Skip to main content
Staff Portal Staff Portal

- We're gonna continue along with the theme of teaching kids skills that consist of noticing their own behaviors, so the behavior of noticing your own behavior.

It's a common thread throughout all of this, but now, with self-monitoring, it really is the target of what we're teaching.

Raise your hand if you have sufficient self control in every aspect of your life.

(audience chuckles) (chuckles) Right?

So again, the skills we're talking about throughout today, again, are not specific to autism at all, but they're specific to humans and functioning as humans.

And they're areas that kids on the spectrum may well need to learn as well.

Someone just came up just a second ago and said, "Hey, by the way, I work with folks with TBI and dementia, and this stuff applies equally to them as well.

Okay, so self-monitoring or self-regulation involves monitoring the effect that one's behavior has on others if it's in social circumstances and/or monitoring the effect that one's behavior has in other ways, maybe a non-verbal or non-social ways if we're talking more work or task-oriented behaviors And yes, I'm a behavior analysis geek, so I always have to bring it back to the B.F.

Skinner, but the neat thing is he did actually… So I don't know if folks know this, but the first couple decades of his career, we pretty much just worked with rats and pigeons, and, by the way, he didn't do punishment.

Like, that's a big thing that's misplaced in understanding of behavior analysis.

Skinner actually only did one study ever on punishment, that involved shock, and I think it was shocking the rats paws, and he felt so bad for the rats that he never studied punishment again for the rest of his career, and he actually spent the rest of his career advocating for the use of positive reinforcement and avoiding punishment.

But in any case, back to the main point here, Skinner spent the first couple of decades of his career, really the first decade, doing basic research with rats and pigeons, and then after that pretty much the rest of his career was spent writing books that were an attempt to push a behavioral understanding of a human life into every aspect of human behavior, so he wrote books like "Verbal Behavior," which is a behavioral account of human language and cognition.

He wrote book called "Science and Human Behavior," which is really trying to deal with everything.

He wrote a book called "Beyond Freedom And Dignity" that was about applying behavioral terms and analysis to making the world a better place.

He wrote a book on aging, being happy in old age.

And the point of all of these books was, again, to push behavior analysis into being a more comprehensive science of everything that humans do, so there's always nuggets there.

If you're interested in how do we push ABA into some other area, there's a nugget of what you're interested in in something that Skinner wrote.

So he did actually talk about self-monitoring and self-awareness.

He talked about, as children in typical development, we start to become aware of our own behavior when the verbal community, in other words, like, our parents and teachers, want us to, when it matters to them, okay?

So basically, translate that into when our behavior as children becomes annoying enough to other people that they start prompting us inadvertently but prompting us to pay attention to our own behavior.

So any of you who have kids will notice, we all do this with our kids, when they do something naughty, you say, "Why did you do that?" or "What did you just do?" right?

So what we're doing without even meaning to is prompting the child to notice their own behavior and to talk to you about it and to talk to themselves about it.

So that's the early development of self-awareness from Skinner's perspective, so it's paying attention to your own behavior and describing, engaging in verbal behavior about your own behavior.

Okay, so self-monitoring is… Most of these areas of executive functions, there actually isn't a lot of research on the application of ABA.

Other than what my friends and colleagues and I have been doing, there hasn't been a lot.

Self-monitoring is a little bit of an exception.

There actually is a good amount of research showing self-monitoring affects behavior.

Raise your hand if you're a runner.

Do we have any runners in the audience?

Got a few, okay.

Have you used any of the apps, like MapMyRun or Strava or any of those?

They track your distance, and they track your elevation gain, and they track your time.

And don't you notice, even if you didn't mean to, you start to look at those numbers and you start to care about them.

Even if you're not trying to necessarily be the best athlete in the world, those numbers matter, They impact your behavior.

In fact, sometimes it's so effective that you stop doing it because you don't wanna deal with that, you don't want that input.

But the basic point is noticing your own behavior has an effect.

It's something that happens in your environment that then feeds back on your behavior.

Checking your daily work calendar, how effective would we be at work if we never looked at our calendar, right?

That matters.

If you wanna change your eating habits, this is a sad thing to say, but it works really well if you just write down everything you ate every day.

There's apps for it, or you can do pen and paper.

If you want to, you can try to calculate the actual calories.

If you want to, you could try to calculate that into how much exercise you would need to do to equal that amount of food.

And I'm not saying you should do that.

All of that gets kinda stressful and kind of aversive and stuff, but doing those things impacts your behavior, and it will change what you eat.

Okay, so that's kind of well known.

There's research on that across large varieties of areas.

Let's talk about how we can apply this to the kids that we work with.

There have been some ABA research studies involving self-monitoring that have shown that it's effective, for example, habit reversal.

Raise your hand if you've heard of habit reversal.

It's a treatment package.

Okay, it's just kind of a funny sounding name for a behavioral analytic treatment package.

Well, so the purpose of the treatment package is to help someone decrease a habit that they know of and that they kinda wanna get rid of, so it could be like thumb sucking, could be nail-biting, could be twirling hair, could be ticks, repetitive behaviors, things like that, but there's research showing habit reversal works for a lotta different folks, and it involves teaching the individual to be aware of their own behavior, so self-monitoring training, taking data on their own behavior, and then getting social support for changes in the behavior, and social support just means someone else in that person's life giving them reinforcement if they meet some kinda goal for reducing the behavior.

Goal setting and feedback, raise your hand if you manage employees.

Do you have other folks that work underneath you, teachers or educational assistants or folks like that, who is performance it's kinda your job to change?

So probably, if there's one intervention that's been proven the most in the organizational and behavior management literature, so the performance management research literature, one procedure that's been proven the most is goal setting with feedback.

So if you're a responsible for supervising somebody whether it's your student, it's your employee, whatever, sorry, and when I said student, I meant, like, you're a graduate student or something like that, if you want their behavior to change, you wanna manage it effectively, sit down on a regular basis, set a goal that's measurable and clear, and then set a meeting when you're gonna follow up, so it could be like you meet with your employee once a week, you make a list of goals for that week, and then you meet the following week, and you just go down the list.

What did you get done, and what didn't you get done?

And you just give them frank feedback.

Stuff they did well, praise it honestly and genuinely.

The stuff that they didn't do so well, point it out and say, "Here's why it matters.

Here's why it has to get done.

