News The latest developments in autism research.
Profiles Portraits of scientists who are making a mark on autism research.
Toolbox Emerging tools and techniques that may advance autism research.
Spotted A roundup of autism papers and media mentions you may have missed.
Opinion Conversations on the science of autism research.
Viewpoint Expert opinions on trends and controversies in autism research.
Columnists Dispatches from experts on various facets of autism.
Crosstalk Debates and conversations about timely topics in autism.
Reviews Exploring the intersection of autism and the arts.
Q&A Conversations with experts about noteworthy topics in autism.
Deep Dive In-depth analysis of important topics in autism.
Special Reports Curated collections of articles on special topics in autism.
Webinars Presentations by leading experts on their latest research.
News / Toolbox

New test captures subtle social difficulties in adults with autism

by  /  5 May 2017
Reading minds: A video-based quiz gauges a person’s ability to determine what someone else is thinking.

Thomas Barwick / Getty Images

A new test can assess theory of mind — the ability to understand others’ mental states — in adults with autism. In the test, people with the condition interpret scenes in a video for white lies, jokes and irony.

Many people with autism have difficulty grasping what others think or feel. But most tests of this ability are designed for children.

The new test is based on the Strange Stories Test, a collection of vignettes covering 12 everyday scenarios in which people tell jokes or say things they do not mean.

The Strange Stories Film Task, described 11 March in Autism Research, recreates more realistic versions of the same 12 scenarios in a video1. It allows researchers to assess how well an individual can tap into social cues, such as facial expression and tone of voice, to understand what another person is thinking.

In one clip of the new task, characters named Max and Alice are sitting in a living room across from each other. Alice is holding a guitar and looks nervous. “I’ve been working on this for ages, and I think I have finally got it. I think my song’s going to end like this,” she says as she starts to play badly and sing out of tune.

Max, nodding and smiling, says, “Well done, Alice … that sounds really good.”

Viewers watch the clip and answer questions such as “Why did Max say that?” and “If you were in Alice’s situation, what would you say next?”

Movie time:

Researchers gave the task to 20 adults with autism and 20 controls with an average age of 30. Scores ranged from 0 to 2 based on the responses to each scenario. For example, participants who said Max did not want to hurt Alice’s feelings earned a score of 2. Those who responded that Max is nice got a 1. Those who answered incorrectly received a 0.

Individuals with autism were worse than controls at guessing what the characters were thinking. They scored an average of 15.5 out of 24, compared with the controls’ 18.8.

The participants also took a series of standard tests of social cognition and emotion recognition, including the Reading the Mind in the Eyes Test — which gauges an individual’s ability to detect other people’s emotions by looking at their eyes. The Strange Stories Film Task was better able to distinguish between the autism group and the control group than the standard tests were.

The researchers plan to make the test freely available online. In the meantime, they say, they will send it to anyone who asks for it.

  1. Murray K. et al. Autism Res. Epub ahead of print (2017) PubMed
  • Recall

    So when the test is given to autisic people, are they telling them that they should give the answer they think a non-autistic person would give?

    If you ask me questions like that, I assume that you what to know what I think, not what I think someone else might think.

    • Zoran Bekric

      I have to agree with this. The problem with most of these ToM tests is they assume that whatever answer the non-autistic participants give must be the right answer, without establishing that it is right.

      The Reading the Mind in the Eyes Test, for example, takes what emotion the majority of NT participants say they see in the eyes as the correct answer, not what emotion the people whose eyes were being photographed report they were feeling at the time.

      Similarly, the Firth-Happé animations (from the 2000 paper “Do triangles play tricks? Attributions of mental states to animated shapes in normal and abnormal development”), the “right” answer was the one that matched the researchers’ interpretation of the animation (the two triangles were mother and child), with other interpretations of the animation deemed “wrong” or “inappropriate”.

      And now we get this. In the example described featuring Max and Alice, I can think of at least half a dozen possible explanations:

      * Max is in love with Alice and is sufficiently dazzled by the halo effect that he honestly believes the playing is good (“love is blind” and all that);
      * Max is in love with Alice and is trying to be supportive;
      * Max desires Alice and is trying to ingratiate himself by appearing supportive;
      * Alice responds badly to criticism and Max wishes to avoid a fight;
      * Max is manipulating Alice, gaslighting her into thinking her playing is better than it is, so when she presents her song to a wider audience, she will be publicly humiliated;
      * Alice is a songwriter, not a performer, and knowing this, Max is ignoring the performance in favour of focusing on the lyrics and the tune, imagining how it will sound when done by a more skilled performer, so his comment is completely accurate.

      And that’s without
      * Max doesn’t want to hurt Alice’s feelings;
      * Max is nice.

      How to determine which of these interpretations is correct (if, indeed, any of them are) would be by observing further interaction between Max and Alice. That would eliminate some interpretations, support others, and perhaps open up new possibilities.

      What Alice would say next would be part of the information gathering process. So what answer the participants go with would depend on which possible interpretation they chose to go with.

      How the two actors deliver the lines would have a big influence as well. A good pair of actors could convey any of the above interpretations through their tone, body language, and delivery. It’s a standard acting exercise to give actors such simple lines and to ask them to come up with several different readings.

      Given all that, deciding that “Max doesn’t want to hurt Alice’s feelings” is the correct interpretation comes across as stunningly naive. If that’s the interpretation the majority of controls come up with, maybe the test just demonstrates the lack of (social) imagination in the controls.

      I must admit that if I were asked (bearing in mind I’m autistic), I would say “Max is nice” since that’s the answer that least commits me to any specific interpretation and leaves the most possibilities open for additional information about Max and Alice’s relationship.

  • Kelvin

    Can you use identity first terms, please? I’m autistic, not with autism. I understand there’s the attempt at being politically correct, but at least mix up the terms to provide some recognition of those that prefer identity based terminology, rather than person first.

  • Otilia Netedu

    Hello, how can I receive the test?


Log in to your Spectrum Wiki account

Email Address:



Request your Spectrum Wiki account

Spectrum Wiki is a community of researchers affiliated with an academic or research institutions. To be considered for participation, please fill out this form and a member of our team will respond to your request.


Email Address:

Title and Lab:

Area of Expertise: