recunoasterea fetei Flashcards
(30 cards)
evolutionary advantage of face recognition
to survive we need to be able to facially recognise other members of our trbie
what is it called when face recognition is impaired
prosopagnosia
what is the history of prosopagnosia
charchot and wilbrand - 1st reports of face-processing problems
bodamer - 1st used the term prosopagnosia - loss of knowledge of faces
3 staged during face processing
the problems can occur at a number of different stages during face processing:
generally 3 subtypes of problems:
- impaired visual analysis
- impaired perception analysis
- impaired face recognition
impaired visual analysis
patients with visual sensory problems find it almost impossible to perceive faces due to problems with shape discrimination
Dr P mistook his wife for a hat
an example of impaired visual analysis
JAF
could read small newspaper print
couldnt recognise herself in a mirror and said what she saw didnt look like a face
she had inability to discriminate shapes
This is a general problem rather than one specific to face-processing
different information derived from faces
once initial visual analysis is complete, there is a rich variety of information coming from a face
expression analysis
lip-reading
gender
age
recognition of unique identity
how do we know that these forms of information are separately processed
double dissociations
between expression analysis and identity recognition
expression analysis shown to dissociate from familiar face recognition
double dissociations in expression analysis and identity recognitino
HJA and DY = able to tell what emotion a face showed BUT difficulty in recognising familiar and famous faces
SM = no problem recognising ppl BUT difficulty in decoding their emotion
double dissociations lip reading and emotional expression analysis
- D = prosopagnosic
unable to judge facial expressions BUT ok at imitating facial expressions = problem in analysing expressions was not perceptual
BUT susceptible to McGurk effect = able to lip read
- T = not prosopagnosic
ok at expression analysis
BUT bad at lip-reading = not susceptible to the McGurk effect
what does the double dissociation between expression analysis and lip-reading imply
shows that facial emotions and lip movement are processed in parallel
IMPAIRED PERCEPTUAL ANALYSIS
perception is different from simply being able to see the object - involved integrating all the information together
how is perceptual analysis tested
same-different matching test
say if a face that is facing to the front is the same as a face that you can see its side profile
impaired perceptual analysis example
S= able to pick out individual features e.g. nose
could distinguish faces from other stimuli
BUT said all faces looked like flat oval plates with dark eyes and couldnt even recognise his mum
Therefore different information can be seen individually but it cannot be integrated
impaired recognition of faces: an example
W - bilateral occipital lobe lesions
OK at picking out picture of face from array of other pictures
OK copying line-drawings of faces
OK telling gender and facial expressions
OK matching unfamiliar faces, pictures of faces from different angles
BAD at recognition of anyone for a year after his stroke
when shown videos of himself and ppl known to him (e.g. health professionals), with no identifying characteristics (hairstyle), he said all faces were unfamiliar
if allowed any other info e.g. voice, her performed better
problems with accessing stored visual representations of ppl from the face
is prosopagnosia a memory problem
amnesics can have problems with face recognition but this is due to a problem in the memory system not in the face-processing system
therefore, cues tend to provide limited benefit to them
HOWEVER, prosopagnosics can greatly benefit from non-facial cues e.g. W benefitted from cues such as hairstyle
are faces special
studies on how human faces are processed in normal control subjects suggest specialised mechanisms for faces
studies on patients with brain-damage suggest differential problems with faces
are faces special - face inversion effect
tested recognition of pictures of faces and of buildings either upright or inverted
recognition of faces is more accurate than that of buildings when upright
recognition of faces is less accurate when inverted
what is the interpretation given for the face inversion effect
that there are 2 different types of processing:
featural processing: bit-by-bit adding up of the jigsaw
configural or holistic processing: seeing the whole
most objects are visually recognised using featural processing
BUT faces are processed using configural/holistic processing
configural processing can only work for UPRIGHT faces and it is AUTOMATIC
When faces are turned upside down they have to be processed using featural processing which distrupts recognition
configural/holistic processing in healthy subjects
presented subjects composite pics of well-known ppl where top half was one person and bottom half another person
ss had to name the top half either when presented with the wrong lower half aligned or misaligned
top halves were much more difficult to name when presented aligned with the incorrect half
BUT when the composites were turned upside down, ss became more accurate at naming the same target
what does this suggest
creating the aligned composite face creates a new configuration which we do not have conscious control of - it happens automatically and so the halves cannot be seen independently
HOWEVER, upside down , since configural processing is disrupted, there is no such composite
are faces special - evidence from brain-damaged patients
double dissociation between object recognition difficulties and prosopagnosia
if human faces were not special, then the prediction would be that prosopagnosics should have problems with all faces
BUT
a farmer who became prosopagnosic still managed to learn to recognise the individual faces of his sheep
configural/holistic processing in patients with brain damage
CK could recognise faces but not objects
hypothesised thatCK could use configural processing but not featural piece-meal processing
Bruce and Young model of face recognition
initial analysis of the face - structural encoding
number of parallel processes:
1. directed visual processing - needed for matching a face across different viewpoints
- facial speech analysis - lip reading
3.expression analysis
- face recognition - through face recognition units = each contains the structural description of a persons appearance and fires when a seen face resembles the description that is held