What are we gonna do next week to make it work?" and just doing that, setting the goals and giving feedback, and by the way, yes, you do actually have to meet and give the feedback, you can't avoid that part even though it's aversive for everybody, just doing that changes behavior because what it does is it gets your employee to then look at their own behavior so they have a list of goals sitting on their desk for this week that they have to accomplish, and everything that they do now for that week, they notice, they see their own behavior, and they're comparing it to the goals.

Like, "How close is my behavior to meeting that goal?" Very, very proven to work.

Self-evaluation in training, so this is an interesting thing, there's a few studies now on staff training where they've shown if you teach staff to… Like, let's say you're training a staff member to do discrete trial training, let's say, or whatever procedure it is you're training the staff member on, the educational assistant, whoever the person might be.

You train them on what to do.

Then you videotape them trying to actually do it.

And then you let them watch the video tape and maybe give them a data sheet to score their own correct and incorrect implementation.

Just doing that makes them do a better job.

So the next time they go to do discrete trial training, they'll do it better, and if they have to score their performance again, they'll do it better, so just the act of noticing your own behavior changes your own behavior.

Seems to be a pretty universal thing.

There's not that much research though on self-monitoring training for folks on the spectrum.

Okay, so I wanna share some clinical data.

We've done this with two children so far, both of them very effectively.

Both of these kids were so cute.

One of them, not the graph I'm showing you now but another one, was the same kid with the tassels on the rug, poor guy.

He had the cutest forms of stereotypy.

We had already helped him decrease all the sorta major forms of repetitive behavior.

He's not doing any rocking or hand flapping, not lining up objects, all that really obvious stuff.

Now he had gotten down to just sorta minor stuff that was cute to us and maybe to… Actually, no, I think it drove mom crazy, but the reason why it was a good idea to decrease it was because it was actually a problem socially, so for this kid, and I'm not making this up, this is real, for this kid one of his forms of stereotypy was contextually inappropriate karate moves, (audience chuckles) so he'd be talking to you, and in the middle of the conversation, he'd be going like this, like that, like this.

And we weren't talking about karate.

We're talking about what you had for breakfast or whatever, but he's just so into karate moves, and it was the cutest thing.

It broke our hearts to even help him decrease that behavior 'cause we loved it so much, but think about how well that works on the playground with the other kids going, "Dude, what are you doing?

Stop." They probably wouldn't even give him the feedback.

They'd probably just walk away, right?

And then another really cute stereotypy that he had was extremely exaggerated eyebrow movements.

So I don't know if I'm standing far enough or too far away from you to… Oh, they're videotaping.

This is not good.

I'm gonna try to model it, so he'd go like this when you're talking.

Again, and we're not talking about facial expressions.

We're talking about something else, but he could just couldn't help doing these really exaggerated, ridiculous eyebrow movements, but they were so cute, and you could see how maybe in the past it had gotten shaped up 'cause it was so cute, and mom probably laughed and hugged him and probably we did too, but it persisted, and again, it was weird.

Like, socially, it wasn't successful.

For this particular kid, the forms of stereotypy were… In the graph each baseline on the multiple baseline is a different form of stereotypy, so we have hand posturing on the bottom, putting fingers in the mouth on the second one up from the bottom, hair stroking and then ear twisting for the other two outta the four.

So what we did was we tried… We thought, well, most existing research has evaluated packages of, first, self-monitoring or combining self-monitoring training with some kind of reinforcement system, like a DRO, like, you'd get a reward if you don't do the behavior, and some other kind of treatment packages.

So we thought these kids actually do you want to decrease the behaviors, they kinda get it, that it's weird, and they actually want to have friends and be successful socially.

So it's not that they don't wanna do it.

The problem seems to be that they're really, honestly not that aware of it, that they don't really notice that they're doing it, so what we decided to do was just teach them self-monitoring and not change anything else.

So in baseline the baseline phase for all of these is just normal, totally unstructured free-play interactions during therapy sessions.

So we do some structured time at the table when we're working on particular skills that they needed tutoring on and then some unstructured, and then some sort of play-based instruction.

These were times that were completely unstructured just hanging out, so breaks, basically, free-play breaks.

And we measured each behavior.

Actually, we videotaped with a hidden video camera, and then we would take data on each behavior.

In baseline there were no contingencies at all, no consequences for the behaviors at all.

The kids already knew that mom doesn't want them to do that stuff, they knew they're not supposed to or whatever, but there was no specific intervention in place.

In the self-monitoring phase, all we did was sit down with them and say, "Okay, I want you to start noticing when you do this behavior like this, twisting your ears," and we'd just role-play with them, "Okay, show me what it looks like.

Yeah, okay, notice that?

All right, that's what we're talking about.

That's that ear twisting, okay.

Can you notice that for me?

Cool, all right, let's practice.

And now I want you to collect data on it." So we'd train the students on how to collect data on their own behavior, one behavior at a time, and the data-collection system just depended on what was appropriate for them.

Think for one kid they had a golf tallier that they would keep in their pocket.

For the other kid, he had a crayon and a piece of paper, and he'd just make tally marks on the piece of paper.

And what you can see is the behavior was really variable in baseline but an immediate reduction and a huge decrease in variability for the first behavior.

Then we did it for the second behavior.

Real similar results.

It's still happening some but way less than in baseline.

Third behavior, same thing.

Fourth behavior, pretty similar.

Now, in that phase, if they didn't record an occurrence, we'd prompt them to record the occurrence, so there wasn't a consequence of doing the behavior, but there was a consequence of forgetting to record the behavior.

Does that make sense?

So as long as they noticed and recorded their own behavior, we had no interaction with them whatsoever about the behavior.

If they forgot to record it, then we'd remind them, "Oh dude, look, you just did the ear twist and you forgot.

Make a tally Mark.

Okay, cool, good job," and then move on.

And what you can see is it worked really well.

Then in the last phase, we just removed the prompting, the reminder in case they forgot to take data, but still no differential consequences at all, and it didn't seem to matter.

They didn't seem to need those reminders anymore.

For hair stroking for that one right there, it got a little worse, right?

But then it went back down.

And I'm not gonna show you the other graph for the sake of time, but it looked really similar.

So for these two kids who had ongoing, automatically reinforced repetitive behaviors that were fairly subtle, just teaching them to take data on their own behavior with no other changes in consequences whatsoever made a difference, a good enough difference to where that's all they needed.

