- We're gonna shift gears a little bit and talk about two different areas of functioning.
Perspective taking and non-literal language.
But you'll notice they're very closely related.
And actually being good at these two areas probably involves a bunch of the EF skills we talked about.
So it's even though they're different sort of domains of functioning, maybe they're more language or social or pragmatic language rather than executive functions.
Very, very closely related still.
Okay, I'm actually not gonna spend time on this, but I highly recommend this book, "Learning RFT" by Niklas Torneke.
Relational frame theory like I said, and we talked about it a little bit already.
That was that sort of overly complex, probably not very interesting theoretical bit that I talked about in the beginning about the different ways of relating stimuli and how it's an ABA application to cognition and cognitive skills.
Well, Niklas Torneke is a psychiatrist from Sweden I think, or Denmark.
And he wrote this book for psychologists to explain relational frame theory to psychologists.
So it's written, it's not a bunch of jargon.
If you go and pick up the original RFT book or if you go look at a research article published in a scientific journal on relational frame theory, don't even bother.
Unless you're a total research, behavior analysis research geek, you're gonna be bored to tears.
And even if you are a behavior analysis research geek you're gonna be bored to tears, and it's not gonna make sense.
The way that RFT is usually taught and talked about is hopelessly complicated and jargon laden.
But this book is not.
This book makes sense and it'll be clear.
I think it'll be quite clear how relational frame theory is actually relevant to what you guys do.
Even though it's not written for folks working in educational settings, it's written for psychologists.
It's about human language and cognition.
And again, that's equally relevant to whether you're a child with autism, neuro-typical child or an adult, person with dementia, so on and so forth.
The first half of the book is just explaining RFT.
The second half of the book is relating RFT to psychotherapy.
So you don't need to read the second half.
The first half is really the important part that's relevant for what you folks do.
But here's the important point.
Here is why it's worth it to learn about RFT is all the most important parts of human behavior, the most fascinating, interesting parts of human behavior involve doing behaviors that were not directly trained and not just stimulus generalization.
So stimulus generalization is when you get generalization because of physical similarity between objects, right?
So if I teach you to call this a pen and now I hold up a different pen and you can call that a pen too, that's stimulus generalization, right?
'Cause the two pens are physically similar in some way.
If I teach you to call this a pen and then tomorrow I ask you, "Hey, go find a pen for me?" And you look around and you pick up a pen, that's not stimulus generalization because you've never done the pointing and picking up behavior in the presence of a pen before or anything like it.
What you were trained on was labeling or naming.
Does that make sense?
Two completely different behaviors that are related because of the relationship between the pen and the name for the pen.
And I know it's getting kinda abstract and kind of, I dunno, maybe not practical again.
But I do highly encourage folks, get the book, it's pretty cheap.
It's probably 20 bucks or something.
If you're interested in application of ABA to more complex skills and behaviors, it's the only game in town.
Okay, the first piece of relational framing or stimulus equivalents is called bi-directional naming.
And that's where when you train, you guys use the terms like receptive and expressive language.
That's a common ABA term, right?
Okay, so if you train receptive, you get expressive for free.
Or if you train expressive, you get receptive for free without direct training for a particular object or a particular stimulus.
Now, when you first start working with a severely affected child with autism, let's say age three with very limited language and you're training the first few object labels, expressive object labels that that child has ever done, there was no way they automatically now can receptively identify that object without direct training, right?
No way, you got to teach it directly.
Now, think about the client, the students you've worked with that have had the best part possible outcomes.
At some point you notice, "Hey, wait a minute.
We don't have to teach everything.
They start to pick it up on their own.
That's really cool." And if we want to we can just say, "Wow, I guess generalization happened." But that doesn't help explain why.
And research by Doug Greer at Columbia Teachers College in New York has documented how this happens and it's through no surprise, multiple exemplar training.
And so the procedure that they have done, and they published a bunch of studies on this now, is they take preschool aged kids who can't do that yet.
So they can be taught expressive language and they can be taught receptive language but they can't generalize between those two.
They have to be directly taught both, okay?
And they do this rapid rotation of multiple exemplar training through expressive or tacting, whatever you wanna call it, listener behavior or receptive, whatever you wanna call it and matching.
They rapidly rotate between trials of those three for a set of stimuli until the kid masters receptive and expressive with all those stimuli.
Then they go and teach another set, receptive only, and see if they can do it expressive.
If they can't, they go through that rapid rotation through matching receptive expressive, all in the same bit again.
And now they teach another new set of just expressive only or just receptive only, see if the kid can do the other.
If they can't, do it again, right?
So they keep doing multiple exemplar training, rapidly rotating between receptive and expressive and matching all in the same trial block with new sets of stimuli until the kid can now do receptive when they've only been trained expressive, or expressive when they only have been trained receptive.
Raise your hand if you've ever heard of the idea that Skinner's verbal operants are functionally independent.
So like mands and tacts and echoics and those things are like separate, right?
