Jason Wolff from the University of North Carolina at Chapel Hill reported on a study of at-risk infants at the IMFAR 2014 press conference today, and it got me thinking.  At-risk studies follow tykes whose family history suggests they have an elevated risk of being diagnosed on the spectrum.  Researchers hope to learn what sets the at-risk kids who develop autism apart from those who don’t. 
What are we learning, and how are we acting on that?  Are we making the right choices?
He says repetitive behaviors may be one area of difference.  Dr. Wolff found distinctly different repetitive behaviors in the children who were ultimately diagnosed with autism.  These differences showed themselves as early as 12 months.
The excessive repetitive behavior set them apart from other same-age tykes who did not go on to receive autism diagnoses.  The behaviors varied from tyke to tyke, but across the board there was more hand or wing flapping, more spinning, more insistence on sameness, and even more head banging and other self-injurious acts.
Dr. Wolff's work is a major achievement and I applaud both his efforts and his success.  The question is, what next?  In his talk, he said, “the earlier we can pick this up, the earlier we can start intervention.” 
That’s when I perked up my ears.  Picking up distinctive behaviors at 12 months is a great achievement, but is quick suppression the correct answer in every case?  As an autistic adult, I had to wonder.  Ten hours later, I’m still wondering.
If I had a baby banging his head on the floor, I would be looking for intervention tomorrow.  I would be terrified that he’d do himself permanent damage.  But what if he flapped his wings and flew?  I did that, and it never harmed me.  How might I be today, if that behavior were “corrected” through early invention 50-some years ago?
Where do you draw the line between stopping something unhealthy and tolerating or even encouraging natural difference?
Might an early intervention have squashed my creativity before it even sprang into view?  No one can know.  And that is the problem with some of today’s intervention, well meaning as it is, and as vitally important as it is for some kids.  We cannot know how many less-impaired kids are being harmed or stunted though our well-intentioned “corrective” efforts.
We say “every autistic person is unique,” but then we talk about our population as if we are all the same.  “Once we detect the behavior, we apply early intervention.”  There’s not much acceptance and encouragement in that phrase.  Yet those are the emerging bywords for the autism community.  Autism Awareness is becoming Autism Acceptance all over the world.  We're recognizing autism as a disability for many and a gift for some; we're also seeing it as an essential variant of the human genome.  It comes with problems, but it also carries gifts.  We've realized the "disease model" doctors embraced 10 years ago does not really apply.
Today we know that some autistic behavior should not be suppressed.  When we exhibit a behavior, it is for a reason.  People – like other animals – do not do things for no reason at all.  We do what we do because it makes us feel good, stops a hurt, satisfies our hunger, or meets some other need.  It may not be obvious to an outside observer what need is being met, but rest assured, there is one.
Rather than ask “how early can we correct the behavior,” I think we would be better advised to ask, “why does an infant bang his head on the floor?”  I was a head banger, long ago.  Try as I might, I cannot recall for you now what I hoped to achieve by denting the wall in our apartment.  But I know it was a very deliberate action on my part.  Bang.  Reflect.  Bang. At some level, I liked it.  I did it a long time.
If you want to truly stop that kind of behavior, you need to figure out what need it fulfills, and fill that need another way.  Saying, “don’t do that” isn’t an answer.  Interventions for head banging rarely work because people seldom get to the root of the issue – why do we do it?  It’s hard to answer that kind of question because it requires getting into the heads of autistic kids and neurotypical psychologists are ill-equipped for that task by their design.
Then there are the repetitive behaviors that are not obviously harmful.  It’s my belief that autistic behaviors that don’t harm us, and don’t harm others, are best left alone.  Why not let them go?  I agree we should teach kids not to do injurious things.  And we should teach kids not to do things that will get them teased or bullied.  But the rest . . . ?
We are beginning to recognize and embrace neurodiversity, especially among adults.  We see how “different” people bring fire to the world, and we need that fire to bring ideas to a boil and pop them out into the world.  Our goal should not be to stamp it out in infancy.  Rather, the goal should be to tend and nurture the fire-bringer, and keep her from getting burned in the process.
At the same time, we might recognize that some people in the fire-bringer’s family won’t have the gifts of fire, but they will have disability.  How can we relieve such burdens as they may have, while encouraging the unique and special among us?

I want to be absolutely clear on one thing - I am not coming out as an opponent of early intervention. I have seen and heard too many stories of positive transformation of major disability to do that.  I'm just suggesting that early intervention - powerful as it can be for some - is not the answer for every child who shows non-injurious autistic mannerisms.
We have no idea how to predict which infant will develop into a star, and which will be crippled as an adult.  And this isn’t just an autism problem.  We can’t tell that for typical kids either.  We can identify certain serious genetic problems – SHANK 3 mutations for example – that are associated with profound intellectual and physical disability.  But beyond that – when we diagnose by watching things like hand flapping – we have no idea of the future development of those kids.  We’ve no clue how they will turn out.
Deciding when and where to intervene, and how, is getting much harder as the detection tools evolve faster than we can understand their meaning.  Intervention is also complicated when the ethical landscape is shifting.
We can’t lose sight of the fact that some infants growing up with autism will suffer greatly unless we intervene successfully.  But others will be the different quirky people who make the world run.  We can’t wipe out one as we ease the pain of the other.  But we have no idea how to tell them apart, especially at 12 months of age.
I don’t pretend to have the answer to that question.  As I wrote above, there are some genetic markers for very serious disability, but when we move beyond those . . . it’s not much different from predicting the future in any other area.  We can’t do it.
It concerns me that this is not really being discussed as we expand our efforts to detect autism and difference earlier and earlier.  Some of today’s autistic infants are surely the Beethovens and Einsteins of tomorrow.  Where will we be, if we “correct” their trajectories before they even read or write?  We’ll never know what we lost.

How might we start this discussion?

John Elder Robison is an autistic adult and advocate for people with neurological differences.  He's the author of Look Me in the Eye, Be Different, Raising Cubby, and the forthcoming Switched On.  He's co-founder of the TCS Auto Program school and Neurodiversity Scholar in Residence at the College of William & Mary.(c) 2007-2011 John Elder Robison