And I'm not saying this is gonna work with every single form of repetitive behavior for all of your students, obviously, but there might be certain cases in which that's all you need, or you could start with that, start with self-monitoring training, see how that works.

If it decreases the behavior a little bit, but not enough you could add in some other system, like, okay, as long as the behavior happens at or below a certain rate and you're collecting data accurately, like, not lying, obviously, then you get some kind of reinforcement, like a DRL or a DRO.

Does that make sense?

Does that seem like something you could train EAs to do in classrooms is get students to take data on their own behavior?

What do you think about it sorta ethically?

Does it seem mean to kinda, like, make kids obsess on their own behavior and evaluate their own behavior?

I mean, I could see how if you were overbearing about how you did it, it could turn into kind of a negative thing, like, "Notice when you're being bad," you know, but that's of course not the route that we took.

It was very neutral.

It was just like, "All right, here's something we'd like you to do.

Great, it seemed to work pretty well.

Yes.

- [Woman] So with the self-monitoring there's no replacement behavior, so you're not getting any type of function, (indistinct).

- Yeah.

- [Woman] So that doesn't train to (indistinct).

- Yeah, great, great question.

Yeah, thank you for busting me on that.

So the comment was, "If you're just decreasing a behavior an automatically reinforced behavior, and you're not giving a replacement behavior, then you're not actually addressing the function of the behavior," right?

So if it's an automatically reinforced behavior, we should be giving them something else to do in order to give them that same automatic reinforcement, you can call it stress relief, you can call it satisfaction, whatever you want to call it.

ABA perspective, that's a similar type of automatic reinforcement, so all of these folks definitely did already have other replacement behaviors that they had been taught in the past before this, but it wasn't enough.

They were still doing some of the target behaviors, but it's a really good point, and I shouldn't even mention this without mentioning that too, that for sure you've gotta have something else for the individual to do, but that is very tricky for those subtle, repetitive behaviors.

Things I've done in the past are give the kid something that'll fit in his pocket that he can fidget with in his pocket that won't disturb the classroom and that won't interfere with his own ability to do tasks.

However, if he needs both his hands to do schoolwork, that might get in the way.

Are those spinner fidget toys taking off in Eastern Canada?

- [Woman] Yes.

- Are they good or are they driving you nuts or what?

What do you think?

Driving you nuts?

Are they distracting the kids from instruction rather than just calming them and keeping them… Yeah?

Yeah.

- [Woman] I think (indistinct) request on it.

- Yeah, yeah, there you go.

So it's interesting.

I mean, we'll see.

There's no research on those spinner stim toys, whatever-you-call-them toys, that I know of, but it would be an interesting thing to evaluate.

For some kids it might actually be a better replacement behavior than totally spacing out or large motor stereotypy or whatever else, but if the kid is just staring at it the entire time and sort of engaging in visual stereotypy, then obviously that's gonna interfere with their opportunity to visually engage with the teacher and instruction, right?

So interesting stuff.

Okay, so the steps for teaching self-monitoring, if you choose to do that, is, just like everything else in ABA, start really simple and build gradually and make it easy to be successful and use lots of positive reinforcement.

Don't make it naggy.

Don't make it about negative reinforcement or punishment.

Make it about positive reinforcement.

Start small, and then gradually increase the complexity.

You'll notice, by the end of this evaluation, this kid was taking data on four different behaviors.

In the beginning we just started with ear twisting.

Then we added hair stroke, then finger in mouth, and then hand posturing, so by the end he's having to notice four different forms of stereotypy that he's doing and collect data, and he was actually doing it accurately, but we started small.

That's why we started with just one.

(child grunts) Okay, so here's another example of self-monitoring, and this is just a very small modification to a worksheet, so this is a worksheet that his teacher wanted him doing anyways, so all we did was added little check boxes next to each component of the worksheet where he makes a check or writes his initials or something like that that indicates he finished each individual component.

When he completes all of the components, he reports that to the EA.

Then he gets some kind of reinforcement

- [Boy] That's a check.

- That's a check.

(chuckles) (audience chuckles) So this is early on when he wants the social praise for each piece of the self-monitoring behavior, but of course we'd want to fade that out 'cause that's a crutch in itself.

- [Boy] Check.

Play! (claps) (audience chuckles) (speaker chuckles) And so you can see self-monitoring training is not just for decreasing misbehaviors, right?

It's also equally effective for increasing good behaviors, like staying on task, completing work, completing problems, so on and so forth.

Yes.

- [Woman] What if he gets it wrong?

- Yeah, so, "What if he gets it wrong?" It's a great question.

So I guess it would depend on the student.

If you're working on… So it depends, so either he's self-monitoring that he completed the problem or he's self-monitoring that he did it correctly.

If you think that's gonna be an issue, then you might need to make sure that it's self-monitoring when he completes the problem correctly.

If you're mainly just trying to teach the child to self-monitor staying on task, then it probably doesn't matter.

It's probably okay to just review the correctness of the work afterwards.

You can kind of think of this as a substitute for the child completing the worksheet totally independently where when a child completes a worksheet independently, no one's looking over his shoulder, telling him after each problem correct or incorrect necessarily.

They do the whole worksheet.

Then they turn it in.

And then you go over it with them.

So I guess it would depend on that.

I don't know if that makes sense.

Okay.

Okay, again, self-monitoring devices, we talked about the motivator before.

Definitely useful for that.

So you can set the motivator to go off every 30 seconds, once a minute, whatever, and the child could record whether or not they're looking at the teacher, so it can be a prompt to look at the teacher, and it could also be a prompt to record whether or not they are looking at the teacher.

They tally those up, record them, and then report back and, again, get some kind of meaningful consequences, meaningful reinforcement, when they meet some criterion, not punishment.

And of course, real-life application is the most important thing.

Some of these slides, I'm just gonna zip through them if I feel like you get the picture, but you have all of the bullets in your handouts and I can also make… Or actually, I guess already you have or Shelley McLean has the PDF copies of all of this, and I wouldn't mind if she distributes that to everybody.

That's completely fine.

Okay, let's talk about planning and goal setting.

So planning seems to involve anticipating future events, setting goals related to those events, developing appropriate steps, and organizing ahead of time to carry out a task or activity, so kind of anticipating what could be a problem before it becomes a problem, may involve imagining or developing a goal and then strategically determining steps to attain the goal, and it seems to help manage or sorta organize future directed behavior.

