What you'll learn

• Why Baron-Cohen et al. developed the Eyes Task as an advanced test of theory of mind.
• How the study was designed, including its participant groups, materials, procedure, results and conclusions.
• How to evaluate the study for AO3: validity, reliability, sampling, ethics and usefulness.
• How to compare it with its paired classic study, Freud (1909) Little Hans.

Why this study matters

Baron-Cohen et al. (1997) is the contemporary core study for the OCR H567 key theme Understanding disorders in the Individual differences area.

It investigates whether adults with high-functioning autism or Asperger syndrome have difficulty interpreting subtle social cues from the eye region of the face.

Today, Asperger syndrome is usually understood within Autism Spectrum Disorder, but the original study used the labels “autism” and “Asperger syndrome”, so you should use those terms when describing the study.

Background: theory of mind

A key idea behind the study is theory of mind.

Earlier theory-of-mind research often used false-belief tasks, such as the Sally-Anne task. In these tasks, a child has to understand that another person can hold a belief that is false. Many autistic children struggled with these tasks, supporting Baron-Cohen’s mindblindness theory: the idea that autism involves difficulty representing other people’s mental states.

However, high-functioning autistic adults and adults with Asperger syndrome could often pass these simpler tasks. Baron-Cohen et al. therefore needed a more subtle adult test.

Aim of the study

The main aim was to test whether adults with high-functioning autism or Asperger syndrome would be impaired on an advanced theory-of-mind task: the Reading the Mind in the Eyes Task, usually called the Eyes Task.

The researchers also wanted to see whether any difficulty was specific to mental-state interpretation, rather than being caused by low intelligence, poor face perception, or a general clinical condition.

Method: telling the story

Design

This was a quasi experiment because the main independent variable was a naturally occurring difference: whether participants had autism/Asperger syndrome, Tourette syndrome, or no diagnosis.

It used an independent measures design, because different participants were in each group. The main dependent variable was the score on the Eyes Task.

Participants

There were four participant groups:

The autism/Asperger group were high-functioning adults with normal intelligence. The Tourette syndrome group was useful because Tourette syndrome is a neurological condition involving involuntary motor or vocal tics, but it is not mainly a theory-of-mind disorder.

Materials and procedure

The main task was the Eyes Task. Participants saw 25 black-and-white photographs showing only the eye region of different faces. For each photograph, they chose which of two words best described what the person was thinking or feeling.

For example, a participant might choose between words such as “concerned” and “unconcerned”. Each correct answer scored one point.

The researchers also used control checks, including:

• a gender recognition task, using the same eye photographs;
• Happé’s Strange Stories, an advanced theory-of-mind task using short stories;
• a basic emotion recognition task, using whole-face emotional expressions.

The diagram shows the logic of the study: different groups completed the Eyes Task, and control tasks helped rule out alternative explanations.

Results

The autism/Asperger group scored lower on the Eyes Task than all three control groups.

The autism/Asperger group did not show the same impairment on the control tasks. This suggests the problem was not simply poor vision, inability to identify gender from eyes, or inability to recognise basic emotions.

In the general adult control group, women tended to score higher than men on the Eyes Task, although the most important finding was the lower performance of the autism/Asperger group.

Conclusions

Baron-Cohen et al. concluded that adults with high-functioning autism or Asperger syndrome can have subtle theory-of-mind difficulties, even when they have normal intelligence and can pass other theory-of-mind tasks.

The study supports the idea that autism involves difficulty interpreting other people’s mental states, especially from complex social cues such as the eyes.

Evaluation: AO3

Strengths

A major strength is the use of control groups. The general adult controls showed typical performance, while the Tourette syndrome group acted as a clinical control. This improved internal validity because the researchers could argue that the lower score was specific to autism/Asperger syndrome rather than clinical status in general.

The study was also fairly standardised. Participants saw the same set of photographs and had the same forced-choice response format, making the task replicable.

Another strength is usefulness. The study helped develop a more sensitive measure of adult theory of mind and influenced later autism research, assessment tools and social-cognition interventions.

Weaknesses

A key weakness is sampling bias. The autism/Asperger group had only 16 participants, mostly male, and all were high-functioning adults. This limits generalisability to autistic children, autistic people with learning disabilities, women with autism, or people with different support needs.

The Eyes Task also has issues with ecological validity. Real social interaction involves moving faces, tone of voice, body language, context and conversation. Cropped, static eye photographs with only two word options are much more artificial.

There are also questions about construct validity. The task is meant to measure theory of mind, but performance could also depend on vocabulary, familiarity with mental-state words, test-taking strategies or cultural experience with facial expressions.

Reliability

The study has some reliability strengths because the materials and scoring were standardised. However, the original Eyes Task had only 25 items and two response options, meaning chance performance was relatively high. Later versions of the task were revised to improve measurement.

Ethics

Using BPS principles, the study is relatively ethical because it involved adult participants, no major deception, and low risk of harm. Researchers still needed informed consent, confidentiality, the right to withdraw and sensitive debriefing.

However, autism research is socially sensitive. If findings are phrased carelessly, they can reinforce stereotypes, such as the inaccurate idea that autistic people have no emotions or do not care about others.

Links to the key theme and area

Understanding disorders

This study helps us understand autism/Asperger syndrome by identifying a possible cognitive difference: difficulty reading complex mental states from subtle social cues.

Individual differences area

The individual differences area studies how people vary in behaviour, cognition and experience. Baron-Cohen et al. fits this area because it compares groups of adults who differ in diagnosis and social-cognitive performance.

It also raises classic debates:

• Reductionism: it explains autism partly through theory of mind, which may oversimplify a complex spectrum.
• Nature/nurture: it focuses on cognitive differences, but social experience and learning may also affect performance.
• Psychology as a science: it uses controlled tasks and quantitative scores, but measuring “mind-reading” is difficult.
• Usefulness: it has practical value for assessment and support, but must be applied sensitively.

Comparison with Freud (1909) Little Hans

Baron-Cohen et al. is paired with Freud (1909) Little Hans under the key theme Understanding disorders.

The strongest comparison is that both studies try to understand unusual or disordered behaviour, but they do so in very different ways. Freud gives a deep, detailed account of one child, whereas Baron-Cohen et al. uses a more scientific group comparison with standardised tasks.