Well, I'm here to tell you that they're not.
For typically developing adults, if you teach me something as a mand, I can now use that same word as a tact or as an intraverbal or whatever other conversational behaviors, no problem.
And I mean, I can demonstrate that to you right now but I think it's so obvious that we don't even need to.
I could tell you, awa is water.
I can tell you that right now.
we could go back and forth a few times as a tact or a mand or intraverbal whatever.
We just do one of those verbal operants.
Now I take you to Mexico tomorrow, don't give you water for 24 hours and you can darn well bet you're gonna go to a store and say, "Awa, awa, awa," 'cause you want water, right?
So you can, generalization between the different functions of language does happen.
The interesting question is, how does it happen?
And for our students with autism, if they can't do that how can we get them to be able to do that?
And the research by Doug Greer's group has shown a really successful way to do it.
So imagine that car is one of the stimuli that the kid's learning.
He doesn't know it yet receptive, he doesn't know it expressive.
And now maybe insert a couple of other.
Maybe bicycle and maybe airplane, whatever.
So you got three different stimuli you're teaching and you've got a block of trials, 10 or 12 trials, whatever it is.
On trial one you might run a receptive trial, like touch the car.
Trial two, you hold it up and say, "What is it?" And you prompt him to say car.
Good job, reinforce it.
Now trial three you hand them a picture of a car and say, "Put the same," or match or whatever.
And he puts it with one of the other cars.
Trial four, randomly choose a different format out of those three formats.
Does that make sense?
So you're rapidly going back and forth.
Perceptive, expressive, matching, not always in that order, in a random order, and with the three different stimuli.
So it's kind of crazy, right?
It's kind of different from how you've probably been trained to do.
Probably receptive first, then expressive, right?
Or maybe matching first, then receptive, right?
And that's the way it used to be done.
That's sort of the Lovaas way was first matching then receptive then expressive.
By the way, there wasn't one shred of evidence, not one shred of research evidence that supported that sequence that Lovaas did.
He just thought it was a good idea, so they did it, and it did work, right?
There's no question that it worked.
But there was no evidence that that was better than a different way of doing it.
That was just his idea and it turned out that it worked.
Now, Doug Greer's research in the last 15 years is showing this other way produces this really, really cool outcome.
So it's something to consider.
Most ABA folks aren't doing this yet.
Like the research is still coming out and everyone's going, "Oh, wait a minute.
Do we have to change everything we're doing now?
Hang on a second." We have 101 reasons why we don't wanna change.
But this is a really cool outcome that might be one of the first building blocks of complex language, sorry, complex human cognition.
Is the idea that if A equals B, B also equals A and you don't need to be trained that.
Okay, that's all I have to say about that.
But it's worth looking into.
Oh and Doug Greer has a book called "Verbal Behavior Analysis" something like that.
Does anyone have that?
Raise your hand if you have Doug Greer's book, ABA book.
It's a good book, you should check it out.
I forget exactly what it's called, but Greer is the last name of the author and verbal behavior is definitely in the title.
But it's like a treatment manual, like one of the ABA books on how to teach kids with autism.
But it's kind of encapsulates his verbal behavior approach, including that procedure we just talked about.
Okay, let's talk about perspective taking.
So raise your hand if you've ever heard of theory of mind.
So tons of research in developmental psychology and cognitive psychology showing as a group folks on the autism spectrum are not as good at perspective taking as folks in the typically developing population.
Most of the theory of mind research just shows that.
Here's an area of perspective taking.
These folks aren't good at it.
These folks are good at it.
Here's another area of perspective taking.
These folks aren't good at it.
These folks are good at it.
And the important thing about that research is, it's documented a whole bunch of different areas of perspective taking that matter and that are important and are a critical part of every day adult social functioning.
And so if our students don't have those skills, we better figure out how to teach them.
I'm gonna skip this study.
But it's in there.
It's in your notes if you wanna take a look at it.
I'm gonna go straight to a different one.
Okay, so and the reason why I'm skipping that one is it's kind of simplistic.
It was the first study we did on perspective taking.
And it was like, "Yeah okay, it worked." But it was not the most interesting stuff and I'm running low on time.
Okay, so talking about identifying.
being aware of what other people want is really important to social functioning, right?
Raise your hand if you've ever worked with a student who kind of only pays attention to what they want, doesn't really notice, right?
And it's not like they're being a jerk.
It's not selfish.
It's just, they're maybe not noticing in a lot of cases.
And in fact, what a lot of the theory of mind research shows is they literally don't understand the concept that what someone else wants is different from what one self wants.
It's a cognitive ability that's lacking in a lot of folks on the spectrum.
So if that's the case, that's totally lacking, you're gonna have a real hard time interacting socially in a successful way, especially with typically developing peers.
And especially as they start to get older and especially anytime preferences change, right?
If you just always do the same thing with your friend and it always works, fine.
But as soon as you make a new friend, they might have different preferences.
