Bruce and young's theory of face recognition

Face recognition complex activity- very important in human interaction. All faces have the same basic features in more or less the same location- yet we are able to distinguish between thousands/may be tens of thousands of faces

We also able to recognise variations of the same face- different emotional expressions with or without make up older versions, under different hairstyles etc. 

Face recognition involves a quite holistic system  -involving the interrelationships between different features

Bruce and Young’s model involves a series of modules that work sequentially and in parallel 

1.       Structural encoding stage

2.      Face recognition units

3.      Personal identity nodes

4.      Retrieve the person’s name  

When our brain identifies something that looks like a human face- structural encoding stage begins a mental representation of the face you are looking at is produced

Basic processing also begins hear- allows you to decide if face is male or female, old or young 

The face is then compared to the face recognition units- include every face we have stored in memory- find a match we feel a sense of recognition 

If there is a match- access the person identity nodes- stores of personal information about the people we have stored faces for 

Another set of processing systems activated at the same time as this process/ weather or not we recognise the face- these are

• Expression analysis
• Facial speech analysis
• Direct visual processing
• Cognitive system 

Makes inferences about someone’s emotional state form their facial features 

Uses visual information from the face when interpreting speech 

This involves additional information  that might aid recognition of the person

• Should be unable to name a face in the absence of other information about that person
• If appropriate face recognition unit is activated- the person identity node is not- should experience a feeling of familiarity- not be able to think of any relevant information about that person.
• Should be able to recognise if someone is familiar before we can remember anything about them
• We should be able to say something about a person before we can name them 

Burton et al- revised original Bruce and Young’s model major changes are that

• No separate store for names 
• A decision about whether or not face is familiar is made at the person identity nodes- intend of at face recognition units 
• Model has been demonstrated in a computer simulation  

Lot of scientific evidence to support B/Y model from people with neurological defects- Young et al 93 investigated face perception in ex-service men who had missile wounds to back of head. Some poor at familiar face recognition some poor at matching unfamiliar faces- others could not recognise facial expressions

All suggest that there are independent systems for different aspects of face recognition.

Rescuer good validity.  

According to the model people  generate the name of a person after we have accessed their personal  details- Young, Hey and Ellis  asked people to keep a diary of problems with face recognition- 1008 incidents  in all but not one involved putting a name to a face without recalling some details of the person

190 occasions when a P could recall some details about a person but was unable to name them 

 Yin Found that face recognition ike buildings/ therefore must be using an independent system and a separate   part of the brain

However if pics were inverted, people much better at recognising objects then faces

It is thought that upright faces are identified holistically   using face recognition units in the B/Y model but when inverted have to resort to recognising them by individual features  

The cog system- not explained  very Cleary – B/Y admit that is a bit of catch all to deal with aspects of processing not accounted for by other prats of the model