Viewpoint Expert opinions on trends and controversies in autism research.
men trying to make eye contact
Illustration by Julien Pacaud
Opinion / Viewpoint

Eye contact is aversive for some adults with autism

by ,  /  16 May 2017
The Experts:

Craig Erickson

Associate professor, Cincinnati Children’s Hospital Medical Center

Rebecca Shaffer

Assistant professor, Cincinnati Children’s Hospital Medical Center

One clue that a child may have autism is that she does not make eye contact with others. This feature appears in the first six months of life, leading some researchers to consider differences in gaze pattern a potential early marker for autism.

One theory holds that people with autism perceive eye contact during social interactions as unimportant: In other words, they are indifferent to it. Alternatively, they may avoid eye contact because it is uncomfortable or aversive.

Many autism therapies encourage children and adults to make eye contact. To determine whether this is the right approach, it is important to understand whether clinicians are teaching people with autism to pay attention to something that doesn’t interest them or forcing them to do something that makes them uncomfortable.

Studies in young children support the first hypothesis of eye contact: an indifference to gaze. Toddlers with autism spend less time looking at the eyes of an actor in a video than do typical children or those with developmental delay. But the children with autism do not actively shift their gaze away from the actor’s eyes or resist looking at the actor’s eyes when prompted to do so1.

Anecdotes from teens and adults with autism paint a more complex picture. These individuals say they do not understand the need to make eye contact — or that eye contact is unpleasant.

We have studied this topic for six years. Based on our research and clinical experience, we believe that these findings are not contradictory. An early lack of interest in eye contact may cause children with autism to miss out on social cues, leading to low social motivation and interest down the road. Feeling obligated to make eye contact when you are not motivated to do so is unpleasant, and this may cause some adults with autism to actively avoid eye contact.

Learned behavior:

A 2010 study supports our theory. The researchers measured gaze in adults with and without autism as they viewed faces with happy, fearful and neutral expressions. Between the face presentations, a cross appeared on the screen in different positions to direct the participants to focus on a particular area.

The researchers found that people in the autism group showed both a decreased preference for looking at eyes and active avoidance of the eyes when the cross cued them to look at the eyes2. But there was variability across the group: Individuals with autism who spent more time looking at the eye region of faces performed better on emotion-recognition tasks than did those who focused on other facial features. They also have better social skills as measured by the Autism Diagnostic Interview-Revised.

This finding jibes with informal observations in our clinic that individuals with autism who are interested in being social, and able to engage in social give-and-take, tend to make eye contact.

We have used eye-tracking technology to examine whether adults with autism have low social motivation, which could contribute to both indifference and an aversion to eye contact3. We looked at 58 people with autism, 37 with developmental delay and 66 controls, all between the ages of 2 and 35. (The data from 37 of the individuals with autism and 26 of the typically developing individuals aged 5 to 17 is published; that from the younger and older individuals is still unpublished.)

We asked the participants to look at a screen showing two videos simultaneously: One depicted social scenes — children playing — and the other showed non-social scenes consisting of moving objects. We found that people with autism spent less time looking at the social videos than the other participants did.

Looking forward:

Our findings suggest that individuals with autism are less interested in social material than either controls or people with developmental delay, regardless of age.

We believe our results reflect low social motivation, which might contribute to eye contact being uncomfortable in adulthood. We plan to explore this possible sequence of events by combining measures of social interest, such as our paradigm, with measures of gaze to the eye region of faces.

It is possible that the experience of eye contact varies across the autism population. Autism is heterogeneous, so some individuals with autism may be indifferent to eye contact, whereas others may experience it as unpleasant.

Showing that indifference to eye contact in toddlerhood gives way to both indifference and discomfort in adulthood requires longitudinal studies. We especially need studies that explore the experiences of adolescents transitioning into adulthood, and those of older adults. Adolescents face unique social challenges as they enter post-secondary education and gear up for job interviews.

Craig Erickson is associate professor of psychiatry at the hospital. Rebecca Shaffer is assistant professor of pediatrics at Cincinnati Children’s Hospital Medical Center. 

