Two Steps Forward, One Step Back

Researchers are starting to realize that “regressive autism” is a myth based on poor methodology. The story tells us about the failings of autism research — and the complexity of human development.

David Dobbs reports in Spectrum [1] that since “autism” was defined and named, professionals have believed that it came in two forms: one present from birth and one defined by “regression.” In this cultural myth, a child seemed to be developing completely normally and then suddenly dramatically lost many social and communication skills. Even such respected researchers as Sally Ozonoff began their careers believing this highly improbable developmental pattern existed.

The Cultural Myth of Regression

This idea of regression has been responsible for a lot of mischief.

In Kanner’s day, the idea that “refrigerator mothers,” or cold, unaffectionate parents, caused autism, was invented to explain regression.

The idea that vaccines cause regression has reinforced the anti-vaccine movement, leading to recent measles and whooping cough outbreaks.

Depictions of regression have been used by Autism Speaks and the media to portray autistic children as “changelings” who replaced the “real child” supposedly hidden “underneath” the autism. Taken to its logical conclusion, this idea led a mother to smother her three year old daughter. Asked who she thought she was killing, the mother replied, “Autism.”

In less extreme cases, parents who believe their child is not present will talk cruelly about him as if he were not there. Those who believe she is in her own world will ignore any attempts she makes at communication — which will stunt any further communication development.

The Reality of Regression

Yet, when we actually measure autistic children’s language and social development, we see a different reality. At best, the regression idea is an oversimplification. At worst, it has no basis at all.

First, there is no consistent data on regression. There’s not even a consistent definition.

In fact, after 70 years of autism research, there is still no clear definition of what regression is. Psychologists Brian Barger and Jonathan Campbell have wrestled with this problem energetically over the past few years, combing through more than 100 studies. They have concluded that the literature on regression is “without a central conception” and has “no universally agreed-upon central definition.”
In one meta-analysis of 85 papers, the pair uncovered a hodgepodge of definitions for virtually every type of regression commonly described in autism research — among them, language regression, social regression, motor regression, ‘mixed’ regression, ‘regression, developmental,’ just plain regression and even ‘regression, unspecified.’ In the case of language regression, for instance, they found no agreement on how many words a child must have had and then lost, or how long she needed to have used them, to qualify as having regressed.
… This definitional disaster, Barger and Campbell wrote in a paper last year, has created a “literature marked by conflicting results.”

After 70 years of autism research, there is still no clear definition of what regression is.

Rather than a rare subtype of autism, some degree of “regression” may be part of most autistic children’s development.

“I think most kids with autism lose some skills, but how many they lose — and when they lose and what they lose — varies across kids,” says Catherine Lord.
“Trying to separate the children who regress from those who don’t can be like drawing a line in the sand” … Katarzyna Chawarska, a researcher at the Yale Child Study Center, said.

There also seem to be many ways that autism first becomes visible, which don’t fit neatly into a “regression” vs. “no regression” binary.

Graph showing trajectories of four developmental patterns from birth to age 6: normal range, slow but steady, plateau, and regression. Note that regression is gradual — not a sudden, dramatic drop as described in the cultural myth of autism. Image from an article in Pediatrics, Developmental problems and the child with special needs.

Second, the loss of skills isn’t actually sudden, and development before the loss wasn’t actually typical.

Researchers noticed that:

the more they scrutinized home videos or other contemporaneous records, such as clinician or caretaker reports from the first year of life in children ultimately diagnosed with regressive autism, the more they saw early signs of the condition.
The strength of these early signs varies, and they’re often subtle, but they show up in multiple domains, from movement and eye-gaze patterns to language responses and social interactions.

In other words, children who seemed to suddenly “regress” already had atypical sensory and motor development that others missed.

Finally, loss of skills occurs (less often) in typical development, too.

A team led by Brignell and her colleague Angela Morgan, for instance, found that about one in seven typically developing toddlers at some point regresses in eye gaze and in showing emotions, compared with one in four children who have autism. A key difference, however, is that the children with autism are less likely than the typical children to make up for those losses.
The researchers found that 26 percent of typically developing toddlers may also regress in other areas, such as eye gaze, social interest, or smiling or head movement in response to social stimulus, compared with 41 percent of children on the spectrum.
“Loss,” Brignell says, “is not unique to autism spectrum disorder.”

