Keynotes | K&Ts | GACs | Talks | Posters | Search

Poster D in Poster Session D: Wednesday, August 5, 2:00 – 3:45 pm, Kimmel Center, Shorin & Rosenthal Rooms

Presence Expectations Modulate the Neural Signatures of Content Prediction Errors

Martha Cottam¹, Nan Chen², Peter Kok¹, Stephen M. Fleming¹, Nadine Dijkstra¹; ¹University College London, ²Max Planck Institute for Biological Cybernetics

Presenter: Martha Cottam

Rather than being a direct readout of sensory information, perception is a process of inference, shaped by our expectations about the world. Predictive processing frameworks seek to explain how expectations modulate perception and awareness. Recent work has distinguished between expectations about what we are likely to see (content) and whether we are likely to detect anything at all (presence). In the current study, we investigate how these two types of expectations interact in the brain. We first demonstrate a behavioural interaction such that the effect of content expectations is stronger when presence is also expected. Using magnetoencephalography (MEG) and multivariate decoding, we find that presence expectations amplify the neural signatures of content-specific prediction errors (PEs). Our results suggest that expecting to see something rather than nothing dials up the gain on content-specific PEs. We discuss computational hypotheses of how presence expectations may modulate lower-level content representations within a predictive hierarchy.

Topic Area: Methods, Tools, Theory & Neural Coding