PreVis · How perceptual history shapes vision: neural mechanisms and predictive properties

What we see at any given time is not solely determined by the information currently reaching our eyes. The visual system is indeed thought to leverage on perceptual history – i.e., the stimuli saw in the recent past – to contextualise and optimise perception. The incorporation of past stimulus information into perceptual processing has important consequences, shaping how we perceive external objects and events according to the context in which they are embedded in. Namely, perceptual history can bias perception in an attractive way, making a stimulus to appear more similar to its preceding one than it actually is. The nature and mechanisms of attractive perceptual history effects are however unclear, due in part to seemingly conflicting findings suggesting either that the bias is perceptual in nature, occurring in low-level sensory areas, or a cognitive effect based on decision-making. With PreVis, I will provide a new perspective on these effects, proposing and testing a novel hypothesis. Namely, instead of being uniquely based on past information as currently assumed, attractive perceptual history biases may entail an active predictive mechanism. In this scenario, the attractive bias would arise from a prediction based on past stimuli (rather than the past stimuli themselves) performed by high-level brain areas, and affect the processing of sensory information in lower-level areas. In PreVis, I will use psychophysical techniques coupled with electroencephalography (EEG) and functional magnetic resonance imaging (fMRI), to test this hypothesis and assess the neural signature and correlates of perceptual history and predictive processes. The innovative perspective of PreVis represents an important step forward from the current state of the art, and the project has the potential to clarify the nature of perceptual history effects and the mechanisms involved in this fundamental process shaping perception.