Or as soon as your friend gets tired of the same thing you always do and wants to do something different and you're not aware of it, and you're not able to adjust your social behavior that's not gonna be successful.
So really, really important.
Lots of kids on the spectrum lack visibility.
Actually lots of typically developing adults lack visibility unfortunately.
Okay, so here's a, I'm gonna show you data from a project that I think we're pretty much done.
Yeah, we're done.
We're writing it up for publication now.
And it was run by my friend and colleague, Adel Najdowski.
She's the same one that wrote the "Executive Functioning" book.
Also the flexible, thank you, "Flexible and Focused" book.
Okay, so she did this project, teaching kids on the spectrum to identify what their friends want and identify how it's going to affect their friend's emotional state.
So there's four different things that we taught the kids to identify.
A friend getting what the friend wants which is gonna cause a positive emotion.
And we didn't tell them exactly what to say.
Maybe it's happy, maybe it's excited, maybe it's whatever.
Okay, but some positive emotion.
Then number two is a friend not getting what the friend wants is gonna cause some negative emotion, could be sad, could be mad, could be frustrated, whatever.
Then there's the friend getting what she doesn't want.
So getting something that they don't want, which is gonna cause a negative emotion.
And then there's the friend not getting or avoiding what they don't want.
They don't wanna eat broccoli and they didn't have to.
How's that gonna make them feel?
So there's four different possible future events of getting or not getting.
Someone else getting or not getting what they want and what emotions that will cause in that other person.
This is what we taught, the verbal behavior to predict all of this stuff to our students with autism.
Based on those predictions, the question we were asking was, "Would our students with autism then change their own behavior to help make their friend feel happy and not feel sad?" And therefore choose activities to make their friend happy versus make themselves happy.
And so the way that we set it up was we had play activities at the center.
And we basically said like, "Oh, Jimmy's gonna come play in a few minutes.
What does he like to do?" Okay, cool.
Yeah, he likes transformers, he likes dinosaurs.
What does he not like?
Okay, he doesn't like Connect Four.
He doesn't like, whatever.
Well, what should we play?
Like what do you feel like playing?
And if a student said, if the student identified something that the other person wants, great.
That's what we're looking for, right?
If the student identified something that they want that's gonna make the other person unhappy, not so awesome.
And it wasn't always stuff that the student didn't want.
We weren't training our students to never care about their own desires, right?
Of course not.
We don't want them only thinking about other people, think it was about 50/50.
So that was the question is, can we get them to accurately, two things, can they accurately predict what their friends, how their friends are gonna feel based on getting or not getting what they want?
And then will they then adjust their own social behavior accordingly?
So here's the graph.
It looks really complex and like there's a lot to it.
Probably what I should have done was only shown one participant's graph.
It's a multiple baseline across three kids.
Just maybe ignore the middle one and the bottom one, just look at the top one.
That'll be the easiest.
It basic real similar results across all three kids.
In baseline what we have is asking all the verbal behavior around it, which is the black triangles.
So asking the kid, "How's your friend gonna feel if they get what they want or don't get what they want or don't get what they don't want or do get what they don't want?" All four trial types.
And would they report accurately how it would make their friend feel?
And so this first kid did get it right about a quarter of the time.
I think they could accurately report, "If he gets what he wants, he'll be happy." But the other three trial types, this particular girl Isabelle was not able to report accurately.
Okay, that's baseline.
Then the natural environment probe there was, "Now okay, your buddy's coming over to play, pick something.
What should we do?" And we didn't influence that at all, right?
We just went through that talking through of how they're gonna feel, okay, what should we do?
And correct would be picking something that would make the peer happy.
Incorrect would be anything else.
So we can see in baseline she's picking what she wants, not what the peer wants.
Then we trained each one of those four trial types individually.
And you could do that or you could train them all four at the same time.
We just arbitrarily decided to train them one at a time because it gets kind of complex, right?
Get what you want, don't get what you want.
Get what you don't want.
Don't get what you don't want.
It's like a little much, right?
So we picked one to teach at a time.
I forget exactly which one we started with, and taught that until we got a 100% correct, no problem.
Then we went and taught the second one and that quickly went up to a 100%.
And then we did a real life probe again.
And so this is where we actually had her decide what are we gonna play with?
And unfortunately she still picked what she wanted, not what the peer wanted.
Okay, then we trained the next trial type and she learned it really fast.
And the next trial type and she learned it really fast.
Note that the duration that it took to acquire each trial type was shorter each time.
It's sort of an overarching skill, right?
A more generalized skill that she's getting.
She gets better at it so it's faster and faster to acquire the new examples each time.
Unfortunately still, even when she's got all the right verbal behavior, she's still not picking what the friend wants.
She's still picking what she wants.
So we thought, "Okay great." So we figured out how, we basically figured, oh and sorry, the last phase is combining all four trial types randomly, unpredictably.
And she's still doing great.
Now, unfortunately though she's not picking what the peer wanted.
And so what that shows is cool, we can teach the cognitive ability to predict how a friend is gonna feel and what we should do based on that.
