In contrast to canonical, stimulus-driven models of perception, recent proposals argue that perceptual experiences are constructed in an active manner in which top-down influences play a key role. In particular, predictions that the brain makes about the world are incorporated into each perceptual experience. Because forming the appropriate sensory predictions can have a large impact on our visual experiences and visually guided behaviors, a mechanism thought to be disrupted in certain neurological conditions like autism and schizophrenia, an understanding of the neural basis of these predictions is critical. Here, we provide evidence that perceptual expectations about when a stimulus will appear are instantiated in the brain by optimally configuring prestimulus alpha-band oscillations so as to make subsequent processing most efficacious.Abstract
The physiological state of the brain before an incoming stimulus has substantial consequences for subsequent behavior and neural processing. For example, the phase of ongoing posterior alpha-band oscillations (8–14 Hz) immediately before visual stimulation has been shown to predict perceptual outcomes and downstream neural activity. Although this phenomenon suggests that these oscillations may phasically route information through functional networks, many accounts treat these periodic effects as a consequence of ongoing activity that is independent of behavioral strategy. Here, we investigated whether alpha-band phase can be guided by top-down control in a temporal cueing task. When participants were provided with cues predictive of the moment of visual target onset, discrimination accuracy improved and targets were more frequently reported as consciously seen, relative to unpredictive cues. This effect was accompanied by a significant shift in the phase of alpha-band oscillations, before target onset, toward each participant’s optimal phase for stimulus discrimination. These findings provide direct evidence that forming predictions about when a stimulus will appear can bias the phase of ongoing alpha-band oscillations toward an optimal phase for visual processing, and may thus serve as a mechanism for the top-down control of visual processing guided by temporal predictions.