Author: Barry Morgenstern

There is a lot of research on teaching children with autism conversation skills. Many times, I’ve seen children who have had a tremendous amount of this training, yet still there is something missing. It just doesn’t seem quite right.

What I believe the children are missing in some cases is enough background knowledge and interest in different topics to meaningfully engage. Some children spend so much of their time on iPads, video games, or extremely obscure interests that they don’t have a lot to talk about.

Although admittedly hard to prove in a scientific study, in my view helping children develop a variety of interests not only opens potential vocational and leisure opportunities, it can help with conversation skills, too. When you have a deep interest and knowledge about a topic, there is a lot more to say.

This goes a little bit against some of the conventional wisdom. Often behavior analysts want to know what is happening in the child’s current environment. What are the 3^rd graders talking about? Let’s teach him to talk about that. Certainly, in some situations that tactic can work great. But often it doesn’t.

Let’s say the kids are interested in the new Star Wars movie that just came out. The amount of background knowledge and language to meaningful participate beyond one or two exchanges is enormous. Often, you won’t be able to get meaningful conversation going using this strategy.

I strongly suspect it is better to take the long view of developing conversation skills. Helping the child develop an appropriate and deep interest in a few different areas will be much more successful in helping develop meaningful conversational skills and relationships than trying to teach the child to talk about whatever the current popular topics happen to be.

There is a lot of research that shows publicly posting data can be extremely valuable. Numerous studies demonstrate how effective this intervention can be. A commonly cited potential problem is that it can be aversive if you aren’t doing well, which might upset some people. But I’m not concerning myself with that problem here.

The main problem comes when people obsess over data that are publicly posted. These data will inevitably become a major focus for everyone. Now, if that is your top measurement, maybe that is OK. But probably not. Any data can be distorted when looked at in isolation.

If you are publicly posting data on things like whether the break room is cleaned up, then you are almost certainly using an intervention that is overpowered for what you need. But it might not cause too much harm.

We tend to underestimate how much data like this can influence behavior in negative ways. For example, I once talked to a teacher that worked in a behavior analytic school. The school had a system for publicly posting the number of teaching trials done by each instructor daily. Seems relatively harmless, right? In general, more teaching, more learning. But what happened is that teachers avoided teaching anything that required a lot of set up. Give him more discrete trial. Forget things like mands for information…

Once you start looking for this, you will see it everywhere. In my view, we don’t appreciate how often this effect happens in behavior analysis. Check out this book for a non-behavioral account of this effect in a variety of fields.

Eddie Murphy guest starred on Saturday Night Live in 1984. In his opening monologue, he said every host tells you what a great show we have tonight. Then, you see some great stuff, but you also see some stuff that sucks.

That’s the way most conferences work. Almost always there are great events to attend where you will learn a lot. But in all likelihood, you will see some stuff that sucks too. We can always try to get better at figuring out what will be great and what will suck when making your selections about what to attend.

But Eddie Murphy tells us the main thing you need to know about attending conferences. You have to go to a lot of presentations. The more presentations you go to, the more great stuff you will see. That means you have to sit through some stuff that sucks, but that’s the price of the Poogi.

It’s good to spend some time drinking at the bar too. It’s fun, and a surprising amount of POOGI can be found there. Just don’t drink so much that you are too hungover to go to a lot of presentations the next day.

Assessment devices like the VB-MAPP, ABLLs, PEAK, and others can be extremely useful. They can serve as a baseline level of skills. They can give useful tasks to get to know the child. They can give the behavior analyst ideas about potential target behaviors. They can give you some progress measures.

But in my experience, there is a downside.

Once you show people the grid – parents, teachers and many behavior analysts – they want to fill in all the boxes. The problem is that this is a natural reinforcer for everyone. We just love to see those boxes colored in with all the “progress” that the child has made!

Now, in some cases that may be perfectly appropriate. But in many, many cases it’s not. Often there are boxes that are extremely critical for a child’s life and others that aren’t. I’ve seen numerous times over the years where children spend enormous amounts of time on tasks that are clearly inappropriate just because a particular box is empty. Don’t waste time on things that aren’t going to last in the long run. Focus on what is going to make a significant impact.

The only thing that matters in the long run is things that maintain. Many of those boxes will require generalization and maintenance programming. If you don’t have time for that maybe that skill isn’t that important.

In summary, there’s nothing wrong with using these types of assessments. They can be extremely useful. Just be careful that you think through your selection of target behaviors. Don’t substitute the assessment grids for thinking about what really matters.

Behavior analysts learn to train others with a procedure called Behavior Skills Training (BST). Part of implementing BST is modeling for the person learning how to implement the skill. Now the usual problem with this strategy is the trainer might have the right letters after their name, but can’t actually work with a person, and/or can’t model the skill well. Of course, that training is ineffective.

But the opposite problem can happen, too. If the trainer is too skilled and is able to do amazing things, it can easily intimidate a new person. Parents or staff who are trying to learn a skill instantly think things like, “I could never do that.” They can become afraid to try, and give up too quickly. Even if the person learning isn’t too intimidated to try, they may attempt to copy fine points that aren’t critical for task at hand, and training can take much too long.  Engelmann warned us about this problem.

Of course, we want trainers to be great implementers. But it is important to try to remember what it was like to be new to a skill. Often, people who have been doing this a long time and are very skilled have forgotten what it’s like to be a learner. It is hard to not model everything you know, but only what the learner needs to learn.

You can improve your training skills by remembering what it’s like to be a learner. You do that by constantly putting yourself in the position of being a learner. You need to do that anyway, because if you aren’t trying to learn new things, you aren’t on a Poogi.