And that doesn't necessarily change what you choose to do.
You could still choose to have your own stuff even if you understand what it matters to other people so that, "Okay, let's teach that too." So then we did, we basically did some kind of just prompting and reinforcement and prompt feeding to get the child to pick what the peer would want.
Like, "Okay, are you sure?
Maybe you should pick X, Y, and Z.
Like you want her to be happy, right?
Let's do that.
Can you do that?
Okay, cool." Reinforcer, fade out the prompts, fade out the reinforcers.
And by the end, she is actually making the decisions to play with the stuff that the peer wants.
So first the cognitive sort of verbal behavior bit then the actual like behavior management bit is how you might think of it.
Lot on that graph there, any questions?
I've got just a bunch of other studies to keep going through.
So I'm just keep going if no one raises their hand.
Okay, and that one we're writing up for publication now.
We haven't submitted it yet.
This study is on teaching children to detect when peers are lying to them for the purposes of bullying.
Well, for the purposes to avoid getting bullied.
And so unfortunately there's a lot of research showing folks with ASD have difficulty with detecting lies and difficulty with telling lies effectively.
And of course, sometimes it matters to lie, right?
You don't always wanna tell someone what you think of them if it's rude or whatever, right?
Socially inappropriate, sometimes you should keep your mouth shut, right?
So both using lies and also responding to lies effectively a lot of folks on the spectrum have a hard time with, because they're overly literal.
They wanna tell the truth.
And man, you can't really blame them for that, right?
Like I actually really appreciate that.
I think us neuro-typical folks are the ones that got it wrong, right?
Like it would be a lot cooler if people could just tell the truth.
But unfortunately that's not the social environment that we live in.
And so this study actually came out of a request from a mom of a client at our clinic.
And she said, "You know, my kid is getting bullied on the playground and it's kind of like low intensity bullying.
And it's where the bullies are just lying to him.
And they're telling them things like, 'You have to give me that ball because it's Tuesday.
And on Tuesdays I get the ball.'" And this poor client didn't get it that it was a lie.
He just thought, "Well, he wouldn't say it if it wasn't true I guess, okay.
You know, give him the ball." Or so it was to take possessions and to exclude.
So, you know, on some other day maybe the same bully or a different one would tell the kid, "Hey, you can't play this game because only girls are allowed to play this game.
And the teacher said, so, so get out, leave." And so he'd walk away sad and get excluded socially.
So no one was beating this kid up, but it was pretty awful actually socially, right?
To get excluded and to have your stuff taken because you're kind of gullible was essentially what it came down to.
So mom asked, "Hey, is this something you could teach?" And I thought, "I don't know, I have no idea, but sure.
It's a bunch of language talking to yourself about how likely it is that what someone else is saying is true or untrue.
Should be able to teach that." Let's use multiple exemplar training.
So the way we did this was we just did practice.
Sounds mean, but we would just lie to the kid and then measure how the kid responded.
And then immediately you stop and go, "Wait a minute, are you kidding me?
You think I was telling the truth?
Think about that, that can't be right.
There's no way that you can't have that hamburger because the moon is yellow or something like whatever, it's crazy right, no way.
What you to do, you say no." And then we prompt the kid to say no.
Say, "All right, cool, good, good," you know, and we back off.
And we weren't doing anything mean obviously to the kid, we're doing stuff that was very low intensity but we were lying to the kid and we wouldn't always lie because then of course we're just teaching the kid that the ABA therapist is a liar, right?
Like that's not the skill that we're trying to establish.
The skill we're trying to establish is a very subtle discrimination skill where the kid themselves identify novel untrained lies.
Is the person telling the truth or not?
And the two types of lies that we came up with were just the same ones that the kid was being bullied on, was taking stuff away from him and excluding him.
We did this with the one child, that worked so great that we did it with two more and we did a multiple baseline across three and then published it a few years ago.
The teaching trials were run by the ABA therapists, but then we also included peer probes to make sure that the students were actually able to do this for real with peers, and that it wasn't just only when the ABA therapist lies I can tell, right?
That's what really matters.
Like we're not gonna lie to him in the future.
It matters that he can actually figure it out with peers.
So the circles were sessions conducted by the ABA therapist and the triangles were peer probes where we'd bring in a peer confederate.
Usually it was like a neighbor or a cousin or a sibling or something.
And the ABA therapist would take them aside when the student is not aware of it and say, "Hey, we're gonna need you to lie to him, okay?
Here's the lie, can you do that for me?" The kid would be like, "Huh?" We're like, "It's okay, don't worry, we'll tell him, you know.
Like you're not gonna hurt him, don't worry." But we do recruit them to help out and they would help out.
And there was the same thing that they'd make up some lie about why the kid couldn't, like we'd play a board game and the peer would tell the student, "I'm sorry, you don't get to play because today is Sunday or Thursday," or whatever.
Okay, then what we're looking for is, oh, and we do new lies every time, not the same lines over and over, right?