So from a behavioral standpoint, this is starting to get really complex and really confusing.

So remember we talked about we got a rat in a box with a lever, then have we a rat in a box with a lever and a light, then we have a rat in a box with two levers.

Wow, really complex, right?

Uh-uh.

Now we're talking about a person now talking about thinking about an event that's going to happen down here, hasn't happened yet, and now talking to oneself about what could happen and what one should do beforehand and then doing it and talking to oneself about the process of doing it and self-monitoring whether it's working right.

And then all the way at the very end, you have a consequence that matters, some natural consequence, like either the plan worked and you get to have fun and it's a good payoff or it didn't work and you're frustrated, whatever, so very, very complex.

And I think that's probably why not a lot of attention to this skill in the ABA literature, because it's kind of hard to figure out what's really going on.

Like, things are supposed to be divided up nicely into the ABCs, antecedent behavior consequence.

Well, that doesn't sound much like that.

That sounds like a lot of antecedents and a lotta behaviors and a lotta consequences.

Do the intermediate consequences matter or just the big one at the end?

So on and so forth.

I've been thinking about this for few years, and it seems like, again, related to flexibility, deriving rules that describe what you're going to do in the future and what's likely to work and not work seems like it's the core skill behind planning.

Okay, and the important point about planning is it's not something you've done before.

If it's something you've done before, it's not really planning, you're kind of just doing a routine.

Does that make sense?

Like, if you always have a playdate the exact same way and always in the same order, you don't need a plan for the playdate.

You just do it, right?

It's like a learned routine.

If your friend is coming over, and you're gonna have a playdate, and you've never hung out with that friend before or something is really different, the Xbox is broken or whatever, and you wanna plan for the playdate, then that is more like what we're talking about, anticipating future events and organizing one's own behavior around those future events based on how likely different behavioral alternatives are to be successful.

So it seems like it's almost always a very complex chain of many antecedents, behaviors, and consequences where lots of the behaviors involved are the behavior of talking about your future behavior and almost always involves deriving new rules, not just following old rules.

Okay, so the way we usually teach planning at FirstSteps is we kinda follow these basic steps essentially like a behavior chain.

So when you think about it, if it's a behavior chain, it might not be that different from teaching a child how to wash their hands.

Raise your hand if you use chaining for like hand-washing or basic like activities of daily living skills, right?

So first you turn on the water.

Then you get your hands wet.

Then you turn the water off.

Then you get the soap, rub your hands together.

So maybe it's kind of like that.

It's just that each one of the behaviors in the chain is a lot more complex than just turning on or off the water.

It's much more verbally mediated, so it seems like the first step in the goal, sorry, the first step in the chain of planning is identifying a goal.

So again, the playdate example, "Well, Jimmy's coming over and I wanna have fun.

I wanna get ready for the playdate.

Okay, cool, so what do we need to do?" So then the next step maybe is organizing and creating steps needed to reach that goal, identifying potential problems that could come up.

Begin the planned sequence of steps, self-monitoring your own behavior.

When problems come up, generate potential solutions.

That's the problem-solving piece, and we'll actually talk more about that in a minute.

So I'm not gonna talk a lot about it now.

If successful, recruit some kind of reinforcement.

So a lot of times when we're making a plan, there's some kind of consequence that matters that someone else is in charge of.

Like, if your mom asks you to clean your room, you might need to make a plan for how to get it all done, and you don't really wanna clean your room.

You don't really care about cleaning your room, but it's worth it because when you're done, mom says you can go outside and play or you can have your video games or whatever it is.

So recruiting reinforcement at the end sometimes could be an important step.

So here's an example from my friend and colleague Lisa Stoddard, who's the clinical director at FirstSteps.

She went to the dentist's office with her kid, who was, I guess, probably about four at the time, and you know how they, like, giveaway… Or do they do that in Canada?

When you go to the dentist they give the kids some prizes to try to keep them coming back and make the dentist office less adversive.

When you think about it, that's a pretty good idea, right?

So it's got some pairing there, got some positive reinforcement there.

So they gave her, I think, what did they give her, like, floss and some other stuff.

And then they happen to have, like, a dishwashing glove at the end.

This is something that just happened naturally, and Lisa thought, "No, this is kind of a good example of planning actually." So if you were gonna plan to do something fun and you were a creative four-year-old and you had a rubber glove, floss, scotch tape, and a stool, what would you do?

Like, what are some fun activities you could plan?

Obviously, you're not gonna be able to guess what she did, so don't even try, but just what could you do?

Practice your flexibility.

I mean, you could floss your teeth, but that wouldn't be very fun.

What else?

- [Woman] You could use the floss as a string.

- Use the floss as a string, okay, excellent.

What else?

What could we do with the rubber glove?

- [Woman] (indistinct) pinata out of it.

- Make a knot out of the… - Pinata.

- Oh, a pinata.

I like that, okay, cool, so let's see what she actually did.

- [Lisa] What are you doing?

- [Girl] I'm doing this.

Can you make a video?

- [Lisa] I am actually taking a video right now.

What is it you're doing exactly.

- [Girl] I am swim like a zip line! - [Lisa] You made a zip line! How did you make this zip line?

(floss hisses) (stool thuds) Whoa! (girl giggles) - So just a silly little example of how this typically developing four year old turned a bunch of random stimuli into a plan for doing something fun and naturally reinforcing.

Okay, so let's talk about how we do some of this stuff.

So start small with something that's not gonna be difficult to plan for and ideally something that's gonna be fun, so it could be pretty much any board game that the student likes to play or any game really that the student likes to play, but rather than just playing it and seeing how it goes, pause right beforehand, and go through those same steps, so you might just talk the student through it, like, "Okay, cool, so we're gonna play X, Y, and Z game.

Cool, so what's the object of the game.

Oh, okay, you gotta get your player over there" or "You gotta make four in a row." Like, connect four is a great planning example for teaching planning.

"Okay, you gotta get four in a row.

Okay, cool, well, what are you gonna do?

All right, you could stack them this way.

You could stack them that way.

Okay, let's let's give it a shot.

Let's try it." Then you initiate the game and then, at each opportunity for a turn, freeze, pause the game, and make a plan and engage in that same steps, those same steps of the student talking themselves through behaviors that they could engage in and what the consequences will be.

