Discuss Bruce and Young’s Theory of Facial Recognition (8+16 marks)
Bruce and Young developed one of the most widely accepted models for facial recognition. The model presents facial recognition as a series of stages (sometimes colour coded for ease) which are accessed serially one after the other. At any time only some of the units are activated. There are two paths within the model, one for recognising familiar faces and one for recognising facial expressions. The model applies only to facial recognition rather than general object recognition or word recognition.
As soon as the face is seen structural encoding occurs. Structural encoding view centred descriptions and expression independent descriptions which produce representations of the face when it is first encountered. This builds a very basic representation of the face and this is where the expression analysis is in its early stages. The individuals features are being processed at the same time as the expressions.
From here, the information from the basic analysis in the structural encoding area is refined in three ways. First there is a conclusion about the emotional state of the person through more detailed expression analysis as there is a focus on the lips and what is being said (if anything). This process is called facial speech analysis. It is also at this point that any particular characteristics are picked out in directed visual processing to aid future recognition. This may be anything like glasses, marks on the face, shape of the nose etc.
The facial recognition units are activated by comparing the incoming information with any stored familiar faces. They look for any similarities in the structure of the face e.g. broadness of the nose or the size of the forehead. To build on these, the personal identity nodes are accessed to put information to the face such as any details about the person. Finally, name generation occurs if they know the person being looked at and this is said to be the end of the facial recognition process.