It was always to take stuff away or to exclude the kid.
Those two facets were consistent, randomly rotating between those.
But the specific example of the lies were different every time.
Again, we're not trying to get kids to memorize.
And so what we wanted to see was the first trial, the first time a specific new lie was told, did the kid detect it correctly?
And like I said, we interspersed lots of real truth telling also.
So there's no way, the kid didn't know he was working on lie detection, right?
The kid didn't know this was an ABA program, we were just playing on breaks.
And so what you can see is pretty rapidly we got up to a 100% correct with first trial lies.
So the first time was implemented, new examples were implemented.
And then we tried new ones that had never been done before.
That's this triangle right there.
And then we also implemented the ones from baseline that we never did any direct training on.
And this was after any feedback.
So during the training phase, we tell the kid, if he messed up, like on any of these trials where he got it wrong, we go, "Wait, freeze, what are you doing dude?
That's obviously a lie.
Let's talk about why it's a lie.
Okay, cool." In the last phase, the post phase, there was no more of that.
So if he didn't detect the lie accurately, okay, just let it be.
If that was gonna be a problem then we would have given feedback.
But as you could see, they did pretty well.
And then we did, what did we do?
One month follow-up for all three of the kids.
And they did pretty darn well.
Jason and Lafayette, the top two were at a 100% correct still during post and at, or no sorry.
Jason went down to 75% correct during follow-up, but that's still pretty good.
And Eric, same thing, not quite as solid as during intervention, but still pretty darn good.
And I don't think that someone on the spectrum needs to be a 100% perfect with detecting lies.
I mean, I'm not, are you?
No, of course not, right?
But as long as it's most of the time they're able to detect this stuff, hopefully we're thinking that is a socially meaningful improvement.
And by the way, we did feel bad when we were running baseline.
We did not like lying to the kid that many times.
And I forget exactly how many trials there are per data point, but not very many for the obvious reason that we didn't wanna lie to the kid a whole bunch.
I think it was maybe three lies or something or four lies per data point.
I think it's four lies for each data point.
Am I a big meanie?
(audience laughs) I think it was worth it, right?
And I mean, what would really would have been cool is if we then snuck into the playground and watch to see if the kid's standing up to the bullies, that would have been awesome.
We didn't do that.
We weren't quite sure how to do it.
Or if we had permission to do it, should have tried to figure it out, but we weren't able to do that.
- [Woman] My question (indistinct).
Oh, with a mom?
Yeah, for sure.
And she, yeah, yeah.
She, yeah, yeah for sure.
I don't usually report that, 'cause it's not like real data, but yeah, yeah.
The mom was like, "Oh, thank goodness, you like saved her life." Yeah, totally.
- [Woman] (indistinct) That was another thing we were concerned about, right?
No, no problem.
Yeah, and there's another study that I'm not gonna show where we did teach, intentionally teach.
Well, actually this one is, actually this is relevant.
Another study though where we taught socially appropriate white lies.
So someone would come in with a ridiculous haircut and in baseline the student would say, "That's ugly, you know?" And so the treatment was, "Well, how about don't say that?
How about say something else?
Like change the topic, or just say cool, or just don't say anything, right?" Or we'd have like grandma or somebody come and give them a present that they really hated, you know, like a Justin Bieber record or something like that.
And, oh sorry, is he Canadian?
(audience laughs) That was a bad example.
But that really was a real example from, this particular student hated Justin Bieber.
So we really did give him a Justin Bieber CD to teach him that you just smile and say, "Okay, cool, thanks," and change the topic, rather than saying, "I hate this, don't give this to me." And we were concerned with that study too.
Is this gonna now over-generalize to where the kids' just lying all over the place and it didn't for any of those kids.
I could see if you spent too much time working online, you might see some of that.
But also that is developmentally appropriate, right?
Typical kids do that too.
They lie a little too much when they first learn how.
We tell them, "Don't do that." Maybe we mildly punish it.
Or maybe we just say, "That makes mommy sad, don't lie," or whatever right.
And then it decreases.
So even if you did see a little over-generalization that would actually be okay.
Like over-generalization shows it's working, like they're learning, you know?
You just have to make sure that you fix it.
Let's talk about playing tricks.
This is one of my favorite studies.
We haven't submitted this for publication yet.
We're just finishing it up.
We finished two kids.
We're running one, a third kid now.
But a student of mine, Megan St.
Claire was working with a client who, really smart kid, super verbal, but really was bad at perspective taking.
So didn't, things like if he was playing hide and seek, he'd leave most of his body sticking out of the closet.
He just didn't get that what he sees is different from what other people see.
He didn't understand that his knowledge was different from other people's knowledge.
And so Megan thought up this great way of teaching that, teaching that your knowledge is different from other people's knowledge.
What you see is different from what other people see and that what you see leads you to know stuff, right?
So if other people haven't seen something or heard it, they don't know the same thing you do because you have seen it and heard it.
