Baron-Cohen et al. (eyes test) Flashcards
Autism characteristics
- forms of communication
- less social awareness
- less emotionally intelligent
- savant syndrome (better memory/IQ)
Baron-Cohen et al. study outline
- mix of disabilities/talents
- Sally Anne Test (had to be revised)
- neurodiversity
- social disabilities
- Asperger’s syndrome
- false belief test:
–> 3/4 children failed test
–> development delay
–> theory of mind (mind blindness)
What is the Sally-Anne test?
- original eyes test (based on Sally-Anne test)
- 1997 (psychometric issues)
- there are ceiling effects (when a test is too easy and everyone’s scores a high mark; can’t identify individual differences)
Strengths and weaknesses of Sally-Anne Test
STRENGTHS:
- engaging for children
- practical
- useful
WEAKNESSES:
- low eco validity
- language
- application for adults
Original eyes test
- 25 photos of eyes
- 2 options
Task analysis
- subject needs to have mental state lexicons + know meaning of terms
- map terms based on facial expressions
- match the eyes to examples of eye region
- arrive at a judgment of which word the eyes closely match to
problem with original eyes test
1 - first version of task involved forced choice between only two response options (score 17 or above/25 to be above chance)
2 - when first version of test was given to parents of children with Asperger’s, they scored too below the general population level
3 - narrow range of scores (above chance on first test) leads to score in normal range being close to ceiling test.
- ceiling effects = undesirable = one loses power to see individual differences
4 - first version of test –> basic + complex mental states
- contained some items that were too easy
- risked producing ceiling effects
5 - original version –> items could be solved by checking gaze directions on face
- words like “noticing” or “ignoring”
6 - original version –> more female than male faces
- unclear if test could’ve been biased in some way
7 - original test –> target word and foil were always semantic opposites which makes the test too easy (eg: concerned vs unconcerned)
8 - eyes-test involves mapping a word to a picture
- unclear if comprehension problems with the words themselves might’ve contributed to individual’s score
1997 problems
- Forced choice
- 50%
- 17 above out of 25
- 17 to 25
- 9 points narrow
- not able to identify individual differences - Parents also scored low
- Narrow range
- ceiling effects
- can’t identify individual differences
SOLUTION:
- increased option –> 4
- 25%
- 13/36 (range) - 24 points - Both basic and complex states
- Gaze direction
- More females than males
- Semantic opposites
- Comprehension
4 words from glossary
- anxious
- cautious
- embarrassed
- decisive
What is the theory of mind?
This is the ability to attribute mental states to oneself or another person. For example:
- mind-reading
- read expressions
- empathy
- mentalizing
Subjects (sample) - Group 1
- adults
- AS/HFA
- diagnosed
- N = 15 (participants)
- male
- volunteer sample
- WAIS
- IQ = 115, SD = 16.1
Group 2
- 122 normal adults
- public library/adult community/education classes
- wide range of time occupations
- unemployment = manual
- clerical = professionals
- mix of education levels
- no education beyond secondary school
- occupationally related training = college degrees
- 88 age known
- opportunity sample
Group 3
- normal
- 103
- undergraduates
- Cambridge
- 53 males, 50 females
- high IQ
- opportunity sample
Group 4
- 14 pps
- general population
- IQ matched group 1
- random selected
- IQ = 116
- SD = 6.4
Revised eyes test (developed)
- 8 judges (4 m, 4f)
- target words + foils were repiloted by 8 judges
- 5 out of 8 had to agree on targeted word for each pair of eyes (inter-rater reliability)
- no more than 2 judges picked any single foil
- items that failed to meet this criterion had new target words, foils, or both generated (validity)
- then repiloted with successive groups of judges until criterion was met for all items
Procedure
Group 1:
- revised eye test
- AQ test (50 statements)
- IQ test (WAIS-R short version)
- gender recognition (revised eyes test)
Group 2:
- revised eyes test
Group 3:
- revised eyes test
- AQ test
Group 4:
- revised eyes test
- AQ test
- IQ test
How was the eyes test developed?
- pilot study
- 8 judges (4m, 4f)
- target words and foils
- most suitable = 5/8
- originally 40 set of eyes
- 1, 2, 12, and 40 did NOT make it
- glossary
Results
- EYES TEST: group 1 did significantly worse off than the other 3 groups
Eyes test scores
- AS/HFA group = 21.9 (mean)
- group 2:
–> fem = 26.4
–> male = 26.0
–> 26.2 - group 3:
–> fem = 28.6
–> male = 27.3
–> all = 28.0 - group 4:
–> all = 30.9
Eyes test (sex difference)
- females scored slightly higher than males in groups 2 and 3
AQ test results
- group 1 scored significantly higher than groups 3 and 4
- no difference between groups 3 and 4
- males in group 1 scored higher than females in group 3
AQ test scores
- AS/HFA group = 34.4 (mean)
- group 3:
–> male = 19.5
–> fem = 16.6
–> all = 18.3 - group 4:
–> all = 18.9
Overall total results
- no correlation between eyes test and IQ test
- no correlation between AQ and IQ
- eyes test and AQ inversely correlated (r = -0.53)
Evaluation of the results
STRENGTHS:
- quantitative data
- easy to compare (grp 1 and grp 4 eyes test scores)
- objective (no need to interpret the information demonstrated)
WEAKNESSES:
- reductionist (no explanation on quantitative data shown)
e.g:
-fems scored higher than males/only looking at one aspect of a person
Conclusions
- the original study replicates the results found in the original eyes test
- AS/HFA pps are significantly impaired compared with non AS/HFA pps
- the modifications made in this test overcame the initial problems of the original eyes test
- revised eyes test is a useful test
- it can identify subtle impairments in social intelligence in adults
Why?
- it uses a series of single case studies
- revised eyes test may be relevant to clinical groups beyond those on autistic spectrum
- adult eyes test has been used during MRI revealing amygdala activity in normal (not autistic) brain
Evaluation (methodological)
METHODOLOGICAL
Strengths:
- quantitative data
- high reliability (standardized procedure)
- inter-rater reliability (when pps were choosing which choice of words to use in the eyes test)
- easy to administer
Weaknesses:
- social desirability bias (self-report to tests from AQ and IQ, control groups)
- low eco validity/mundane realism
- no qualitative data
- low generalizability
Evaluation (ethics)
Strengths:
- confidentiality (we know nothing about the participants except for how many there are)
- consent (not informed consent)
- right to withdraw
Weaknesses:
- possible psychological harm (pps could’ve been under stress when choosing between images in eyes test)
(under distress/mostly people in AS/HFA group)
