Class Notes (1,100,000)
CA (630,000)
UTSG (50,000)
PSY (4,000)
PSY100H1 (1,000)

PSY100H1 Lecture Notes - Dark Skin, Categorical Perception, Elisabetta Canalis

Course Code
Nick Rule

This preview shows pages 1-3. to view the full 10 pages of the document.
PSY421: Person Perception
Lecture 2: Race
January 30, 2012
Depending on how you think about race, there are many ways to think about it
(i.e. Princess of Sweden, skin closer to Elisabetta Canalis, finger print closer to
Japanese lady, malaria resistance closer to other race, etc.)
Definition of Race:
Invented “Caucasian”
Refers to people living at food of Caucasus mountains
“Most beautiful” = original humans, birthplace of humanity (considered
most pure)
Divided into 5 different groups (based on beauty)
Linnaeus 1758, 4-group division
1. Caucasian
2. American
3. Asian
4. Malay (Blumenbach added this 5th group to the original group)
5. African
Latter 4 groups degenerations of original Caucasians
1. Result of adaptation to climates (i.e. Asian eyes = snow blindness)
2. Customs/artistic practices (e.g. skull-binding), could become
3. Changes could be reserved (over generations) by moving back
Not racist:
1. Believed groups to be equal in value
2. Abolish slavery
Black slaves > White masters (morally)
3. Hierarchy based primarily on beauty
If race were based on genetics:
94% or more of DNA is identical across racial groups
Greater variation within groups rather than across racial groups
West Africans may be more similar to White people than to East Africans
DNA polymorphisms show 5 groups:
Sub-Saharan Africans

Only pages 1-3 are available for preview. Some parts have been intentionally blurred.

Europeans & Asians West of Himalayas
East Asians
Inhabitants of New Guinea & Melanesia
Native Americans
Phenotype vs. Genotype
Different looking brothers; similar genotypes, different phenotypes
Race: Mostly based on appearance
Environmental adaptations (e.g. skin color)
Geographic origins
Racial variation confounded by geographic variation
Clines incremental variations in traits across geographic areas
Similarity in appearance does not equal genetic similarity
Skin color of African/Aboriginal Australians
3 most important appearance traits in selecting a partner?
Personal ads: skin colour, hair colour eye colour
Race: A moving target
Racial groups today may not be true
Can see differences where they may not exist
37% of babies described as Native American on their birth certificates described
as another race on their death certificates
5 most salient racial groups in US in the late 1800s?
Whites, Blacks, Germans, Italians, Irish
Irish White; Jews White
Race: A social construction
Important social variable
Proxy for SES, effects of racism
Eye of the beholder, eye of the beholden
How we see others and ourselves
How we assign and perpetuate group membership
Multidimensional Space Models (MDS)
Multiple dimensions are used to perceive and categorize faces into groups
Derives from multidimensional scaling
Statistical procedure
Used to distinguish between different wines
Using dimensions on axis

Only pages 1-3 are available for preview. Some parts have been intentionally blurred.

Each dimension has an axis on the model
MDS of faces:
Dimension 1 Hair: long/short?
Dimension 2 Eyes/Brows: close/far?
Dimension 3 Brows: thin/Thick?
Dimension 4 Jaw: big/small?
Dimension 5 Skin: smooth/rough?
Multidimensional face space Valentine (1991)
Every point represents a face
Every face ever seen is represented somewhere in the field
The dimensional code for basic aspects
Over time, add new dimensions as new faces/types of faces encountered
(i.e. move to foreign country)
Typical faces close to original (similar to average or prototype)
Density of faces on each dimension normally distributed around origin
*Dimensions are theoretical
Norm-based encoding
Prototype at origin
1. All faces stored as vectors from prototype
2. Vectors account for distance between prototype and representations
Exemplar-based encoding
Origin empty
1. Cluster around origin NO PROTOTYPE
2. Similarity between faces is monotonic distance between them
(assume to be Euclidian)
Two white faces are closer together and further apart from Black
Direct comparison of the two types of encoding
- Assume normal distribution
- Use prototypes (theoretical
- Everything relative to
- First observed data better
- Assume normal distribution
- Uses actual faces
- Each face relative to
individual faces
- More parsimonious
Process of MDS
Perceive a face, attempt to map it to a location in the face space
You're Reading a Preview

Unlock to view full version