And so that's straight out of like the Sally Anne task on theory of mind stuff.
But this was a way that we came up with, or that Megan came up with to try to teach all of those skills.
So she thought it would be really funny to teach our students with autism to play friendly pranks on other people, on family members and peers.
So when you think about it, successfully executing a trick, a friendly trick, involves identifying what other people know and that that's different from what you know.
Identifying behaviors that you'll do to prevent other people from knowing what trick you're gonna play on them.
Doing something that the other person will think is fun.
We're not talking about mean tricks, right?
That'll hurt the person's feelings, but something fun, and preventing them from knowing that you're doing it.
And then executing all of that in a way that maintains the deception that doesn't give the trick away.
And that is fun and friendly.
So we thought, "Okay cool.
Whole bunch of perspective taking, and we're gonna have some fun." So these are kids with autism again, who needed perspective taking, highly verbal probably like near average intelligence but really poor on perspective taking in particular.
They couldn't keep secrets and they couldn't keep surprises.
So if they bought their mom a present they'd say, "I bought you a present momentum, it's whatever." Like, okay.
Okay, so Megan created a task analysis of what it takes to do a good trick.
Oh, and by the way, we didn't want them to just be able to do a trick.
We wanted to teach these kids to be able to create their own tricks because we thought that would be even cooler and even more difficult to teach.
So why not try it?
So the task analysis is this.
The student had to create their own trick, think of their own trick, and tell it to us.
They had to describe it and explain why it's a trick.
That explanation had to explain what the victim of the trick thinks and what the truth is.
And then they had to execute the trick without giving it away.
And they had to end the trick appropriately like by saying, "Gotcha, or tricked ya." It's not polite to do a trick and then just walk away, Right?
Okay, we initially we had a few sessions in a row where we teach the same trick, but then we moved on to new tricks every session where we told the kid, "Look sorry, kid, you got to think of something new.
We can't do the same one we did yesterday." "Why?" "Because it's not really a trick anymore if it's always the same thing.
If mom knows you're always hiding in the closet, 'cause it's always the same thing, it's not really a trick anymore." So new tricks every day.
And we scored the percent correct on those four pieces, those four steps of the behavior.
And we did three tricks per ABA session.
And the sessions are usually about two hours long.
So we wouldn't do three in a row.
Otherwise the victim would know like, "Okay, it's trick time," you know?
Because we were actually trying to trick the victims for real.
And so the victims where whoever else happened to be in the house.
So mom definitely got tricked a lo, but siblings got tricked and nannies.
Anybody else who was there, workers in the house and construction workers, anybody.
Anybody was fair game.
And the types of tricks were things like, one of my favorite, this one could be a little bit mean, I don't know.
One of my favorites was asking mom," Hey, are you thirsty?
Do you want some water?" If she says, "Yes," you go into the kitchen, get some water and put a bunch of salt in it and stir it around and the salt completely dissolves, it's invisible, right?
If you stir it and then you give it to mom, "And man, oh, it doesn't hurt, but it tastes so disgusting." And you know, she's not gonna drink a lot.
So just taste enough and go, "Ah, you know," but it's hilarious.
And I will say we never actually hurt someone's feelings substantially.
(audience laughs) Another example was disappearing ink.
Disappearing ink's really fun.
And so the student would like pretend like, "Oh, hey mom, there's something wrong with my pan and go, 'Oops." And then spray ink all over mom's blouse.
And it's like bright blue ink on a white blouse.
But it really does disappear.
Like that's what it does, you know?
So it doesn't ruin the clothing.
One of my favorite examples, this was one of the coolest examples, and it was near the end when it clearly showed like this works.
The student's big brother just got a brand new Chevy Camaro.
Really nice and really good condition, brand new, parked in the driveway.
And the student says to me in front of his brother, "Hey, let's go outside and play catch." I said, "Okay, cool." So we go outside, we go in the front and the the brother can't see us.
And then we make a loud noise and then the student runs inside, He goes, "Oh, I'm so sorry.
We were playing catch and I think we dented your car." And the brother goes, "What?" And runs out there, you know?
And the student goes along with him and runs out and looks at his car and the student goes, "Gotcha." (audience laughs) And the brother totally bought it.
He believed it, it was so awesome.
(audience laughs) And you know, I mean he's, and I have to say like, he's actually proud of his little brother, his little brother with autism is actually able to play a trick now and pull it off.
And it was funny after he got over being mad for a second.
So here's the data from the first two kids.
In baseline this is the percentage of those four steps of tricking that are executed accurately without prompting.
And the kids couldn't, the first kid couldn't do it at all.
The second kid could kind of come up with some of the elements of a trick sometimes.
Like he could tell you, "Well, I'm gonna hide in the closet or something." Like, okay, so maybe we'd score that as a correct trick but then like he couldn't explain why it was a trick.
And then he certainly couldn't go do it.
