The good, the bad, the complex: Investigating the behavioural consequences of EEG alpha oscillations in visual detection and selective attention

Abstract

We heavily rely on vision to represent the world around us. Quite surprisingly, our ability to perceive identical visual stimuli varies considerably. Understanding why we sometimes miss a stimulus that we could easily see just a moment ago promises exciting insights into how we perceive and interact with the world. This variability has to depend at least to some extent on variability in the how the brain processes these stimuli. Recent research has related fluctuations in ongoing brain activity, especially in the alpha band (8-12 Hz) of the scalp-recorded electroencephalogram (EEG), with variability in the detection of visual stimuli. Lower alpha power over posterior cortex is generally assumed to lead to superior detection performance. It has also been shown that alpha oscillations play a role in the orienting of selective attention, which allows us to focus on task-relevant stimuli and to inhibit the rest. Alpha power is usually shown to be lower over the side of the head where the attended stimuli will be processed and higher over task-irrelevant sites (i.e. it is lateralised). This thesis provides an in-depth investigation of the behavioural consequences of alpha power fluctuations in visual detection and selective attention. A series of three studies demonstrates that although alpha power fluctuations are related to the behavioural responses in visual tasks, low or high power cannot easily be related to either good or bad performance. Instead, lower alpha power is related to a less cautious and faster response. This association between alpha power and factors that are orthogonal to performance is similar regardless of whether alpha power fluctuates spontaneously or whether those fluctuations are controlled by attentional cues. In fact, the cue-related lateralisation of alpha power seems to index some other mechanism that is not directly related to the behaviourally relevant bias. We thus reveal the complexity of the relationship between alpha power fluctuations and behavioural performance in visual detection and selective attention tasks. EEG alpha oscillations influence how we respond to visual stimuli, and not directly their sensory coding.