprint first posted online Feb. 12, 2018; 3 psychophysical task performance, even during trial baselines. These non-stationarities do not appear to be two manifestations of a single process but rather two independent effects. These findings inform signal processing and neuro-stimulation techniques and advance our understanding of the physiology of functionally relevant neural oscillations.

Directional coupling of slow and fast hippocampal gamma with neocortical alpha/beta oscillations in human episodic memory

“…Indeed, one could argue that the observed differences are driven by fluctuations in the frequency of a single oscillator. While we are unaware of such a phenomenon in hippocampal gamma, such an effect has been reported in neocortical alpha 53 . Notably however, the reported alpha-band fluctuations were very subtle (<0.5Hz), so it'd be highly questionable to interpret the much larger 25Hz shift between "fast" and "slow" hippocampal power as originating from this alpha-band 'fluctuation' mechanism.…”

“…The 1/f noise was subtracted using the method described above to help pronounce the peaks in the power-spectrum. The data was then smoothed using a Gaussian kernel (full-width half-maximum 200ms; 1Hz) to attenuate inter-and intra-individual differences in spectral responses 53 and to help approximate normally distributed data (an assumption frequently violated in small samples). The data was averaged across all time-points, trials and contacts (separately for the hippocampus and ATL).…”

Prestimulus neural alpha power predicts confidence in discriminating identical auditory stimuli

“…Next, we controlled for potential influences of non‐normality of single‐trial alpha power values and linear change of alpha power (and confidence) across the duration of the experiment (Benwell et al., ). To this end, we first log‐transformed single‐trial alpha power values and, second, removed the linear change in alpha power and confidence across trial number, using the residuals of two separate linear regressions of alpha power and confidence on trial number.…”

Pulsed Facilitation of Corticospinal Excitability by the Sensorimotor μ-Alpha Rhythm

“…The experiment was split into multiple blocks, separated by ϳ10 min breaks to allow for coil cooling and relaxation time for the participant. To account for slow power drifts with time on task (Benwell et al, 2019), in the first 16 subjects, the break was also used to perform a recalibration of -alpha power thresholds (see below) based on 3 min resting-state EEG recordings, whereas in the last 7 subjects a continuous recalibration was implemented in form of a sliding distribution of -alpha power values based on the last 60 s of clean data (excluding 1.5 s intervals post-TMS), as this procedure had been shown in the meanwhile to prevent unnecessarily long intertrial intervals (ITI) that occur when the algorithm waits for the power criterion to be met in the face of slow -alpha power fluctuations (Thies et al, 2018). This resulted on average in slightly shorter and more homogenous ITIs for the last seven compared with the first 16 subjects (3.7 Ϯ 0.7 s vs 4.6 Ϯ 1.1 s), but did not produce any differences between experimental conditions.…”