And in fact, during baseline where the kid on top, what he would do is when he, he might even come up with a trick and say, "Okay, cool, let's do it." He walk over to mom and say, "I'm gonna check you mom." And it was super cute, but obviously like didn't get the perspective taking piece, right?
So the first three sessions were acquisition trials where we allowed the student to just do the same thing he did yesterday if he wanted to, right?
Repeat the same one until he got reasonably good at it.
And then after that the white data points were, there always had to be a new trick every time.
And it was kind of hard for these students to create their own tricks.
It's actually kind of hard for us to create enough new tricks to even prompt correctly.
But by the end, both of the kids were tricking their family members a 100% correct without any help.
We're still there hanging out saying, "All right, you wanna trick?
Let's do a trick.
What are we gonna do?
Like, what do you think?" But then a 100% of the rest of the steps were done independently, correctly.
They weren't giving it away.
They weren't giggling, you know, when they're trying to execute it and they're doing it independently, and it was a lot of fun.
Okay, let's talk non-literal language.
So I guess the larger point there with the perspective taking stuff is putting yourself in somebody else's shoes, like everything else we've been talking about today is a larger overarching skill that needs to be practiced a lot across multiple exemplars, probably broken down into lot of steps, just like everything else we do in ABA and taught sequentially.
But again, make it fun.
Lots and lots of practice opportunities across a lot of situations.
I don't think there's anything specific about playing tricks or about the desires thing that I showed.
It's just more examples of how complex verbal behavior in particular perspective taking can be taught.
All right, let's talk non-literal language.
This is another one of my areas of interest.
Non-literal language is tough for a lot of folks on the spectrum because a lot of folks on the spectrum are overly literal.
They just wanna say it the way it is.
And again, I actually really can't blame them.
Again I think folks who just wanna tell the truth and just say it like it is, is really refreshing.
Like we could use a lot more of that.
When you think about it, the reason why that doesn't work for us is because we get all bent out of shape, right?
We get all frustrated or offended or insecure or whatever else.
If we were able to just be flexible and just hear what we need to hear, it actually wouldn't be such a big deal.
We wouldn't need to teach all of these skills, right?
Like folks on the spectrum could just be overly literal and people wouldn't have to get offended.
But you try teaching that to the rest of society.
That's not gonna happen, right?
So given society that we all live in, we all do indeed need to learn some non-literal language and be able to detect it in other people and respond appropriately and also use it ourselves.
We cannot just be overly literal all the time.
It's not gonna work socially.
All right, so metaphors is one type of non-literal language.
And it's a strange thing that makes sense that a lot of folks on the spectrum have a tough time with it because it's when we call something other than what it really is.
So we call a thing by a name other than what it really is.
It's not that thing, we're calling it something else.
Like when I say like you know, Halifax is an ice box.
It's not literally an ice box, it's just cold here.
It's not literally an ice box, okay.
Or, you know, you name it.
I mean, there's research actually that shows that metaphorical language is ubiquitous throughout all normal conversational interactions or I shouldn't say normal, just daily conversational interactions, we're constantly using metaphors.
Even the term of reinforcement, positive reinforcement, is a metaphor.
We're not literally physically making a behavior stronger.
We're making it happen more in the future.
Like reinforcement is what you do with like concrete and rebar and two by fours and stuff to make something physic, right?
It comes from that.
So like everything we do on a daily basis involves, everything we talk about on a daily basis involves metaphors.
And so if you have a tough time understanding people's use of metaphors, it's gonna be a problem socially.
And there is research showing a lot of folks on the spectrum have a tough time with it.
So we used, you guessed it, multiple exemplar training to teach kids on the spectrum to identify and decode sort of the reasoning piece of identify what other people mean when they use a metaphor.
So to do this, we would teach, or sorry, we would state short stories that talked about three different properties of a situation.
And then we would give the student a metaphor and then we'd ask him what we mean by that.
And then we just multiple exemplar training, talk them through it and then faded out the prompting.
So I think I have an example here.
Yeah, so one example is, all right, so I knew this kid that I went to school with when I was young, he was really strong.
He always wore yellow and he stayed up really late at night.
If I told ya that he was a banana, what would I mean by that?
Always wearing yellow.
If I told you he was an owl, why would I say that?
Stays up late at night.
What if I told you he was a superhero?
Why would I call him that?
Because he's really strong, okay.
So these are totally non-functional made up, arbitrary, examples of metaphors.
And the reason why we didn't teach socially conventional metaphors is because we don't care about teaching memorization of specific metaphors.
We care about establishing the cognitive ability to figure out what metaphors mean when you contact new ones in the future that you've never been directly trained.
So, we presented a bunch of these over and over and over and just walked them through it with prompting, leading prompts like I said.
Where we kind of talk the person through, coming up with a solution.
Prompting and reinforcement, fade out the prompting.
For folks for whom just the vocal prompting and reinforcement wasn't enough, we use a visual prompt as well.
The visual prompt was like this.
We'd have a piece of paper or a worksheet or a dry erase board or whatever that looks something like this.
