Seeing and hearing lecture 5

?
What can we work out from peoples faces?
Gender, Identity, emotions, what they are like,
1 of 66
Why is facial recognition a hard problem?
Because in real life people don't differ by that much (have eyes, noses, mouth), the features that change between us are subtle distinctions. Recognising a face involves subtle distinctions between items containing the same features
2 of 66
How does a task association test show that we recognise faces differently to other objects.
When we manipulate a face in a certain away and that affects how we recognise it but we manipulate an object in the same way and it doesn't change how we recognise it it shows that faces are perceived differently to objects
3 of 66
Is facial recognition disproportionately affected by inversion?
Yes
4 of 66
What is the thatcher illusion?
That if you flip their head upside down but make the eye and mouth face the right way people don't notice and just perceive it as upside-down, or it takes them a really long time to notice!
5 of 66
What is the face inversion affect?
The meaning of the face changes when it is upside-down
6 of 66
What is the conclusion did Yin make about the face inversion affect?
We are putting the features together as a whole and that is how we recognise them, and when they are being turned upsidedown that is what disrupts them. We see faces holistically (as a whole)
7 of 66
What is the part-whole effect study?
People learnt peoples faces and they were asked to either then point out what his nose was or out of two pictures which one was his face (and they just changed one thing).
8 of 66
What were the results of the part-whole effect study?
Found people can recognise whole faces better than parts but not when they were upside-down (inverted)
9 of 66
What was the conclusion of the part-whole study/idea?
We put all the parts of a face together and that is how we recognise faces, not by parts.
10 of 66
What is the composite effect of faces study?
People learned the top half's of someone's face. Then that half was placed with the bottom half of someone else's face. Ps were told to ignore the bottom and try and recognise the top. People were worse when doing that instead of just tops alone
11 of 66
With the composite effect of faces study if three faces share the same forehead what happens?
People do not recognise that the foreheads are all the same
12 of 66
What happens if you invert or mis-align the tops and bottom halfs in the composite effect study?
The effect is undone and people can recognise the foreheads again
13 of 66
What happens if you give two people the same facial features (the comopsite effect(?
They are often seen as the same people
14 of 66
What is the negation effect?
If you change the polarity of faces people struggle to recognise them
15 of 66
What is the bogart effect?
Reversing the contrast has a strong effect on gaze perception (where you think someone is looking)
16 of 66
Face parts are ___ affected by inversion
Disproportionately
17 of 66
Face parts are recognised more easily when in a ______
Face context
18 of 66
The fusing of two halves of a face into a composite happens automatically and ____ recogntion
Disrupts
19 of 66
Face processing is ____ or ___
holistic or configural
20 of 66
What is the Bruce and Young model of facial recognition?
Describes the processes involved in recognising a face, starts with visual face input then goes to structural encoding (view -centred or expression-independent descriptions). So have centred and independent pathway
21 of 66
What is the Bruce and Young model of facial recognition step 1?
You need to encode the description of the face
22 of 66
What is a view-centred description used for? ( Bruce and Young model of facial recognition)
Recognises expressions, based on current view, are they speaking or not
23 of 66
What does a view-independent description used for? ( Bruce and Young model of facial recognition)
Learning across multiple views, recognise people from different angles and expressions
24 of 66
What is the View-independent pathway used for? ( Bruce and Young model of facial recognition)
How we learn how someone looks like
25 of 66
What does a view-centred pathway support? ( Bruce and Young model of facial recognition)
Processing of expression and speech. Used for facial speech analysis and expression analysis
26 of 66
What does a view-independent pathway support? ( Bruce and Young model of facial recognition)
A generalised view of the face that leads to familiarity, identification and naming. Face recognition units, person identity nodes -> name generation
27 of 66
When accessing a stored representation of a face what does the Face Recognition unit do? ( Bruce and Young model of facial recognition)
They are activated according to similarity and signal that the face is familiar. Looks at memories/ templates to see if there is a match. Identifies familiar faces
28 of 66
When accessing a stored representation of a face what does the Person Identity Nodes do? ( Bruce and Young model of facial recognition)
Link the face to the facts about the person (e.g. job) then links to name generation
29 of 66
When accessing a stored representation of a face what does the Name generation do? ( Bruce and Young model of facial recognition)
Retrieves the person name, this comes last
30 of 66
When accessing a stored representation of a face what order does the person identity nodes, name generation and race recognition units come in? ( Bruce and Young model of facial recognition)
Goes Face recognition units, person identity nodes and then name generation.
31 of 66
What evidence did Bruce and Young et al provide for their model of facial recognition?
In the form of decision time, familiarity decisions are the faster, then matching to occupation is faster than naming and that names are never recalled without the above semantic information
32 of 66
How has the Bruce and Young model of facial recognition model been extended?
Burton, Bruce and Johnston's IAC model uses a connectionist architecture to represent information in the Frus (facial recognition units) and Pins (personal identity notes) , their model has different modules connected to nodes
33 of 66
What is priming?
The process that if you repeat something people get faster the second time around, and by changing what you repeat you can find something out about the representations involved.
34 of 66
What is semantic priming?
You are quicker to recognise a person if you see something/a person which is related to them before seeing them
