Visual perception is an inferential process, in which prior expectations are combined with actual sensory input. How these prior expectations are represented in the brain and how they impact learning or memory of visual stimuli is still debated.

Individuals differ substantially in how they weight prior and sensory input, and the inappropriate weighting of those has been linked to clinical conditions. In autism, for example, studies have found an over-weighting of the sensory input and an inability to appropriately evaluate priors, while the findings for psychosis are more controversial, and may shift with disease progression, from over-weighting sensory input in early to over-weighting prior knowledge in later stages.

In this symposium, we will tackle some of the open questions by bringing together research on how prior expectations influence visual perception in the healthy brain and clinical conditions, using a variety of methods (behaviour, computational modelling, multivariate fMRI, and MEG). First, Helen Blank (UKE, Hamburg) will discuss how expectations of face-identity are represented and weighted during presentation of context cues and how these expectations influence face processing. Second, Franziska Knolle (TUM, München) will discuss how visual priors are used during stimuli disambiguation in psychosis. Third, Janine Bayer, (UKE, Hamburg) will explain how visual category learning relates to autistic traits. Fourth, Alex Clarke (University, Cambridge) will explore how learned contexts shape the neural reactivation of expected sensory details; and finally, Andrea Greve (MRC, Cambridge) will present behavioural evidence supporting a theoretical framework on expectation violations during learning and memory processes.

Zeitfenster der Veranstaltung (1)

Attention and perception

Chair(s): Helen Blank (Hamburg), Franziska Knolle (München)

Presenter(s): Helen Blank (Hamburg-Eppendorf), Franziska Knolle (München), Janine Bayer (Hamburg-Eppendorf), Alex Clarke (Cambridge, UK), Andrea Greve (Cambridge, UK)