Skip to main content

Verified by Psychology Today


Is Early Autism Intervention About to Become Too Early?

Our developing ability to detect autism in infants raises some new questions

A few short years ago, early intervention was the hot thing. Diagnose autism early, the thinking went, and deliver interventions to help kids communicate in the conventional ways and get along before they fail in school.

It was a noble idea. Parents or pediatricians would recognize the early signs of autism. They would seek out testing. Results in hand, they would push for early intervention, and by the time the child was a teenager, the autistic disability would be minimized.

There was a lot of evidence that this was a good idea. Kids who got speech therapy when they were little supposedly developed better language than kids who began therapy as teens. But we may have been comparing apples and tomatoes. We’ve no way of knowing how similar the two kids really were to begin with, so the claim that the kid who got therapy earlier turned out better was really unprovable.

Yet the idea made sense if for no other reason than the time. If you have two kids, one 5 and one 15, and you give both of them speech therapy until they are 18, who would you expect would be the better speaker?

What if you gave both of them ten years of therapy? Many people would still bet on the younger kid, and the evidence would be on their side because kids pick up language best at an early age and it makes intuitive sense that language intervention would be most effective then too.

Many studies supported that, and most made sense to me. I supported the idea of helping kids with challenges we can see or hear as soon as we notice. I still do, in the case of visible impairments. But what about eccentricities? Then it's not so clear cut. Some stimming behavior would be an example - comforting to us autistics, generally harmless to others. Something to "control" if it leads to bullying or trouble, but leave alone otherwise.

Now there's another wrinkle . . . adults who were the beneficiaries of early intervention have grown up, and spoken about their experiences. Some thank the wonderful people who helped turn their lives in a good direction. But others talk about supressing behaviours that embarrassed parents even as they comforted the autistics themselves. They talked about adults imposing their will where it was most assuredly not wanted. When autistics look back on their childhood therapy the reviews are decidedly mixed. It's something to think hard about going forward.

It still is a good thing, some of the time. Maybe even a lot of the time. But when considering intervention for a child today we should be asking: Is this going to help the child, and help the child feel good about herself, or is it for the parents? The answers should be reviewed anytime things change.

And what about the younger children, the infants . . .

Researchers have been chasing two goals, with respect to early intervention

- Detect autism even earlier

- Develop more effective interventions for the many possible issues

Progress on the first goal has now far outstripped success with the second. Today, using several new experimental protocols, we can detect signs of autism in 6-month-old infants. Already there is a cry to start intervention “right away!”

Therein lies the problem. What intervention, and to what end?

When we test a 4-year-old, we can recognize problems that cry out for solution:

- A child may not talk or seem to understand

- A child may not engage parents or others

- A child may show self-injurious behavior

When we sign up for an intervention for a 4-year-old, we know what issues we are trying to help with. Consequently, we can see whether progress is being made, how it sits with the child, and adjust plans accordingly.

We don’t have those advantages when working with infants. We may pick up a sign of autism, but what kind of autism? Will the child be verbal or silent? Will the child be a lovable eccentric, unable to care for themselves or talk, or somewhere in between? It’s too early to know.

If we have learned one thing about autism it is this: The spectrum spans the whole breadth of humanity. Just as there is no one “human therapy,” there will never be just one “autism therapy.” Therapy must be delivered in response to an actual need, or one we KNOW will arise.

It’s time to take a step back, look at what we know, and where we can take constructive action. Knowing an infant is “developing autistically” is probably not enough to choose an intervention or even to know if one is needed. How autistic? In what ways? The last thing we need is to hammer kids with possibly unneeded interventions that may harm as many as they help.

As much as I believe in science and therapy to help our population, we need to know who needs what, and what will work. In the case of infants, I suggest we don’t know enough to act wisely in most cases.

As always, more work is needed...We want to act from a position of knowledge.

More from John Elder Robison
More from Psychology Today