Evaluating the Evidence for the Functional Inhibition Account of Alpha-band Oscillations during Preparatory Attention.

The functional inhibition account states that alpha-band (8-14 Hz) power implements attentional control by selectively inhibiting task-irrelevant neural representations. This account has been well supported by decades of correlational research showing attention-related changes in the topography of alpha power in anticipation of task-relevant stimuli and is a viable theory of how attention impacts sensory processing, namely, via alpha power changes in sensory areas before stimulus onset. In addition, attention is known to modulate neural responses to stimuli themselves. Thus, a critical prediction of the functional inhibition account is that preparatory alpha modulations should explain variance in the degree of attention-related modulation of neural responses to stimuli. The present article sought evidence for or against this prediction by scouring the literature on attention and alpha oscillations to review papers that explicitly correlated attention-related changes in prestimulus alpha with attention-related changes in stimulus-evoked neural activity. Surprisingly, out of over 100 papers that were examined, we found only nine that explicitly computed such relationships. The results of these nine papers were mixed, with some in support and some arguing against the functional inhibition account of alpha. Our synthesis draws out common design features that may help explain when effects are observed or not. Even among studies that do find correlations, there is inconsistency as to whether preparatory alpha modulations are predictive of sensory or postsensory components of stimulus responses, highlighting avenues for future research. A clear outcome of this review is that future studies on the role of alpha in attentional processing should analyze correlations between attention effects on alpha and attention effects on stimulus-evoked activity, as more data pertinent to this hypothesized relationship are needed.

Informations de copyright

© 2023 Massachusetts Institute of Technology.