Author: Barry Morgenstern

In high school, I had the opportunity to play a game of chess with a prodigy. He was around five- or six- years old (maybe a bit younger or older), and was an extraordinary player. He won the game easily. After the game, we had a long discussion where he was able to explain very advanced technical aspects of the game. Later, his dad came by, and he immediately jumped into conversations about little kid things like going to the park to play and watching cartoons. A few months later, I saw him on public television giving his thoughts on the live broadcast of the Chess World Championships.

This was long before I had ever heard of ABA, but I remember thinking how incredible it was that this child’s incredible knowledge of the game allowed him to engage on a deep level much beyond his current age, yet he still had “little kid” interests, too. He probably had typical social skills for a child his age (or at least one who was a chess prodigy). But he could engage with adults for long periods of time, and hold their interest in a meaningful way, because of the chess skills that he had developed.

I believe for many children with autism, this is our best shot to developing meaningful conversations and friendships. Although typical social skills programs can be very helpful, they often do not lead to friendships. It is hard because the process of developing a deep interest will likely not lead to immediate benefits, but in the long run, the procedure seems promising.

Years ago, I was working with a first-grader in an inclusion classroom. The student made fantastic progress with a 1-1 aide in the classroom. He had great reductions in problem behaviors, increases in appropriate behaviors, and was doing well on a variety of programming. For context, he had mostly completed Level 2 of the VB-MAPP.

But one area where the team still had significant concerns was his ability to socially interact with peers. There was a variety of programming in place for these skills, but he didn’t quite have the ability to keep up when his peers were talking at snack or recess.

A common tactic employed in this situation is to first determine what the other kids are talking about, and then teach our students to talk about the same things. Well, one of the little FIRST GRADE BOYS was saying things like, “Who is the hot girl with Johnny?”  when talking about the paraprofessional working with Johnny. That was a long way from the types of conversational exchanges that we were working on like, “I have chips for snack.” “What TV shows do you like?” The gap was obviously large, but a bit hard to quantify.

Although data looked good on increasing peer interactions, those interactions were (obviously) still awkward, and did not lead to meaningful friendships with peers. This is another situation, where it might look good in the short-term, but the results are not very meaningful towards improving anyone’s life in the long-term.

In my view, the solution to this problem is to focus on developing a deep interest. Friendships are usually based on having at least one similar interest that both people like to talk about. When you have at least a few appropriate things that you know a lot about, it becomes possible to find “your people.” This goes a long way towards helping children with autism develop meaningful friendships.

Don’t worry so much if they aren’t interacting with peers a lot. The goal isn’t to make the child popular. The goal isn’t to make the data look good to meet arbitrary criteria in the goals. There is a big difference between one meaningful friendship and zero meaningful friendships. The way to go from zero friends to one friend is to develop at least one skill, interest, or activity that is deep enough that it becomes possible to find others that like to talk about and participate in that activity.

There are a number of reasons why a child does not use a skill in real life which he has supposedly “mastered” in therapy. I have previously argued that one commonly overlooked reason is that the skill wasn’t fluent enough to be useful. I think that is true, but I’ve discovered it can sometimes lead us to think too narrowly about how good someone has to be at a particular skill in order for it to be useful.

When I was in high school, I entered the New York State High School Championship in Chess and won 1^st place. I have a trophy and everything. Now, to put my achievement in perspective, there was a masters level division, but I didn’t compete in that division. There was also a Class A Players (Experts) division, but I didn’t compete in that division either. Same with Class B, C, and D. I won first place in the E division.

To summarize, out of all the kids in the state who were the worst players on their respective teams, I won first place. My high school didn’t even have a team or a coach. I was self-motivated and went as the only representative from my school. Now, by any objective standard, I wasn’t a very good player. Isn’t it unusual to be self-motivated when you aren’t very good at a skill? Still, all of us E-division players had an enjoyable hobby.