"Well, I could go there, but then that'll let Jimmy win the game" or "I could go there and that'll trick him and make him think that I'm gonna do that, okay, great." So you go through that same process of running through those steps, of talking about what to do, making a plan, and doing it, and then pausing multiple times, and basically then letting natural consequences kick in, like, either it worked or it didn't work, but make it fun to begin with.

So basically it's like you're overthinking the hell out of everything is essentially what you're teaching the kids to do but in a fun way.

Then of course there's lots of sort of contrived ways to practice this stuff.

Packing a suitcase, That's really boring if it's just random and arbitrary, but what if you're going… Well, do you guys go to the beach here?

Do you ever have beach weather?

(audience laughs)

Not really?

Where do you go?

You go to the cottage?

(woman speaks faintly)

Okay, go to the cottage.

So let's say you're going on a trip or you're going for an outing that the student really wants to go on and is excited to go, so you're building in the positive reinforcement to begin with, packing a suitcase for it or packing the bag or whatever is a great opportunity to practice planning.

Problem with that, of course, is how often does that really happen?

And repeated practice is critical to learning this stuff, so if you're gonna do that, that's great, but you've gotta build in lots of other practice opportunities too: taking a shopping trip, packing for a picnic; building with LEGOs or any blocks is a great opportunity to practice planning; a maze, a really simple maze.

And you can start with a maze that only has two turns at first.

You walk the kid through it, "So, okay, we're starting here.

We wanna get there.

What do we have to do?" "Okay, well, let's see.

First I go straight.

Then I turn right." "Okay, you have a plan, first straight and then right.

Let's try it.

Go for it." Then the student starts to do it, and as they do it, prompt them to talk to themselves about how it's going.

"Okay, I turned right.

It's working.

Now I went straight.

That's working.

Oh, I got to a dead end.

It didn't work.

I gotta try something different." So just getting that verbal behavior going of noticing the student's own behavior, talking to themselves about it, and then fading out your prompts and your reinforcement so that they're not depending on you to constantly tell them what to do.

And then of course, there's lots of compensatory strategies and environmental supports too that can be helpful with planning, so decision trees, to-do lists, all of that stuff, and to the extent that compensatory strategies are a normal part of what we all use, there's no reason why they necessarily need to be faded out.

So for example, if we're teaching a student to make a to-do list as part of a planning program, that's probably not a prompt that needs to be faded out.

Like, we pretty much always use to-do lists when we're trying to plan an important task whether it's shopping or packing for a trip or whatever.

So that's fine.

In other cases, if it's stuff that's really an additional level of support that shouldn't be needed hopefully, then you would wanna plan for fading those out.

Planning a playdate is another good one that we already talked about.

Making a gift for mom is really easy.

That's also an opportunity to work on perspective-taking, thinking about what mom really likes rather than what you like and maybe what colors mom loves, and you can draw a picture for her with her favorite color, and that combines planning and perspective-taking at the same time.

Cooking a meal or a snack is another great way to work on planning, but any of these things, again, make sure that there's lots of practice opportunities, that's the important part, and that the prompting that you're giving is getting faded out.

Okay, let's talk about problem-solving.

This is one of my favorite topics right now.

Here we go.

Here's Skinner again.

So Skinner talked about a problem as a situation where an outcome would be positively reinforcing if only you had the behavior needed to produce it.

So here's this consequence that could happen later on.

You see the consequences, or you're thinking about the consequence, or you know about the consequences, and that's gonna be a reinforcer, but you don't have the behavior needed to produce that consequence.

So a classic example with chimps is they had a an experiment where they had a big chamber that the chimps were in with bananas hanging from the ceiling but it was too high up for the chimps to reach.

Then they had some random boxes of different sizes laying around on the floor, So the test was to see will the chimps figure out to get the boxes and stack them on top of each other and then climb up and get the bananas, and, sure enough, chimps could figure that out.

So that's an example where the bananas were present, there's a clear reinforcer, the chimps want them, they're hungry, and they don't know what to do.

They've never been in this situation before, exactly like this.

Will they figure it out?

And the chimps… Well, we actually don't know.

I'm assuming the champs are not sitting there, talking to themselves in English.

How would they know English?

But they're not engaging in the complex verbal repertoire that we teach our students, but it's the same basic idea.

There's a problem.

There's reinforcer that you want.

You don't know what to do to get it, so you do something.

You rearrange your environment until the needed behavior becomes apparent.

Then you do that behavior and you get the reinforcer.

But it's not just an accident.

If we wanna teach this as a skill to our students, we don't just put them in situations of trial and error like the chimps and just kinda hope that they develop this or hope that they figure things out.

In fact, if we did just that, that would show that they don't need the instruction.

If we just put our students in problem environments and they solved the problems on their own, then, great, they don't need us to teach them problem-solving.

So for us to teach problem-solving, it's teaching that repertoire of being able to figure out a solution when you don't have one.

So in other words, it's the skill of figuring out what you need to do in order to get what you want.

So problem-solving is super critical to all human functioning.

If you look around in your life and your career and family life and social lives, and you look at the people that are the most successful, generally living a pretty valued life, they're getting what they want out of life, and I don't just mean, like, money and objects but outcomes that matter to them, I can guarantee you that they're not only just doing the same thing over and over, that they're engaging in flexible and creative ways, and, when problems come up, they're willing to try things differently than they have in the past.

So by Skinner's definition of a problem, there is an outcome that you care about, but you don't have the behavior needed to produce it.

By definition, if that situation comes up, you don't know what to do, so if you do the old thing that you used to do, it's not gonna work.

Does that make sense?

So have you guys worked with students with autism who have trouble with this, like, a problem comes up and they just try the same solution over and over and get frustrated and then quit?

Okay, so there is certainly research showing that there's deficits in problem-solving as a skill in folks with autism, very little research on teaching that ability.

So raise your hand if you've heard of the Self-Determined Learning Model of Instruction.

Many heard of that here?

It's pretty cool.

So there's a whole resource on the web.

It's a manual written by Palmer and Wehmeyer that's available for anybody to download for free, and most of the treatment studies or at least a lot of them were published by Agran and colleagues, and what they did was application of this self guided problem-solving model, teaching that model to students with intellectual disabilities, some of them autism, in educational settings to help what they described as problem-solving.