The Methodology Behind the Myth

Besides an overreliance on clinical intuition, the myth of regression seems to stem from bad methodology.

Clinicians relied on studying developmental trajectories after the fact, even though researchers in other areas of psychology are well aware of the inaccuracy of human memory:

The final but central problem in the regression literature is its heavy reliance on retrospective studies — those that depend on reconstructing events, rather than observing them as they occur.
… It’s only during the past decade or so that Ozonoff and others have used prospective studies or home videos to check parent recall. Their work has shown that parents’ memories, like most human memories, can be astoundingly unreliable. People tend to misremember things such that they fit their current impressions or beliefs — a form of confirmation bias…
This illusion may be further magnified by something called the telescoping effect: The more time that passes after a significant event occurs, the more likely a person is, in remembering that event, to move it forward in time. … telescoping encourages a false perception of regression.”

So, a gradual plateau or decline, or even a “two steps forward one step back” situation, might be remembered as a recent, sudden drop.

What the Regression Story Tells Us About Autism Professionals

This isn’t the first time autism researchers and clinicians have made assumptions, supported by faulty methodology and bias, that turned out to be wrong.

Autism researchers and clinicians made a similar error with echolalia (apparently out-of-context quotations instead of speaking in one’s own words) and pronoun reversals (switching pronouns such as “me” and “you”). These were asserted to be characteristic of autism. In all these cases, clinicians:

  1. Observed a new, unusual behavior;
  2. Came up with theories to explain it, based on their personal interests and professional background (for example, immediately interpreting autism through a psychoanalytic framework[2]).
  3. Pronounced the behavior to be universal in autism (without testing this hypothesis);
  4. Pronounced it to be unique to autism (without testing whether other groups behave the same way);
  5. Later started measuring the behavior;
  6. Discovered the behavior was neither universal in autism nor unique to autism;
  7. Failed to publicly acknowledge and apologize for their mistake; actively retract and re-educate the public; or change the poor research practices that caused the mistake in the first place.
  8. Instead, researchers will quietly stop talking about it as if they hadn’t based all their work on it. (For example, in the mid-2000’s, Theory of Mind and Extreme Male Brain Theory, which had been mentioned in every autism paper no matter how tangentially relevant, suddenly disappeared from papers, without comment).

It’s telling that, although evidence against regression has been building in the scientific community for the past 15 years, the theory is only just now being rejected in academic conferences — rather than in communication with non-academic audiences.

Watch for your favorite (or least favorite) assumption about autism to be debunked, and then, without explanation, disappear from the public conversation. “All this has happened before, and all shall happen again.”

What the Regression Story Tells Us About Development

The Spectrum article hints that regression may be part of the normal developmental process, but fails to consider the fascinating, revolutionary implications of this idea.

“[Regression] might be part of a larger, normal development process. Maybe it’s not specific to autism; maybe there are more kids that go through losses and delays and spurts”… says Campbell, professor of psychology at the University of Kentucky.

So, if we take this idea seriously, here are a few implications:

  • Regression — or temporary loss of skills — is a normal part of development that should be expected, not a cause for alarm.
  • Development happens faster when a child’s environment is happy, stable, and nurturing, and it appears to slow or reverse when the environment becomes less optimal, more stressful,or gives the child fewer opportunities to use their skills.
  • The main difference between neurodivergent and typically developing kids is how stressful their environment needs to be before they regress (neurodivergent kids need a more supportive environment. They may be more sensitive to their environments, in general).
  • The idea of development as a continuously increasing line is as ridiculous as the idea of the stock-market as a line always going up. Kids can go up and down at different times and in different amounts, and end up in a similar place.
  • Regression won’t ruin a child’s life forever, but may signal that the child might need less stress or extra help right now.
  • There isn’t one single right way to develop.

There isn’t one single right way to develop.

Further Reading

For detailed discussion of the mistakes researchers made about echolalia and pronoun reversals, see the paper I co-authored with Morton Ann Gernsbacher and Elizabeth Grace, Language and Speech in Autism.

For more information on how poor quality research harms autistic people, see autistic researcher Michelle Dawson’s blog, The Autism Crisis.


[1] David Dobbs (Aug. 2, 2017). Rethinking regression in autism. Spectrum News.

[2] Creak M. (1961). Schizophrenic syndrome in childhood (Progress Report of a Working Party, April, 1961). Cerebral Palsy Bulletin, 3, pp. 501–4.