Stimulus-driven brain rhythms within the alpha band: The attentional-modulation conundrum

Abstract

Two largely independent research lines use rhythmic sensory stimulation to study visual processing. Despite the use of strikingly similar experimental paradigms, they differ crucially in their notion of the stimulus-driven periodic brain responses: One regards them mostly as synchronised (entrained) intrinsic brain rhythms; the other assumes they are predominantly evoked responses (classically termed steady-state responses, or SSRs) that add to the ongoing brain activity. This conceptual difference can produce contradictory predictions about, and interpretations of, experimental outcomes. The effect of spatial attention on brain rhythms in the alpha-band (8-13 Hz) is one such instance: alpha-range SSRs have typically been found to increase in power when participants focus their spatial attention on laterally presented stimuli, in line with a gain control of the visual evoked response. In nearly identical experiments, retinotopic decreases in entrained alpha-band power have been reported, in line with the inhibitory function of intrinsic alpha. Here we reconcile these contradictory findings by showing that they result from a small but far-reaching difference between two common approaches to EEG spectral decomposition. In a new analysis of previously published EEG data, recorded during bilateral rhythmic visual stimulation, we find the typical SSR gain effect when emphasising stimulus-locked neural activity and the typical retinotopic alpha suppression when focusing on ongoing rhythms. These opposite but parallel effects suggest that spatial attention may bias the neural processing of dynamic visual stimulation via two complementary neural mechanisms.

Alpha oscillations do not implement gain control in early visual cortex but rather gating in parieto‐occipital regions

“…Our results are in line with recent studies utilizing a similar cuing attention paradigm when tagging at lower frequencies (Antonov et al, 2020; Gundlach et al, 2020; Keitel et al, 2019). In general, these studies show that the magnitude of the alpha‐band activity does not modulate steady‐state visual evoked potentials (SSVEP).…”

“…It should also be noted that such correlation across participants was not significant in a similar study (Keitel et al, 2019). One plausible explanation is the difference in frequencies of the tagging signals: while we used stimulation at 60–70 Hz, the study by Keitel et al used tagging frequencies within the alpha band (10 and 12 Hz).…”

“…Hence, the attentional dynamics of early sensory gain control and alpha‐band oscillations may represent two complementary mechanisms of spatial attention. In a study, Keitel et al (2019) examined modulations of SSVEPs for stimuli that flickered in the alpha band. Evoked SSVEP signals increased for the attended versus unattended stimulus, whereas the opposite pattern was observed for induced alpha‐band oscillations, although both signals overlapped in the frequency domain.…”

Shared and modality-specific brain regions that mediate auditory and visual word comprehension

“…One prediction from our results is therefore that if the nature of the task was changed from speech comprehension to an acoustic task, the perceptual relevance of word representations would shift from left anterior regions to strongly word encoding regions in the temporal and supramarginal regions. Similarly, if the task would concern detecting basic kinematic features of the visual lip trajectory, activity within early visual cortices tracking the stimulus dynamics should be more predictive of behavioural performance ( Di Russo et al, 2007 ; Keitel et al, 2019 ; Tabarelli et al, 2020 ). This suggests that a discussion of the relevant networks underlying speech perception should always be task-focused.…”

EEG spectral power abnormalities and their relationship with cognitive dysfunction in patients with Alzheimer's disease and type 2 diabetes

“…During eyes-closed wakefulness, one of the most prominent features of the EEG signal is αband (~8-13Hz) activity, leading to the characteristic α peak in the power spectrum (Klimesch, 2012;Keitel et al, 2019). We sought to investigate group differences in this dominant frequency, and whether these differences were related to cognitive function, using two independent metrics.…”