Alpha-band frequency and temporal windows in perception: A review and living meta-analysis of 27 experiments (and counting)

Abstract

Temporal windows in perception refer to windows of time within which distinct stimuli interact to influence perception. A simple example is two temporally proximal stimuli fusing into a single percept. It has long been hypothesized that the human alpha rhythm (an 8-13 Hz neural oscillation maximal over posterior cortex) is linked to temporal windows, with higher frequencies corresponding to shorter windows and finer-grained temporal resolution. This hypothesis has garnered support from studies demonstrating a correlation between individual differences in alpha frequency (IAF) and behavioral measures of temporal processing. However, non-significant effects have also been reported. Here, we review and meta-analyze 27 experiments correlating IAF with measures of visual and audio-visual temporal processing. Our results estimate the true correlation in the population to be between 0.39 to 0.53, a medium-to-large effect. The effect held when considering visual or audio-visual experiments separately, when examining different IAF estimation protocols (i.e., eyes-open and eyes-closed), and when using analysis choices that favor a null result. Our review shows that 1) effects have been internally and independently replicated 2) several positive effects are based on larger sample sizes than the null effects and 3) many reported null effects are actually in the direction predicted by the hypothesis. A free interactive web-app was developed to allow users to replicate our meta-analysis and change or update the study selection at will, making this a "living" meta-analysis (https://randfxmeta.streamlit.app). We discuss possible factors underlying null reports, design recommendations, and open questions for future research