  1. Moriuchi J.M. et al. Am. J. Psychiatry 174, 26-35 (2017) PubMed
  2. Kliemann D. et al. J. Neurosci. 30, 12281-12287 (2010) PubMed
  3. Shaffer R.C. et al. J. Autism Dev. Disord. 47, 506-513 (2017) PubMed

6 responses to “Eye contact is aversive for some adults with autism”

  1. p_seven says:

    I’m a 59 yo Aspergers woman. Why do I avert my eyes when talking to someone? Because it makes it much easier for me to comprehend what the other person is saying to me and to follow along with the flow of the conversation.

    If I make eye contact during a conversation, I have to put a great deal of energy into concentrating on the conversation. So much so that I can lose chunks of the conversation, and also lose track of “It’s your turn to speak … It’s now my turn to speak” (this doesn’t come naturally to me), and in general come out of the conversation not understanding what was said.

    Perhaps this has to do with social phobia, in that looking into someone’s eyes causes me anxiety, which consequently causes my inability to concentrate on what’s being said. But personally I don’t believe that to be the case, since I know that it’s much easier to follow a conversation with my husband if I avert my eyes. Definitely no social phobia there. I think my brain’s just wired differently.

    • C Stargazer says:

      I agree with p_seven and what Planet Autism said.

      Personally I think it’s maintaining concentration that causes anxiety, not the act of making eye contact. Eye contact is distracting at the very least and can contribute to being overwhelmed by sensory.

      Consider Neuroscience studies that show our senses do not have the same dampening as NT brains. Also consider a recent study on babies that showed they experiencing the same sound differently every time. NT brains can “get used to it” due to dampening, they also won’t measure additional details because of that dampening, at least not without focusing on noticing other details; it’s the opposite for me. I’m going to notice all those tiny patterns in the eyes, the specs, veins, etc without trying. If I’m already overwhelmed by other factors, it will become pain. There are times that if seated and needing to concentrate on listening to what is being said; I will close my eyes to completely shut off that sensory input.

      Children generally are not carrying on complex conversations. I can look at eyes on TV or in pictures because that is not interaction. I’m able to put all my concentration into watching TV while comfortably sitting down, but if there is even the slightest distraction it will disrupt me greatly. NTs can sit there and carry on a conversation while watching a show, I can not do both. If I’m talking to someone and there’s a TV on in the background, I won’t be able to listen to the person without muting that TV. Same goes for listening on a phone.

      The way our minds breakdown and process language is also a major factor, including when it’s “our turn” to communicate especially when our minds are wired to think with our senses more.

      Also consider how exhausting it is for Parkinson’s patient to walk. My ASD father recently developed Parkinson’s and found that he was putting in the effort to take many steps in order to take just one step. It’s the same with being exhausted and overloaded from processing too much sensory for a long time period, social concentration will exhaust you even faster in those circumstances. Not making eye contact is more efficient use of our resources.

      for reference, Autistic 36 yr old, diagnosed before the Asperger’s diagnosis was added to the DSM.

      We live in an era where more and more people are using text to communicate, I certainly find it more comfortable; with room for articulation that does not exist in back and fourth speaking. You could be communicating with a deaf, blind, non-speaking or paralyzed person with text and not even realize it. Feels more like equality to me.

  2. Planet Autism says:

    I have Asperger’s and I find eye contact uncomfortable. I can look at someone when they are talking but when it’s my turn to talk I need to spend most of the time looking away to focus on what I need to say. It is sensory input to look at someone’s eyes and it’s overload to look at them and think at the same time. It’s also too intrusive to look at someone’s eyes, it feels like they can see inside your brain and it’s too personal. Agree with what p_seven says below.

    • merchantfan says:

      I feel sometimes like I have a meter of how long I can make eye contact during the day before it becomes painful. It’s also easier with some people or with kids. Looking at eyes in a picture isn’t painful. I don’t think studies with videos are going to quite match the real thing.

