Please enable JavaScript.
Coggle requires JavaScript to display documents.
4 Face Recognition, Is processing of faces different from processing other…
4 Face Recognition
IMPAIRED VISUAL ANALYSIS
Shape Discrimination
So an impairment at this stage would mean that the patient would have visual sensory problems, where they'd be unable to process basic information such as shapes, motion, colour etc. If you can't see shapes etc. then you're not going to recognise objects.
Patient JAF
This patient had good visual acquity, but couldn't even recognise herself in the mirror, she was unable to distinguish and match shapes to a 'target shape', hence her deficit was at the early stage of visual processing.
Variety of visual information that comes from a face
Expression Analysis
= emotions,
Lip reading
= we hear what we see,
Gender
= if people are male or female,
Age
= young, middle age, old,
Recognising unique identity
= who someone is e.g. say their name.
Double Dissociation for expression analysis
(what emotions people show)
and identity recognition
(who the person is) = classic prosopagnosic can't recognises faces (identity recognition) but is good with emotion recognition (Jansari, 2015). Adolphs et al (1994) showed that patient couldn't recognise emotion (expression analysis) but could recognise who someone was (identity recognition - they had no prosopagnosia).
Double Dissociation for lip reading
(lip reading helps you hear)
and emotional expression analysis
= patient D problem with analysing emotional expression, but was able to lip-read (susceptible to the McGurk illusion), but patient T could analyse emotional expression, but was bad at lip-reading (the McGurk effect didn't fool them).
IMPAIRED PERCEPTUAL ANALYSIS
A coherent percept
This is where the individual can take all the info in, integrate all the information together, and then mentally rotate the object in our minds eye,
Tests to measure perception
Same-different matching task
= two images are shown of people from different angles, are they the same or not?
Face-matching task
= target face at the top, six faces underneath, which face is the face at the top
Brain-damaged patients
Patient S (Bodamer, 1947) could see individual features, but couldn't integrate these features to be able to recognise others.
MODEL OF FACE RECOGNITION
The process of face recognition
Structural encoding, parallel process, directed visual processing, facial speech analysis, expression analysis, facial recognition, personality identity node, name generation.
Dementia/Alzheimer's
This happens in reverse = first the name is wrong, then personality identity nodes break down, then eventually the initial analysis of face recognition goes.
EVAL
= HOWEVER, this model can't explain covert recognition of faces, where
correct face-name associations
were learnt faster by prosopagnosic patients vs. incorrect face-name associations. Additionally,
studies of skin conductance
show there's an automatic unconscious response to familiar faces. Moreover, prosopagnosic patients show
an interference effect
, showing that they unconsciously recognise that the picture and name are different.
WHAT IS PROSOPAGNOSIA
Evolution
Face Recognition had developed through evolution because individuals needed to be in groups to survive in the past, and in order for this to happen, you need to know who is part of the in-group (ally) and who is part of the out-group (foe); hence, face recognition developed.
History
Prosopagnosia is an impairment in face recognition, Charcot (1883) was the first to report a case of face processing problems. Bodamer (1947) developed the term 'prosopagnosia'. This is not a unitary disorder because problems in face recognition can occur in 3 stages: can be impaired in visual analysis, perception or recognition.
IMPAIRED RECOGNITION OF FACES
Patient W
He could see the individual shapes and features on a face, he was able to recognise gender and emotions, he could to the matching and rotation tasks, but he couldn't recognise anyone. This is because his access to his visual representation for faces was impaired.
Matching percept with memory data banks
So for recognition, this is where you use your memory. So after identifying the shapes, integrating all the information/rotating it, you create a coherent percept, and this image has to be matched with your memory of faces for you to recognise whether the face you're looking at is of someone you know i.e. so you can recognise them. After this happens, you then find information about this person e.g. they're the president of the united states.
Patient W can't access this visual representation in their memory to match their face they're seeing in their minds eye.
ARE FACES SPECIAL - HEALTHY PARTICIPANTS
The Inversion Effect
Recognition of faces and buildings that were upright & upside down were tested, found that pps had poorer recognition for faces when faces were upside down = we process faces in a holistic way, and when faces are upside down, this holistic process can't occur. So when faces are upside down, we use featural processing, to look at all the features of the upside down face so we can integrate and match the face to memory - this takes a bit of time hence that's why it's longer to accurately guess a persons face when it's shown upside down.
Feature Processing
= processing info bit by bit.
Holistic Processing
= automatically see the visual scene in one go/the metric distances between features on people's faces
Composite Face Effect
Top half is of someone elses face, bottom half is of someone elses face, either aligned or misaligned, and either upright shown or upside down shown. Found that
misaligned and upright
led to better recall.
Aligned and upright
led to poor recall (holistic happens automatically, then forced to use featural processing).
Aligned & misaligned and upside down
also led to better recall (gone straight to feature processing).
ARE FACES SPECIAL - BRAIN DAMAGED PATIENTS
Disagreement among researchers
Some researchers, such as Damasio, suggest that face recognition isn't special bc nearly all prosopagnosic's also have agnosia (object recognition disorder), hence bc of this co-occurance of symptoms, face recognition isn't special.
They would say that face recognition is part of a unitary structure e.g. other recognition, object recognition and face recognition - hence face recognition isn't special
Double Dissociation between prosopagnosia and agnosia & featural processing and holistic processing
Patient CK
- was agnosic (so he can recognise faces, but not recognise objects), hence configural processing intact but featural processing damaged.
The painting of the man
- CK could point to the nose on the face but not the pear.
Chopping up an image of the face
- CK couldn't recognise a face when it was chopped up, but could recognise the face when the image was stuck back together.
Patient David
- When shown a picture of Ashok, he used featural processing to recognise Ashok's earrings, goatee etc. & said he was George Micheal. I.e. he can do featural processing (which is for objects).
Chopped up image
= showed David the chopped up image and the intact image, he could recognise the image faster when it was chopped up, took him longer to recognise image when it was intact.
Hence, CK can process process intact faces (holistic processing) but not the chopped up face/or objects (featural processing), David can process the chopped up face faster/and earrings (featural processing) but takes longer in the intact face image (holistic processing).
CAPGRAS SYNDROME
Patient S
Had damage to the right hemisphere and frontal lobes. Said that he had two families, his original family he has not see since his accident, his current family is nice but they're one year older than his old family
Combining Prosopagnosia and Capgras Syndrome
Prosopagnosia has damage to the ventral route (which is responsible for conscious visual recognition) i.e. patient David can't recognise his family based on seeing their faces but still has loving feelings for them. Capgras Syndrome has damage to the dorsal cortical route (which is an emotional route to recognition) i.e. patient S is ABLE to recognise his family from their faces, but has no emotional recognition for his family, hence why he thinks his current family is someone different.
Is processing of faces different from processing other objects?
So shows that when faces are upright, we use holistic processing, but when it's upside down, we use featural processing.
These 3 are like the
process
of face recognition