Now, it's a little bit different from what I'm gonna talk to you about, and it's a little bit different from how Skinner conceptualized a problem because most of these studies, I do recommend checking these out, they're worthwhile, but most of these studies, the problem wasn't something that the kids saw as a problem.

It was usually something that the teacher saw as a problem, so, like, the problem was the kid having aggression.

Well, that's not really a problem for the kid if he's getting what he wants, you know what I mean?

It's a problem for everybody else.

Or the problem was the kid talking out in class without raising their hand, something like that.

So remember Skinner's definition of a problem is it's a problem to the person themselves meaning they want some outcome that they can't get.

So it's a little bit different, but the basic model and process that they use for teaching kids to modify their own behavior in classroom settings is very relevant, so I recommend checking them out.

So let's talk about the relation between problem-solving versus planning.

So we just talked about planning.

They're very similar, planning and problem-solving.

Both involve deciding what you need to do in the future in order to produce a particular outcome, but it seems to me that planning is what you do before there is a problem, so if you're really good at planning, you don't have to do as much problem-solving, right?

If you actually plan something out really well ahead of time, fewer problems come up and you're able to just get it done.

If you don't do as much planning or if it wasn't possible to plan because something random happened, that's where problem-solving as a skill comes into play.

So let's go over some examples, so if you're planning for a playdate beforehand in order to make it as successful as possible, that seems like a planning skill.

If in the middle of your playdate, you accidentally hurt your your friend's feelings, and they're crying now, and you're trying to figure out what to do to fix the situation socially, that seems more like a problem-solving situation.

If you're planning how to complete a school project beforehand, so you've got a report that's due or a project that's due tomorrow, and you write down the steps needed to achieve that, that's more like planning.

If you're working on a project and the tool you need breaks while you're in the middle of the project, and you've gotta figure out what else to do to fix the tool or to use a different tool, that's more like problem-solving.

If you buy a new skateboard, and it comes with directions on how to assemble it, that's more like planning.

If a bolt on your skateboard breaks, and you need to figure out how to fix it or get a new bolt or ride it without the bolt, that's more like problem-solving.

So let's talk about what's involved in problem-solving.

So the way that we've been teaching problem-solving, and I'm gonna show you some data from real kids, is we teach it in these steps, and they sound almost exactly like planning.

It's just that you're doing it after a problem has started rather than beforehand.

So in a situation where a problem comes up and there's a consequence that the kid cares about and they want it but they don't know what to do to get it, then these are the steps that we teach them to do to walk themselves through: first, identify the problem; talk to themselves about why it's a problem; talk to themselves about possible solutions that they can engage in; choose a solution out loud, like, say what they're gonna do out loud; and then do it while monitoring their own progress; and then if it works, great, done, if it doesn't work, go back to step four, and pick a different solution outta the ones that you identified.

So again, it's a complex chain of behaviors.

Most of them are verbal behaviors, and they're behaviors of the student talking about their own behavior in addition to actually then implementing the solution.

So this is kind of a special skill 'cause it involves pretty much all of the other repertoires of behavior we've been talking about, so even noticing the problem involves paying attention.

Explaining why it's a problem involves some working memory.

Generating multiple possible solutions for sure involves flexibility rather than just doing the same old solution that doesn't work.

Choosing a solution and implementing it is basically planning.

This whole thing is planning.

Monitoring progress of whether or not that solution's working is basically self-monitoring, which we talked about before.

If it's unsuccessful, doing a new thing is flexibility rather than just trying the same old thing over and over.

So it basically involves all of those, so here's an example of a silly little game that you can use that, again, if the kid actually likes it, if it's reinforcing to the kid, it's a way to practice this planning and problem-solving stuff.

Have you heard of Peek a Boo Bunny, anybody?

Okay, maybe it's not popular here, but basically it has a bunch of parts, these parts on the left, so, like, a little doll, little bunny doll, and blocks and shapes and things like that.

And then it has picture cards, and you have to put together the parts to make it match the picture card, and there's lots of ways to do it wrong.

So it's a way to prompt the child to talk through what they can do, what's going to work, and then actually do it, and then talk to themselves about whether or not it worked and why it did or didn't work.

So there's lots of ways of where you put it together and it doesn't look quite right.

So it's an opportunity to engage in that behavior of, "Oh, it didn't work because the ears are behind the yellow block instead of in front" and then "Let me try it again" and then take it apart and do it again.

Okay, so in terms of the chaining bit, like any any other skill you teach with chaining, you could do forward chaining where you start with just one step at a time, like, first just teach the student to identify when a problem has happened.

Like, "Uh-oh, this is a problem." You could do that.

Then once they're good at that, move to the next step, like, "Oh, this is a problem because now I can't play the game because there aren't enough game pieces." Or you could use total task chaining where you teach all the steps all at once.

That's what we've done so far and that's worked, but with some kids with more limited verbal repertoires or less rapid learners, you might wanna do a forward chaining approach instead.

Use lots of prompting at first, like anything else, and reinforcement.

Support the successful verbal behavior at first.

Then remove your supports.

One thing that we've done a lot with these higher-functioning skills is rather than using model prompts where we just tell the kid what to say, we use what we call leading prompts, which is where we kinda guide the kid into coming up with their own solution.

So, like, let's say the toy doesn't work.

Uh-oh, we have a problem.

The toy's not working.

What should we do?

Well, instead of just telling the kid, "Check the batteries," we might say, "I don't know.

What can we do?" Then as the kid says, "Well, I dunno.

It doesn't work." "Okay, so what about it doesn't work?

What do you think?" So you use multiple leading questions to try to gradually, subtly shape the kid into coming up with an answer rather than just telling them what to say.

There's no research on whether or not that's actually even a good idea, but we think that it is because that in itself is prompting the child to do the behavior that matters, which is think about it for themselves, not just do what someone else said.

So here are some examples we've been working on, examples of problems, so the kid's doing an art project that he actually cares about.

It can't be one that he's doing because he has to.

It has to be one that he actually enjoys doing.

And let's say it involves coloring with a crayon and the crayon breaks.

Well, what can we do?

What are some solutions?

You could tape it back together, you could pick a different color to color with, or you could just use it anyways, the little stub, so those would be the actual solutions that we teach the kid to talk through.

Or we're doing a craft project, again, that the kid actually likes doing.

