Autistic voices on the Autism Quotient (AQ) test

Kajsa Igelström, PhD

If you're autistic or suspect you are, you've probably encountered the AQ test. It's widely publicized as an online tool to get a "score" reflecting autistic traits. It's not diagnostic – your score will not tell you whether or not you are autistic. On average, autistic people do get a higher score than non-autistic people, but 50 questions can never capture the complexity of the spectrum and some autistic people score pretty low. Many autistic people like the AQ, and some feel it helped them realize they were autistic and understand themselves better. These positive views aren't represented in this article, simply because we haven't received much positive feedback on the AQ in our studies. (I'd love to see comments below the post pointing out the positives to make it more balanced!)

We have included the AQ in many of our questionnaires because it provides a recognized measure that helps our studies "fit in" in the scientific world. It can also identify major problems, like when we did a pilot study that returned a completely unrealistic AQ score distribution. We didn't know which of the participants were actually autistic, but we didn't trust our recruitment success, so we discarded the data and changed to a more robust recruitment strategy. Even though the total score is in no way diagnostic, the test also returns some sub-scores, some of which can be relevant for correlation analyses. The AQ has limitations, but we know about them – and so do other scientists in general. 

I'm writing this post, because the AQ has triggered negative reactions from some of our participants. I won't write a research paper on it, because an acceptable scientific analysis would require a more diverse population and specific questions. But the reactions are illuminating and deserve to be heard. 



A logically oriented mind is more likely to see all the complexities. This the main topic that participants touch upon. Many AQ questions are context-dependent. While a non-autistic mind might be able to make a rough approximation and pick a response option, many autistic minds get stuck. The answer always seems to be "IT DEPENDS!". This can be especially upsetting when you know someone is attempting to probe autism – a huge part of your life and yourself.  

Here are some examples of context-dependent questions that causes problems for many of our participants: 

Q1. I prefer to do things with others rather than on my own. 

(Response options are Definitely Agree, Slightly Agree, Slightly Disagree, and Definitely Disagree)

In case you have a less literal interpretation of phrases, I'll break this question down for you. First, there is no modifier like "I often prefer..." – for a literal mind, this can be hard to get around. Second "prefer" – a preference for being alone may be caused by something other than a true preference (e.g. social interactions are difficult, but still desirable). Third, "things"! This encompasses EVERYTHING. Fourth, "others" – that encompasses EVERYONE. A detail-oriented mind may see all activities and all people, and get lost in a maze of complexities. I'll let the participants illustrate.  


"I would love to have lots of friends and be able to engage with people but I often feel trapped inside my own head and can't make a connection with others." 
"I would prefer to do things with others but it's easier to [do] them alone." 
"I want to be sociable but I don't know how to do it." 
"One of my biggest difficulties is realizing and remembering that others do not think like me, that they are often looking out for hidden deceptiveness and are deceptive and manipulative themselves and will /.../ believe that there are manipulative or insulting messages in perfectly innocent actions and words. They make life unnecessarily complicated and negative. /.../ I have really thinned out my group of friends in recent years. Most people are more trouble than they're worth and take up too much time and energy."
"Do I prefer to do things alone? It depends what the thing is! I prefer to have a crap alone. I prefer to have a conversation with other people. /.../" 
"I prefer being on my own than with other people, unless the other person is someone I'm very comfortable with /.../"
"I like talking to my sister and my partner, and when I had a few friends in school I liked talking to them, but it's really hard for me to find people who are easy to talk to and most people misunderstand me a lot and being around them makes me feel bad about myself and get tired."
"Weird survey. Too many variables, not exact enough. I enjoy SOME people, some social situations, but your questions group them all together."


Q13. I  would rather go to a library than a party.
Q24. I would rather go to the theatre than a museum.

These two questions are similar and rather easy to see the problem with. You are given a choice of two venues. The question requires you to make some type of generalization, say, by visualizing the library as a peaceful place where you're free to wander around, and a party as a sensory chaos with huge social demands, and – based on that – analyze which you prefer on average.