35 of 66
Is there a difference between names and faces when doing semantic priming?
No, because it activates the same nodes
36 of 66
What does semantic priming have to do with facial recognition models?
The idea is that priming spreads activation in models which makes it easier to recognise people. Activation feeds back to the personal identity nodes (Pins) making them easier to activate later
37 of 66
What has Schweinberger and Burton added to Bruce and Young's facial recognition model?
They have a route based on the current view for expressions and another route based on name recognition. However they have added an emotional route to recognising who someone is, you recognise someone and you feel an emotion towards them,
38 of 66
Why is Bruce and Young's model good?
It makes explicit the differences between familiar and unfamiliar faces and identity vs expression recognition.
39 of 66
What is the deficit in recognising faces called?
Prosopagnosia?
40 of 66
How do you get Prosopagnosia?
From brain damage
41 of 66
Can prosopagnosia been discriminated between visual object agnosia?
Sometimes. Sometimes people can recognise objects and not faces, sometimes faces and objects and sometimes can't recognise either!
42 of 66
What is a double dissociation?
When a group of people who are impaired in one thing and another. With mirrored impairments.
43 of 66
What is a signal dissociation?
When a group of people are impaired in one thing but not another thing
44 of 66
Why is it suggested that there is a double dissociation between prosopagnosia and visual object agnosia?
Because sometimes people can recognise objects but not faces and faces but not objects suggesting a double dissociation
45 of 66
What does it mean if there is a double dissociation between prosopagnosia and visual object agnosia?
That there are two different parts of the brain used to do the two different things
46 of 66
What don't prosopagnosics show?
A holistic processing
47 of 66
How can we think about Prosopagnosia in terms of the facial recognition models?
There is a difference between detection and recognition could relate to the different stages, as they still can identify a face just they don't know who's it is.
48 of 66
How to prosopagnosics recognise familiar and unfamiliar faces?
They are quite bad at recognising familiar faces however are worse at unfamiliar faces and struggle to learn new faces. This could be down to a stronger activation of the network or the use of semantic pins (personal identity nodes)
49 of 66
How does how prospagnosics struggling with who someone is relate to Bruce and youngs model?
Prosopagnosics usually struggle with knowing who someone is, not the facial expression which relates to Bruce and Young's model and they might have issues with the view-independent route.
50 of 66
What is covert recognition in prosopagnosia?
That even though Prosopagnosics cannot name a face they can show semantic priming effects which shows that some information is being processed . They show some sort of implicit recognition, they also show physiological responses to familiar faces.
51 of 66
What does prosopagnosics having covert recognition suggest?
That their is another pathway to facial recognition
52 of 66
What is Capgras delusion?
The belief that someone has been replaced by someone else who looks the same
53 of 66
What does covert recognition and capgras delusion suggest?
That there is double dissociation between an emotional route and a facial recognition route which suggests two pathways
54 of 66
What is it which is impaired if someone is suffering from the Capgras delusion?
They have issues with the emotional route of facial recognition however they can still explicitly recognise them they just cannot feel the emotions towards them
55 of 66
What is the part of the brain which is beleived to be responsible for how people recognise faces?
The fusiform gyrus
56 of 66
How many fusiform gyruses do we have?
Two in each side of the brain under the occipitable lobes by the temporal lobes
57 of 66
What is the fusiform face area (FFA)?
The area which activates when we see faces
58 of 66
What is evidence for the fusiform face area (FFA)?
There is greater activation there than in other areas of the brain when seeing faces compared to non face objects
59 of 66
What is evidence for the fusiform face area (FFA)?
Prosoprasnosics often have damage to the FFA.
60 of 66
What is evidence for the fusiform face area (FFA)?
If the FFA is stimulated people see objects as faces and peoples faces changes
61 of 66
What is the argument against the fusiform face area FFA?
That we are all face expects, and the FFA may just be a domain general system for classes of that object we are experienced with. We see them all the time and are rewarded for that , the FFA isn't for faces but for things we are experienced with.
62 of 66
What is evidence for the fusiform face area FFA being for things we have expertise in and not faces?
Tested people in brain scanners looking at faces, cars and birds and those who are car experts looking at cars the FFA is activated.
63 of 66
What is evidence for category-specific agnosia?
That people with prosopagnosics can often recognise things they are experts in like sheep or cars
64 of 66
What did Gauthier do to test the FFA?
Came up with Greebles and got Ps to study them and then showed in FMris and facial recognition tests that they interacted with greebels the same way as faces. So they concluded that the FFA is flexible
65 of 66
What is the argument against Greebles?
Patient who was a prospasnoic could do Greebles showing that there is something special and extra about faces.
66 of 66

Other cards in this set

Card 2

Front

Why is facial recognition a hard problem?

Back

Because in real life people don't differ by that much (have eyes, noses, mouth), the features that change between us are subtle distinctions. Recognising a face involves subtle distinctions between items containing the same features

Card 3

Front

How does a task association test show that we recognise faces differently to other objects.

Back

Preview of the front of card 3

Card 4

Front

Is facial recognition disproportionately affected by inversion?

Back

Preview of the front of card 4

Card 5

Front

What is the thatcher illusion?

Back

Preview of the front of card 5
View more cards

Comments

No comments have yet been made

Similar Psychology resources:

See all Psychology resources »See all Visual System resources »