About four weeks after we started with a new occupational therapist, my son invited another child to play with him.
Nobody prompted it. It was not a target on a plan. It was the thing that years of social skills work had been aimed at, and when it finally happened it arrived sideways, from a direction nobody was measuring.
I have thought a lot about why, and the short version is that he had not been ready before, and being placed in a group is not the same as being ready for one.
Placed in a group is not the same as ready for one
A child can be physically present in a social setting and be doing nothing that resembles socialising. They can be enduring it. They can be watching from behind a wall of effort so high that nothing else is possible.
From the outside, presence and participation look similar enough that a session note cannot always tell them apart. My son spent a long time being present.
What I could not see at the time was that being placed in the group repeatedly was not building the skill. It was building the association that groups are exhausting.
Why the compliance route does not get there
When my son was in ABA, he learned what that room expected of him. He did not learn a skill that travelled. It was compliance rather than capability, and compliance is specific to the place that demands it.
Social initiation is the clearest example of why that matters. You cannot comply your way into wanting to invite someone to play. There is no prompt for it, because the entire content of the thing is that it comes from the child. A programme that produces reliable responses to adult instructions has not touched it.
The idea that changed how I read this
The double empathy problem, developed by Damian Milton, holds that the difficulty between autistic and non-autistic people runs in both directions. Autistic people find neurotypical social norms hard to read, and non-autistic people are equally poor at reading autistic ones. It is a mismatch, not a deficit sitting in one person.
If that is right, a lot of social skills teaching is aimed at half the problem. It trains the child to produce behaviour that non-autistic adults find legible, and calls the result social skill. That is a real thing to be able to do, and it is worth something. It is not the same as connection, and it has a cost, because performing legibility all day is work.
It also explained something I had noticed and could not name. The settings where my son did best were the ones where the adult adjusted to him, not the ones where he was successfully adjusting to the adult.
What the evidence supports
I want to be careful not to overstate this, because the honest picture is modest.
The 2023 BMJ review known as Project AIM pooled 252 studies and 13,304 children. Under its strictest analysis, excluding caregiver and teacher reported outcomes and studies at high risk of detection bias, the approach that retained a significant effect was naturalistic developmental behavioral intervention, and the effect size was modest. These are approaches built around play, natural settings and the child’s own initiations, described by Schreibman and colleagues in 2015.
So there is some support for following the child’s lead rather than driving toward targets. It is support, not proof, and a modest effect is a modest effect.
What I would look for now
Not in a programme, but in what is actually happening with your child:
- Is your child initiating anything, ever, unprompted? That is the measure that matters and it is rarely on the form.
- Is participation increasing, or just attendance?
- What does your child have left after the session? A social win that costs the whole evening is not free.
- Is support being faded because your child is ready, or because the plan says week six?
- Does anyone in the room adjust to your child, or is all the adjusting expected in one direction?
The part that is hard to hear
It takes longer this way. There is no version of this where a child who is not ready becomes ready quickly because the programme is intensive.
For us, the visible change took about four weeks, and it was small: more confidence, better impulse control, and that one invitation. Not a transformation, and I would not present it as one. But it was more than I saw in many hours of the other thing, and it was his, rather than something he had been trained to produce.
If your child is being pushed into groups and coming home wrecked, you are allowed to say that the pace is wrong. Readiness is not a scheduling problem.