EU research results


English EN
Predictive coding in social perception: a social neuroscientific approach to study the dynamic social brain

Predictive coding in social perception: a social neuroscientific approach to study the dynamic social brain


It is well known that racial stereotypes can influence the way people evaluate and treat people from other racial groups. Recent findings suggest that social knowledge can also directly influence very basic perceptual processes. These findings are in line with computational models that see the brain as a ‘prediction machine’: based on knowledge about the world, the brain attempts to match incoming perceptual information to top-down predictions and expectations. Here, I explore whether stereotype-based social predictions directly alter basic visual perception, and how the brain supports this integration of high-level social predictions with low-level perceptual processes. I propose a series of experiments that use binocular rivalry (in which different pictures are presented to the left and the right eye, which results in only one picture entering consciousness) to test whether stereotype-evoking primes (such as a middle eastern face) can alter how and when socially relevant objects (such as weapons) enter awareness. To explore how the brain supports the dynamic integration of stereotypes with ongoing perceptual processes, I will measure the electrical activity of the brain (the electro-encephalogram or EEG). To chart integration of top-down and bottom-up information across brain wide networks, I will study synchronization over EEG frequency bands associated with these two streams of information. I will employ a novel method of analysing neural data, Granger causality, to test whether oscillatory activation evoked by social primes has a direct causal relation with oscillatory activation related to altered visual awareness. Together, these experiments will elucidate whether basic perception is indeed shaped by social knowledge embedded in stereotypes, and how the brain supports the dynamic integration of existing social knowledge with incoming perceptual information.
Leaflet | Map data © OpenStreetMap contributors, Credit: EC-GISCO, © EuroGeographics for the administrative boundaries


University of Sussex


Sussex House Falmer
Bn1 9rh Brighton

United Kingdom

Activity type

Higher or Secondary Education Establishments

EU Contribution

€ 231 283,20

Administrative Contact

Sarah Mcdonald (Ms.)

Project information

Grant agreement ID: 329134


Closed project

  • Start date

    1 August 2013

  • End date

    11 September 2015

Funded under:


  • Overall budget:

    € 231 283,20

  • EU contribution

    € 231 283,20

Coordinated by:

University of Sussex

United Kingdom