Autistic minds are often incredibly good at seeing all the details. A library can be boring if you aren't in the mood for reading or already have enough books at home. Perhaps you use e-books. Perhaps the lights in the local library are too bright. Perhaps you've been alone for too long and actually feel like a social occasion. But then how you define a party? There are so many kinds of parties... We can go on and on!

Autistic thinking styles commonly involve precision. The questions aren't precise, so how is it possible to give a precise answer? If you also feel that it's important that the person analyzing the responses actually understands you, what do you do? You're intelligent enough to guess that you'll get a higher "autism score" if you pick library and museum – but in many cases IT DEPENDS... This really isn't nit-picking! It just reflects a tendency for certain types of minds to see details and complexities. Again, let's listen to some reactions. 


"Which party? I hate weddings. But the parties my disabled friends throw are great."
"It would depend where the party is and who is going to be there. I don't always hate parties."
"/.../ I prefer going to a museum than a theatre but I do love to go to the theatre for a performance I expect to enjoy and which will absorb me enough that any problems with the environment or other people will not distract or disturb me /.../ [T]he museum would also have to be one I'll find sufficiently interesting! /.../"
"Which museum vs which theatre? I hate art museums but love history museums. I love musicals but hate Shakespeare."
"Very difficult to answer! Answers would often depend on circumstances. e.g. if i would prefer theater or museum depends on museum." 


Q40. When I was young, I used to enjoy playing games involving pretending with other children.

This one probes imagination (which some autistic people have lots of and some have less of). There can be many complications in answering the question. The quotes below show some participants reason around this. Again, the answer was "it depends...", and they felt conflicted in picking an answer because they saw and appreciated the complexities. 


"I wanted to be included and feel part of it, but often wasn't. I'm not sure I really knew what to do."

"I did enjoy pretending games but I liked to be in charge."

"Really enjoyed imaginary games as a child but would reenact scenes from books or films rather than inventing new things."
"My pretending games were often played with my childhood friends. I didn't play with new kids much. It was easy for me to play with my steady childhood friends because they understood me. I was afraid of strangers."
"I liked playing imaginary games with my brother and other kids where I was the oldest and could dictate the terms of the game."


Camouflaging complications

I cannot emphasize this topic enough: Many autistic individuals have learned to work around their weaknesses by compensation and camouflaging. In our AQ comments boxes, it comes up frequently and is often eloquently expressed. 

If you do a lot of camouflaging, some questions in the AQ require you to choose whether to respond in accordance with your outward appearance, or in accordance with how you'd intrinsically function if you were allowed to relax and be yourself. While some people don't mind this and simply answer according to their "natural" behavior despite being able to camouflage, this choice can be difficult for many, and also be very emotionally upsetting, as much of the feedback to us showed. (In addition, many people take the AQ before they have gained this type of insight.) 

Please listen to these women and transgender individuals. 


General comments about the AQ: 

"Scale should include how often one thinks about, practices, or puts in energy or effort to mask, or pass. THIS is very important. I pass well, but have schooled myself."
"/.../ When you've only realized you're autistic later in life and had to find so many ways to adapt socially - how can you answer these questions without denying part of yourself? Either your autistic self or the skills you have learned which have had a masking effect - enough so that you didn't know yourself. /.../"
"/.../ Lots of them are learned skills - my ability to read people is actually brilliant pattern recognition and is real as anything. I am easy to coerce though. The other main point is I spent my entire life habituating myself to be as flexible as possible. /.../"
"Some of the question asked don't take into account that I can do some of these things but only because I've practiced them over the years and that I sit and observe any situation first and only then, when I've worked out the pattern, can I participate, often by playing the right role."

"I have gotten very adept at passing as neurotypical in most social situations, but I know I put an inordinate amount of conscious effort into remembering the lines to "scripts" people expect, reading people's tone/facial expression and constructing appropriate responses. It's exhausting /.../. I compulsively censor myself and am terrified of making a social misstep; I am afraid people will not like me anymore /.../. I hide certain aspects of my personality and play up others to seem quirky, endearingly awkward, and relatable."