It involves gluing stuff.

Now the bottle of glue is jammed.

What can we do?

We could squeeze it harder.

Hopefully, it won't explode, spraying glue everywhere.

We can poke it with a paperclip or we could use tape or staples instead of glue.

How about if you can't open a box that's taped shut.

Maybe you got a game that the kid really wants to play or a toy that the kid really wants to play, and it's inside a container that's too hard to open or a box is taped shut.

Well, what are some solutions?

You can get scissors to cut the tape.

You could peel the tape off or just tear the box.

How about you're playing a game, again, that people wanna actually play, but there's not enough chairs for everyone to sit on.

What are some solutions?

Well, we could use something else as a chair.

We could find a crate, maybe, to sit on.

We could sit two people on one chair if their backsides are small enough, or you could find another one.

Whoa, sorry.

(laptop sounds indistinctly) No, okay.

But the main point is you work on problems that the kid is actually interested in and actually cares about that outcome at the end.

The natural consequences they actually want, it's a reinforcer if they can figure out how to get it.

Then these are the steps that we're teaching: identify the problem, explain why it's problem, generate potential solutions, choose a solution and implement it, evaluate whether or not it works, and if it doesn't work, pick a new solution.

Have a video, but I don't think I like it.

It's like too dark and it's too long.

It's not very good and I'm not gonna the video.

And it's not very cute.

(audience chuckles) Okay, now I'm going to show you data from a project that we're done with now, and it was our first project on teaching problem-solving, and here are data from four different students, and each data point is one problem, and we have 0 to 100% correct for each data point, and the percentage correct is the percentage of those six steps that the student did correctly without any help at all, so if we're at a 100%, they did everything right without any support at all.

If we're at about 80 to a 100%, that's probably good enough.

Maybe they got one thing wrong or they left one thing out.

Maybe they didn't state the solution before doing it, they just went and did it.

Oh, and let me note also that they needed or we prompted them to come up with at least three possible solutions every time that we're training them on problem-solving, at least three possible solutions.

What do you suppose the reason for that is?

(woman speaks faintly) Yeah, yeah, exactly, and it promotes flexibility too and creativity, and they have something to fall back on if their first solution doesn't work.

With a lotta these problem-solving kids that we're working with, I've noticed that there tends to be an impulsivity issue, like, they immediately just try something.

It doesn't work and they try and try and try and try and they don't take the time to stop, breathe, look at the situation, and think about different alternatives, so that in itself is a piece of what we're trying to establish here, stop, think, and then do it.

Okay, so saw pretty similar results across the four kids.

The first student actually took a while.

It was about 8 or 10 sessions before his accuracy started to increase, so still pretty much fully prompted.

Right around here we started to get worried that he was lacking some important prerequisite skills.

We weren't sure what those were really.

We thought he was ready for this, but we weren't sure, but then, sure enough, he started to respond pretty well independently, and by the end he's solving most of these problems pretty well by himself.

An important thing to note about this graph is every data point is a new problem.

So we never repeated the same problem.

Well, actually, that's not true, sorry.

The first few data points, we let them repeat the same problem if they want to do, basically, until they got it right.

Then after that, no repetition of problems, it was always new problems, so at least the latter half of the intervention graph, every data point is a measure of generalization in itself.

So what we wanted to do was continue until we saw substantial increases in accuracy with new problems.

And what was cool about this project was the parents, at least one of the parents, reported, "Oh yeah, he's actually using some of the language you taught him.

Like, when something doesn't work, he'll say, 'Oh man, there's a problem.' Then he'll start talking out loud about what to do." Now, do we always want kids talking out loud about everything they should do until the end of time?

No, right?

It would probably be great if they could just start to think it instead of always say it out loud.

Why do you suppose we required them to say it out loud at first?

- [Woman] (indistinct) monitor?

- Yeah, to make sure they're actually doing it.

If we just told them, "Think about it.

I really hope you think about it," how would we know?

We wouldn't know.

What we're doing is teaching them the thinking but teaching it out loud in a way that we can actually make sure they're doing it and we can help them to do it.

Then we probably should have included a phase at the end where we said, "All right, kid, you can't do it out loud anymore.

Just think it." We didn't bother with this particular project, so I think some of them are definitely still doing some of that self-talk out loud, but also, this is, like, the first time that these kids are learning how to solve problems on their own.

All of the kids involved in this project were kids where their parents reported are highly dependent on adults to fix things for them.

Whenever something went wrong, they would turn to an adult whether a teacher or a parent to fix things for them, so this is one of the first times that they're actually able to function and figure stuff out on their own, so, frankly, I'm kind of okay if they sound a little bit weird, talking to themselves out loud.

That's still pretty nice improvement.

(woman speaks faintly) Yes.

- [Woman] Seems like along the (obscured) topped out at maybe 80 or 90% accuracy in (indistinct) process.

What are the steps that was, like, the most common one they weren't really getting?

- It's not that they weren't getting it.

It's that it would fade out because it actually wasn't very helpful, and it was saying the solution out loud.

So early on, if they didn't say the solution out loud, even if we knew, like, in the early phases of intervention, even if we knew that they had a good idea, but they didn't say it out loud, we'd stop and we'd say, "Hang on a sec, bud.

You gotta tell me what you're going to do." I mean, you can see the light bulb go off.

You know that the kid's got it figured out, they're gonna go do something, but if they didn't say it, we'd go, "Wait, wait, wait.

You gotta tell me.

You gotta tell me.

Hang on.

What are you gonna do?

Okay, cool, go for it, awesome, but towards the end, we started to loosen up on that a lot because, you know, who cares?

It's a little weird and a little overly rigid for us to always require them to tell a teacher before they solve their own problem.

That's kinda defeating the whole purpose which is independence, so for the most part, it was that.

By the way, it's not on this graph, we also recorded did the solution work, did they come up with a solution that actually fixed the problem, and actually, 100% of the time, they did.

And the reason that is early on it was 100% because we wouldn't let them do it wrong.

If they came up with a solution that was unreasonable, like, one of them said something about a rocket ship will come down and fix it, like, "Well, do you have a rocket ship bud?

Maybe we need to come up with another solution," early on we wouldn't let them but then later on we would.

It was just, "Well, natural consequences.

Let's see." And it worked, so the solutions worked just fine, but what didn't work 100% was still them always doing all the verbal steps out loud, which, again, is probably actually better, I guess.

