Skip to main content

Autism and looking preferences: The ethics of pre-symptomatic detection

As I have written before, researchers at the Marcus Autism Center are working with eye tracking technologies to identify Autism Spectrum Disorder (ASD) in young children and infants. As Katie Strong described in this blog post, a recent article in Nature, titled “Attention to eyes is present but in decline in 2-6-month-old infants later diagnosed with autism,”[1] presented the Marcus team’s most recent findings related to the early identification of ASD traits. They argue that, although there are many different ‘autisms’ with many likely causal pathways, the developmental pathway to ASD is similar. This work is an effort to capture this pathway by focusing on differences in early looking patterns. In this article, they “propose that in infants later diagnosed with ASD, preferential attention to others’ eyes might be diminished from birth onwards”(p. 427). After a brief refresher on the article’s findings and background, I will provide a deeper discussion on the neuroethical concerns. 

This hypothesis was built on previous research showing differences in children’s and adults’ preferential looking behaviors during social situations. As Ms. Strong described in her blog, an earlier study found differences between where adults diagnosed with autism and adults without autism looked during socially or emotionally salient scenes from the 1966 film, “Who’s Afraid of Virginia Woolf?”[2] A more recent article, found that two-year-olds with a diagnosis of autism preferred to look at mouths, rather than eyes, and objects, rather than bodies.[3] From this work, the authors formulated a hypothesis that these differences were present at birth.

Some of the eye tracking results from the 2002 study using scenes from “Who’s Afraid of Virginia Woolf?” The red lines show where the autistic adult looked and the yellow lines show where the non-autistic adult looked during this scene.

The current study, however, both supported and contradicted this hypothesis. Children recruited at birth had their eyes tracked while watching social videos a total of nine times between two and eighteen months of age. Of a total of 110 infants included (59 having an older sibling with autism and so at ‘high-risk’ and 51 with no 1st, 2nd, or 3rd degree relatives with autism and so at ‘low-risk’), the authors did a deep analysis of the eye-tracking results of the eleven males diagnosed with autism at 36 months. Compared to the children who did not obtain a diagnosis, these children showed different looking preferences. Overall, the infants who were later diagnosed with autism preferred to look at mouths versus eyes and bodies versus objects. 

Thus, the hypothesis was confirmed as far as showing that looking differences were present early in life. However, the differences did not appear until around 6 months of age, becoming a trend at 9 months of age. The authors conclude the findings suggest that the variety of genetic and possible environmental factors that lead to a person having autistic characteristics all present with a similar developmental path early in life. Knowing that the differences in looking patterns are not present at birth, they argue, “suggests a neural foundation that may be built upon” (p. 430) for treatment. In other words, the authors suggests that if differences in looking patterns were identified in infancy then one could assume that a later diagnosis of ASD is likely and, potentially, interventions could be implemented to avoid that later diagnosis. 
This is a finding to which neurodiversity advocates object. As I have described elsewhere, neurodiversity is the concept that autism and other neurologically-based disabilities are part of the natural variation of human neurological development. This variation is necessary for human diversity and so should be neither cured nor normalized. There is a spectrum of this opinion but most adherents recognize that people with significant autistic impairment should get some sort of aid to communicate needs and achieve a higher quality of life; however, neurodiversity advocates draw the line when efforts are made to make someone’s autism diminish to the point of being unrecognizable or eliminated. 

As a neurodiversity advocate who has also been deeply involved with families struggling to find help with their significantly autistic children, I am ambivalent about these findings. I recognize that some sort of skill development is incredibly useful for many children and families. This is particularly true for children who are engaging in frequent self-harm or harming other people. I do recognize, however, that there is a fine line between this level of intervention and normalization attempts. I appreciate my autistic friends and the variety autism brings to the world. I also appreciate that people can be happy with being different. Eradicating this kind of diversity could further narrow our community’s definitions of normality and, thus, diminish acceptance for diversity. It is for this reason that neurodiversity adherents are skeptical of research aimed at intervening with the development of autism. 
Beyond the concerns of the neurodiverse community, other ethical concerns include how to accurately communicate this information to parents and the high cost of using this technology. The concern over the lack of available treatment for children showing early traits of ASD, however, spurred the most conversation. The central question being this: if we are able to detect the possibility of ASD in infancy, what will we tell parents to do? Although you can work on pre-linguistic skills in young children, there are no real interventions for that age group. While we cannot discount the possibility of the development of reliable, evidenced-based treatments, there is also a strong possibility that individuals and companies looking to profit from these findings will promote ineffective and possibly harmful treatments. The history of autism is saturated with unfounded, often dangerous treatments from holding therapy to chelation to the latest: bleach.[5] With the possibility of identifying signs of autism in infants come the possibility of unfounded infant-directed treatments.
The most likely outcome will be that the clinical world will have to grapple with the same questions the DSM 5 creators encountered when considering the inclusion of a diagnosis for risk of psychosis. Psychosis risk syndrome would have been made during the teen years and been used to identified a person as being at ‘risk’ for developing schizophrenia. Discussions about this proposed diagnosis—which was eventually rejected—brought up a host of ethical issues beyond the question of available treatments. These issues included inevitable stigma towards diagnosed individuals and the likelihood that big pharmaceutical companies would use the inclusion of this diagnosis as an opportunity to generate and market a host of unnecessary or possibly dangerous drugs.[6] There are also issues related to inevitable false positives. With both psychosis risk syndrome and the early identification of ASD traits there will be people identified who will never develop the actual syndrome. Although psychosis risk syndrome ended up being excluded from the DSM 5, early detection technologies are continually being developed and so the concerns are timely and paramount. While it could be argued that any behaviorally or socially based interventions for early ASD identification is not harmful and even potentially beneficial to children not diagnosed with ASD, these interventions are, nevertheless, costly and time consuming. 
The eye tracking technology.
It is critical for clinicians, professionals, families, advocates, and autistic individuals to begin thinking about and discussing issues related to preclinical detection. Earlier diagnosis, and even pre-natal screenings, are a close reality. 


  1. Jones, W. & Klin, A. (2013). Attention to eyes is present but in decline in 2-6-month-old infants later diagnosed with autism. Nature, 504: 427-433.
  2. Klin, A, Jones, W, Schultz, R, Volkmar, F, & Cohen, D. (2002). Visual fixation patterns during viewing of naturalistic social situations as predictors of social competence in individuals with autism. Arch Gen Psychiatry, 59(9): 809-816.
  3. Jones, W., Carr, K. & Klin, A. (2008) Absence of Preferential Looking to the Eyes of Approaching Adults Predicts Level of Social Disability in 2-Year-Old Toddlers. With Autism Spectrum Disorder. Arch Gen Psychiatry, 65(8):946-954.
  4. Only two girls were diagnosed with autism. Because of the low number they were excluded from analysis.
  5. Bleach is being marketed as Miracle Mineral Solution (MMS). The creators of MMS claim it will cure everything from hepatitis to malaria to cancer to herpes. To find out more about the history of false cures for autism, see Offit, P. (2008). Autism’s False Profits: Bad Science, Risky Medicine, and the Search for a Cure. New York: Columbia University Press. 
  6. Corcoran, C., First, M., & Cornblatt, B. (2010). The psychosis risk syndrome and its proposed inclusion in the DSM-V: A risk-benefit analysis. Schizophrenia Research, 120(1): 16-22. 

Want to cite this post?

Sarrett, J. (2014). Autism and looking preferences: The ethics of pre-symptomatic detection. The Neuroethics Blog. Retrieved on , from


Emory Neuroethics on Facebook