"I feel like I struggle to communicate my challenges because medical/psychological questionnaires seem too open ended and subjective. /.../ Women are expected to be friendly and engaging in conversations. One on one in a quiet clinical setting, I can appear normal. I have been so socially sanctioned for being different that I struggle to be myself in front of strangers. Then I am doubted when I talk about my many struggles. Growing up, other kids, especially girls, made it very clear that I wasn't acceptable, then adults saw me as mature because I was smart and could hold mature conversations. My inner screaming was never heard. /.../"
"There are a number of things this questionnaire asks about which I have learned to be better at over time, that do not come naturally to me but I have had many years to rehearse them and find them less difficult, though often still a strain, e.g. meeting new people (rehearsed approaches and masking), recognizing when someone is bored by my conversation (hypervigilance)."


Some specific examples: 

Q7. Other people frequently tell me that what I’ve said is impolite, even though I think it is polite.
Q39. People often tell me that I keep going on and on about the same thing.

"/.../I've learned how not to be offensive/impolite by counting in my head to let enough time pass before I speak. /.../ So by employing my coping mechanism, people end up not telling me I'm impolite so often. Thus the response to that question would be that no, people don't tell me I'm impolite but that's only because I intervene. It takes a ton of work and that's what should be measured here, not the end result."

"Some of these would be masked over. /.../ going on and on about a subject; I did when younger, but the response from peers was that this was wrong and I was terrible to be around. I am afraid to speak about my own interests now."


Q38. I am good at social chit-chat.
Q48. I am a good diplomat.

"Some things have changed over time. I learned how to make small talk at work. /.../ You (try to) adapt."
"Some of these answers have changed over time as I have developed coping strategies that make it now easier to do certain things. But I am always aware that my way of approaching things is not the same as other people's. So for example I now CAN make small talk but I'm not generating it spontaneously, I'm going through a script in my head that I've memorized."
"I've gotten lots better at social chit chat /.../ since learning I'm autistic, because I can look at these things as set, finite skills to intentionally learn rather than just things people either can or can't do."
"/.../ [I]f I follow a script I can manage it but naturally I'm very quick to anger and highly opinionated. So....I'm not a natural diplomat but I can keep up an act short term if I follow a script."


Q18. When I talk, it isn’t always easy for others to get a word in edgeways.

"/.../ I limit my speaking time on a topic to a rough number of sentences per turn so people don't tell me I go on and on. It's not that I don't WANT to give all I know to people who don't know but I'm also aware that not only do they not want it but I've learned that my knowing more makes them feel bad about themselves somehow. So I modify my behavior to limit unpleasant responses from other people because that's stressful too... but I'm still me, inside. /.../"


Take-home messageS

for autistic people

If you struggle with the AQ, you're not alone! Your mind isn't necessarily compatible with this type of questions. It was designed before we knew much about all the different manifestations of autism, and it has been overvalued as a clinical tool on websites or by people who didn't know better. The creators never meant for that to happen. If you see it in any of our studies, don't worry about having your answers over-interpreted, or that we're going to think you're any less autistic because you love reading fiction or hate numbers. And after finding out how negative many of you feel about it, we'll think carefully about whether or not we need to use it. 

For Scientists

Let't try to be aware of how difficult and frustrating this questionnaire can be for autistic people. It can trigger feelings of having the autism diagnosis tested or questioned (a diagnosis that might have taken decades to obtain), and it can trigger anger because it seems to probe for stereotypical views of autism (e.g. a fascination by dates or numbers). Given the emotional triggers, the forced-choice format and requirement for answering all 50 question can make it very confrontational, painful and tiring. 

For the future

It's important that questionnaires meant for autistic people are designed in a way that's compatible with a wide variety of autistic "thinking styles". Ideally, scientists and autistic people could collaborate when designing questions and response options. Autistic people are great at discovering problems with questions, because they often have a low threshold for reacting to a lack of clarity. And in our experience, putting some extra thought into wording actually benefits everyone.

Footnote: We've changed spelling to US English to make it less obvious where participants are from, and corrected some obvious typos.