I don't know.

That was one piece of the study that we weren't actually sure about.

Like, how much of that do we require, the talking out loud?

And how long do we require it for?

When do we let that one go?

So we still required it the entire time for data collection because when you're doing a study, you can't change the data collection in the middle of the phase because that invalidates the study, So you have to stay with the same data collection even if it's stupid, basically, so yeah, But the data on did the solution actually work was totally solid, so we're happy with that.

Any other questions on this?

It's kind of a complex protocol.

It's a lot to squeeze into a 5 or 10-minute summary.

Does it seem like something that you guys could do with your students?

You kind of already do anyways, right?

Like, when a student is being unsuccessful, you talk to them about it, right?

And if they have the language, you try to get them to talk about it.

But here's a way of just doing that in a more structured way and doing repeated practice on it with repeated problems until you see that generalization to novel untrained problems being solved successfully, and I think that's the message here.

The take home point, if there's anything, here is I'm not teaching you guys on stuff you totally didn't know already.

If anything, maybe I'm giving you a structure for how to do it until you're sure that it's actually working.

Like, just talking to a student about their behavior once a day probably doesn't work.

It's gotta be repeated practice with fading out the talking because the talking itself is a prompt, fading that out until you see generalization to untrained and new circumstances.

Yeah, so parents reported generalization outside of session when the ABA therapists were not there which is really nice.

That's what we like to see.

For sure we need to evaluate social problems.

For this first project, we kept it to just non-social problems 'cause we thought that might be easier.

Honestly, we just wanted something simpler to start with.

Some social problems get really big and really heavy really fast, so we figured let's not deal with that for now.

So that's that's for the next study, but it's very important and needs to be done still, but I do think this same basic strategy should work.

It shouldn't really matter whether it's a social or non-social problem.

Either way it's the repertoire of talking to oneself about what's gonna work and what isn't and picking something that works.

Other compensatory strategies that are probably a great idea that we didn't try in that study are, like, mindfulness training or relaxation training, deep breathing, positive self-talk, counting to 10, things like that, are probably going to be helpful also especially if it's like an emotionally charged problem.

Handouts, like we said, so there's lots of resources out there on social problem-solving and conflict resolution.

And everyone's seen that stuff.

it's available.

What's missing is the systematic way to do it until you see the repertoire emerge.

A lot of this stuff you could do with the resources you already have.

It's just a matter of being really systematic about it.

And then, of course, real life application, super important.

By the way, all of the data points on the graph, on the graph that I showed you, were problems where, and I shoulda mentioned this 'cause really important, our staff pre-sabotaged the environment without the students witnessing it.

If the students knew that we caused the problem, I really think it would've hurt the treatment 'cause they could see it's not a real problem, so it was all stuff that there's no way they coulda known, just suddenly, the toy doesn't work anymore, or suddenly, there's not enough chairs, "Where'd the other chair go?

I don't know," very important.

So really all of the data points on the graph were real life application.

Okay, wrapping up executive functions, I'm gonna finish up executive functions.

Then we're gonna go on to perspective-taking and non-literal language.

So in summary with executive functions, the big themes here are flexibility, rigidity, we wanna increase flexibility, we wanna decrease rigidity, talking about one's own behavior with respect to the future, talking about one's own behavior and what's going to work and not going to work, taking the time to think before acting.

All of these are repertoires of behavior.

They're all skills in themselves that are common threads across all of these different areas of programming.

That's why I think all of these areas of programming will support each other, so that book that I showed early on, I think I have another slide later on that's gonna show it, called "Flexible and Focused" by Adel Najdowski, it was published in 2016, few months ago, basically has practical programs for all those different areas that we talked about.

And they'll all support each other.

It's not like you have to do one before the other.

They're all working on different facets of the same overarching higher-order skills.

So I hope none of you are gonna take away from this that I think that the brain doesn't matter in executive functions.

Of course the brain matters.

It's critically important to executive functions, and, yes, it's likely, although we don't know for sure, but it's likely that folks on the autism spectrum and of course every person's different, but that folks on the autism spectrum may have more of a neurological challenge with respect to EF skills than neurotypical folks.

Sure, that's totally possible.

Although I will say there's no direct evidence for that.

Like, they haven't found the part of the brain that's messed up that makes EF skills be a problem.

That doesn't exist yet.

They may someday, but I don't think that's actually how neuroscience works.

Like, there isn't some object or mechanism in the brain that says memory, that, oh, it's too small or, oh, it's too big.

It's not how it works.

Like, they can find underconnectivity or overconnectivity, meaning, like, too much activity or not enough activity compared to neurotypical people, in particular regions of the brain, and particularly regions of the brain are important for particular repertoires of behavior, but it's much more complex than that.

It's never, like, one part of the brain does one behavior.

That just doesn't exist, so what we know is the brain matters a lot, and the brain never does anything except when there's a human being alive interacting with their environment, so every time the brain ever did anything, it was in relation to the person behaving in relation to their environment, always.

It's always brain, behavior, environment always together, and it's not a sequence.

It's not like first brain, then behavior, right?

It's always brain and environment and behavior, all three always influencing each other, and there is actually research now showing that behavioral intervention does change brain connectivity, and of course brain connectivity changes how someone behaves, so it's always a coevolution between these two sources of influence, and, by the way, the only thing we can actually do anything about right now is the environment, which helps us improve behavior and skills.

So in conclusion with EF, you're not escaping today completely, got plenty more to talk about, but in conclusion with respect to executive function skills, it's a new area of application of ABA, but it does seem like there's evidence showing that this works.

And I will say to you, evidence-based practice is critically important.

It's the foundation and the core of everything we do, and you cannot wait for research on every single thing that you have to do with your students, right?

You can't wait for 100 studies published on ABA to improve working memory if your students have a problem with working memory, right?

I mean, we can't just pretend the problem doesn't exist because there isn't enough research, right?

I'm not saying go do some wacky stuff that has no evidence to support it.

I'm saying use evidence-based procedures that you're well-trained on to address new areas of human functioning, new areas of executive function skills that maybe there isn't a lot of research for yet.

And what's cool is your practice will feed back on researchers' behavior just like our research hopefully feeds back on your practice, so that's the goal.

That's where we're going.

So thank you very much for your attention to executive function skills.