Baron Cohen - reading the mind in the eyes

cognitive approach

Assumptions of the cognitive approach:

1. There are similarities and differences in the way that humans and computers process information.

2. There are individual differences in cognitive processes like learning, thinking, attention, and memory this explains our differing behaviors and emotions

Aims:

1. ASD group would score significantly lower on the eye test as compared to the control group.

2. The ASD group would score significantly higher on the AQ test as compared to the control groups 

3. The females of the “normal” group (2&3) would score higher on the eyes test than the males of that same group

4. Males in the normal group would score higher on the AQ test as compared to females

5. The eye test and AQ test results will be negatively correlated.

Psychology being Investigated:

Theory of mind

1. The ability to attribute mental states to ourselves and other people

2. This can include desires, emotions etc

3. Its about using this knowledge to understand, explain and predict the actions of others.

4. Its also about understanding that other people have unique perceptions and world views that may be different from our own.

ASD: 

A neurodevelopmental disorder impairing a child's ability to communicate and interact. Symptoms include inability to identify facial expressions, and repetitive behaviours.

• AS and HFA fall under this

  
  IQ: 

  A measure of intelligence that produces a score representing a person’s mental age. Average=100

  WAISR scale used to measure  

  
  EQ:

  Measure of emotional intelligence

  
  
  
  Ceiling effect: 

  When the majority of participants meet the highest score possible. Individual differences in participants are difficult to identify, which is an issue.

  
  

Background: - Issues with original procedure

1. FORCED CHOICE 2 OPTIONS and opposite words too - revised version has 4 and 3 foils arent the opposite

2. CEILING EFFECT 25 - 36 eyes in revised eye test

3. BASIC AND COMPLEX EMOTION - only complex in revised

4. WORD MEANING - glossary given in revised

5. MALE AND FEMALE IMBALANCE  - equal male and female eyes in revised

6. DIRECTION OF EYES - removed

Research method:

Quasi-experimental - almost an experiment -  has controls, but no random allocation of participants

Sample:  Group 1

• 15 adult males

• Mean age 29.7 yrs

• Average IQ 115

• Volunteer sampling - magazine adverts (Autistic Society magazine)

Sample: Group 2

Comparative control groups of adults

• 122 adults

• From the adult community and educational classes at Exeter UK and the public library in Cambridge

• 55 male 67 female

• Mixed occupation and education

• Mean age 46.5

Sample: Group 3

Student comparison group

• From University of Cambridge

• Mean age 20.8

• Smart - not representative

• 103 students

Sample: Group 4

IQ matched with ASD group

• 14

• 116 IQ

• 28 mean age

• Selected randomly from the general population

Problems with original test and the solutions:

1) FORCED CHOICE BETWEEN TWO OPTIONS - foil was opposite of target: 3 foils that arent exact opposite

2) CIELING EFFECT - 25 questions to 36 so it isn’t as easy to get high score

3) BOTH BASIC AND COMPLEX EMOTIONS - only complex emotions

4) COULD TELL EMOTION FROM DIRECTION - removed

5) MALE/FEMALE IMBALANCE - equal

6) WORD MEANING - glossary given

Creating the test:

• 40 sets of eyes, target words, and foils

• Atleast 5 of 8 judges had to agree target word was appropriate

• If more than 2 judges selected a foil word then a new target word, set of foils or both were generated.

• When the group 2 and 3 results came for four questions, the groups consistently chose the foil, as the normal group would ideally get most questions correct, these 4 were removed from the eye test and not considered in the analysis

procedure

To begin the eye test each participant read the glossary to familiarise themselves with the words, and were told they could refer back to it whenever necessary.

All groups completed the eyes test, group 1 completed a gender identification test to prove that they could accurately identify other aspects of the eyes not associated with theory of mind. A pilot study showed that ‘normal adults’ scored 100% on the gender recognition test so the control groups were not tested on this.

Everyone except group 2 completed the AQ test at home and sent it back via post.

All participants were aware of the nature of research, and data was kept confidential, individual participants cannot be identified from it.

results

n mean age eye test AQ

15 29.7 21.9 34.4

122 46.5 26.2 -

103 20.8 28.0 18.3

14 28.0 39.9 18.9

real life applications

Evaluation:

Strengths:

Internal validity - 

→ 8 judges - 5 had to agree for a set of target and foils to pass

→ only complex emotions

→ 36 eyes to avoid ceiling effect

→ participants tested individually in a controlled setting - to ensure others’ responses don't influence their own

→ glossary provided to ensure that all participants knew what the words meant

Reliability - 

Standardisation

→ images were the same size black and white

→ 4 words, 1 target and three foils for each image

→ all words judged on the same criteria by 8 judges

→ participants were all administered the test individually

→ all participants were provided a glossary

Aims the replicability of the study. This makes it easier to test for reliability.

Weaknesses:

Mundane realism - ecological validity:

The images are not dynamic while real life is, theory of mind is about more than eyes in real life you can see the rest of a person. 

Generalisability - 

The study’s Asperger syndrome/high-functioning autism (AS/HFA) group consisted entirely of male participants, which limits conclusions about how these findings apply to females with AS/HFA. Given that research suggests potential differences in how autism presents in males versus females, the lack of female representation may reduce the study’s ability to generalize its findings across genders.