      • Planet Autism says:

        I know what you mean! Some days are worse than others. On an autism forum once people discussed it and some said things like it’s an ’emotional rape’ to have to meet eyes with other people. I understand that too. Yes painted eyes aren’t real so they aren’t the same, although that weird thing about how the eyes can appear to follow you round a room when you move is freaky. I agree, there will be some collective similarities but it’s that thing where every autistic is as different as everyone who’se NT and you cannot have a blanket cause.

  3. Sharwan Kumar says:

    Please try to analyze your work in view of these latest conclusions:
    Eye contact with your baby helps synchronize your brainwaves
    November 29, 2017
    University of Cambridge
    Making eye contact with an infant makes adults’ and babies’ brainwaves ‘get in sync’ with each other — which is likely to support communication and learning.


    This is Lucy Kivlin and her baby Ginny.
    Credit: University of Cambridge
    When a parent and infant interact, various aspects of their behaviour can synchronise, including their gaze, emotions and heartrate, but little is known about whether their brain activity also synchronises — and what the consequences of this might be.


    Brainwaves reflect the group-level activity of millions of neurons and are involved in information transfer between brain regions. Previous studies have shown that when two adults are talking to each other, communication is more successful if their brainwaves are in synchrony.

    Researchers at the Baby-LINC Lab at the University of Cambridge carried out a study to explore whether infants can synchronise their brainwaves to adults too — and whether eye contact might influence this. Their results are published in the Proceedings of National Academy of Sciences (PNAS).

    The team examined the brainwave patterns of 36 infants (17 in the first experiment and 19 in the second) using electroencephalography (EEG), which measures patterns of brain electrical activity via electrodes in a skull cap worn by the participants. They compared the infants’ brain activity to that of the adult who was singing nursery rhymes to the infant.

    In the first of two experiments, the infant watched a video of an adult as she sang nursery rhymes. First, the adult — whose brainwave patterns had already been recorded — was looking directly at the infant. Then, she turned her head to avert her gaze, while still singing nursery rhymes. Finally, she turned her head away, but her eyes looked directly back at the infant.

    As anticipated, the researchers found that infants’ brainwaves were more synchronised to the adults’ when the adult’s gaze met the infant’s, as compared to when her gaze was averted Interestingly, the greatest synchronising effect occurred when the adults’ head was turned away but her eyes still looked directly at the infant. The researchers say this may be because such a gaze appears highly deliberate, and so provides a stronger signal to the infant that the adult intends to communicate with her.


    In the second experiment, a real adult replaced the video. She only looked either directly at the infant or averted her gaze while singing nursery rhymes. This time, however, her brainwaves could be monitored live to see whether her brainwave patterns were being influenced by the infant’s as well as the other way round.

    This time, both infants and adults became more synchronised to each other’s brain activity when mutual eye contact was established. This occurred even though the adult could see the infant at all times, and infants were equally interested in looking at the adult even when she looked away. The researchers say that this shows that brainwave synchronisation isn’t just due to seeing a face or finding something interesting, but about sharing an intention to communicate.

    To measure infants’ intention to communicate, the researcher measured how many ‘vocalisations’ infants made to the experimenter. As predicted, infants made a greater effort to communicate, making more ‘vocalisations’, when the adult made direct eye contact — and individual infants who made longer vocalisations also had higher brainwave synchrony with the adult.

    Dr Victoria Leong, lead author on the study said: “When the adult and infant are looking at each other, they are signalling their availability and intention to communicate with each other. We found that both adult and infant brains respond to a gaze signal by becoming more in sync with their partner. This mechanism could prepare parents and babies to communicate, by synchronising when to speak and when to listen, which would also make learning more effective.”

    Dr Sam Wass, last author on the study, said: “We don’t know what it is, yet, that causes this synchronous brain activity. We’re certainly not claiming to have discovered telepathy! In this study, we were looking at whether infants can synchronise their brains to someone else, just as adults can. And we were also trying to figure out what gives rise to the synchrony.

    “Our findings suggested eye gaze and vocalisations may both, somehow, play a role. But the brain synchrony we were observing was at such high time-scales — of three to nine oscillations per second — that we still need to figure out how exactly eye gaze and vocalisations create it.”
    I look forward to your kind reply. Thanks.

Leave a Reply

Your email address will not be published. Required fields are marked *