In my view, the key factor that allows E-division-level players to enjoy chess is that our skills are “good enough.” If we struggled to remember how the pieces moved, or didn’t understand the basics of strategy, we would be bored to death. It actually takes a fair amount of practice to get to the minimal level necessary for the game to be interesting. Now, more than 35 years later, I even have a high appreciation of dramatic TV shows about chess due to my understanding of the game.

Sometimes, we give up teaching skills too easily. We tried teaching him drawing, bike riding, robotics, or Connect Four, and he didn’t like it. Sure, maybe the child just doesn’t like that activity. But there is a fairly high probability that he or she just didn’t learn enough skills to appreciate why the activity is interesting.

We often teach skills to fairly low levels and expect them to generalize. That rarely works. You need to teach skills to the “good enough” level to be useful, which will be more likely to lead to the child developing a natural interest in the activity that will persist over time. It is easy to be discouraged looking at the high levels of skill that the other children have, and to think this will be impossible. Remembering that the child doesn’t have to be great, just good enough that the activity is fun, is often more than enough for the child to use skills outside of therapy.

Frequently, I’ve discussed a variety of problems that occur when we attempt to measure progress in ABA for children with autism. Even when the data are collected perfectly, a variety of factors can make it look like the child has made more progress than he or she really has. Those issues can be problematic, because they can lead to bad treatment decisions.

I believe that one way to partially overcome some of those problems is to decide this ahead of time:

What will we consider success in ABA treatment?

There is only one answer, regardless of the child:

What happens after treatment ends?

We are thrilled when the child reduces tantrum behaviors, learns new words, eats something that isn’t orange, uses the toilet, or learns a new leisure skill. But if the child only uses those skills in therapy, it probably is not very meaningful for the child’s life.

Now, many times we have to start in a place that is very far from what it might look like when therapy ends. Sometimes, it can take a long time to get to the point where the skills are useful outside of therapy. We understand that. But if you start with the idea that success is what happens when therapy is over, you are much less likely to fool yourself with great looking videos, reports, or graphs of hundreds of changes that were made. There are lots of ways to do that which won’t last over time.

Of course, we don’t want to wait until therapy ends to see if we are successful. By then, it is too late. We need to do testing to determine if the long-term success if likely.

I have frequently discussed on this blog that while ABA has been extremely successful in the treatment of problem behavior, there is a very high risk that the effects will not last over time. That’s probably because the new behavior changes did not work under real world conditions. This has been understood since the earliest days of our field. Teodoro Ayllon and Nathan Azrin called this the relevance of behavior rule.  The rule simply states that you don’t change any behaviors that won’t work in the real world after treatment ends.

This puts BCBA’s in a conflict. On the one hand, we want to focus on the long term to ensure that our treatment will work in the real world and maintain over time. On the other hand, we have pressure to handle immediate problems. The parents are upset about the behavior in church or not being able to find a babysitter. The art teacher is complaining about behavior in class, the social worker wants intervention on peer interactions, and the bus driver says he won’t stay in his seat.

The basic problem is that we only have one good long-term solution. That long-term solution is to teach the skills needed to be successful in natural situations. It requires careful programming done in a step-by-step fashion until successful. Unfortunately, for many children, that can take a significant amount of time. It is not likely to help the bus driver worried about tomorrow’s ride. If you try to put the long-term solution into effect in every natural situation, right away you will almost certainly fail to be successful.

Now, we do have a lot to offer that may help all those other situations relatively quickly. But those procedures are not long-term solutions. Those are short-term fixes to get us to the point where the long-term solution becomes possible. But those short-term fixes can actually look really good for a long time. Don’t get fooled by the data. You need to do a lot more to ensure long-term success. 

In my view, in most cases, the solution to the problem is comprehensive treatment with two behavior plans. First (and most important); a plan to teach the necessary skills to ensure success over the long-term. Second; develop procedures to immediately help parents, teachers, and others who interact with the child. Those procedures will likely look very different from the procedures we are using to build the long-term skills. Although this is different than the traditional advice, I think the logic of two behavior plans is compelling.