We'd present the story to the kid.
And then we'd write the two main pieces of the story on there.
So for that example I just gave you, we'd write boy on the left and owl on the right.
And then there'd be a column underneath, right?
So we say, "Okay, what did I tell you about the boy?" We have the student write it down.
If the student didn't know how to write we would write and we'd have them read it.
"What did I tell you about the boy?
Well, I told you he's strong.
I told you he stays up late at night.
I told you he wears yellow." Okay, now let's practice each one of the metaphors.
"If I told you he's an owl, what would I mean by that?
Well, let's talk about owls.
Tell me something about owls, and whatever the student would say about owls, We'd write down.
So maybe the student would say, "Well, an owl's a bird.
Lives in trees.
Stays up late at night.
Oh wait, what did you say?
Cool, okay now draw an arrow there that connects those two things.
He stays up late at night.
And I told you the boy stays up late at night, so there you go, there's your answer.
Does that make sense?" The kid would go, "Huh, but he's not an owl, he's a boy.
He's a human, right?" Overly literal, right?
Say, "Okay, cool, let's do another example." And then we just walk him through another example, and another example, another example, using the prompting, fading out the prompting.
We use this visual prompt in a least to most fashion, meaning the first time we presented a trial of a metaphor, we wouldn't use the prompt.
We'd let the kid do it on their own or try to.
If they got it wrong, then we implemented the prompt.
The reason why we did that was because the prompt would then fade itself out by itself.
We wouldn't have to fade the prompt out.
Because when the kids started to get it correct on their own, the prompt wouldn't be done.
Does that make sense?
For interest in the geeky relational frame theory, ABA analysis of the complex verbal relational responding involved in metaphors, you can take a look at this.
We don't have time.
Let me get to the graph.
So this is a multiple baseline across three students.
Two of the students required the visual prompt, one of them didn't.
So let's look at the bottom graph first.
We've got a percent correct here down near zero, zero to a 100 on the vertical axis.
And this is baseline.
And these are metaphors that we did not use in training.
Then we have the teaching phase.
In the teaching phase we plotted both the first trial data, which is the first time a metaphor was presented and all the rest of the data which is not the very first time a metaphor was presented.
And each session we would do three new metaphors and three metaphors from the previous session.
So there was some repeated practice but always half new, half old, every data point.
And so you could tell like if you've paid attention over the last six and a half hours, we've been kind of playing around with that, with multiple exemplar training.
Do we always do something new every session?
Or do we stick with the same thing until it's mastered and then do something new?
Or do we do kind of a combination?
And so for this particular setting we arbitrarily chose, you know, I think we wanna do half brand new every day and half from yesterday.
So that there's some novelty, but not too much.
I don't know what the right combination is, but it's just one of those variables that you can play with when you're designing programs for your students.
And if it isn't working, step back and go back to repeated practice.
For these kids it worked pretty well.
For this first kid he got up to a high level of accuracy with first trial data.
And then we went to the post-training phase.
The post-training phase we repeated the metaphors from baseline that we had never directly trained and no feedback and no prompting.
And the kid did pretty darn well.
There's two sessions where he didn't do that so hot.
And those were sessions where we realized we did not choose the best metaphors.
So some of the metaphors we used in that for this, I think it was this particular, let's see, this particular session here was talking about snow covering everything like a blanket.
And these are kids that live in San Diego, California, and they're like six years old and they've never been to the snow.
So maybe that was kind of a stupid, poorly prepared example.
And that wasn't one of the ones that we directly trained.
So we're hoping that the kid would just figure it out, and that one not so much.
But overall compare the performance to baseline.
Take a look at that.
Take a look at that.
No practice, no exposure whatsoever, no prompting, no reinforcement.
They were able to just figure it out now.
For the other two kids they required just prompting and reinforcement alone wasn't enough.
We had to go to the visual prompts.
So see, notice they're doing okay.
But those first trials they weren't getting.
So it shows a lack of generalization to novel exemplars.
So, and even here, it's pretty good, but not good enough, was somewhere around 60% or so.
We wanna get them up closer to a 100.
So then we implemented the visual prompt with both of these participants, the top and the middle participant.
And the cool thing is, remember, this visual prompt fades itself out.
So if the kid is doing well, getting independent correct, that means they're doing well without the visual prompt.
And then the last data point, the solid black circle data point in each phase was completely novel metaphors that had never been tried at all.
And that worked well.
And then the post phase was the same ones from baseline.
Again, compare baseline to post, baseline to post, not perfect.
A couple errors here and there, but pretty darn well.
Here's the cool thing.
One of our happiest findings that unfortunately we didn't take data on and I kick myself to this to this day is, they started to create their own metaphors.
And so when they'd see the ABA therapist walked through the door, they'd say, "Hey, guess what?
Guess what I had?" Like here's one example and it's not perfect but you'll see that they're connecting the dots.
Students said, "So if I told you that I had lava for snack today at school, what would I be talking about?" And the ABA therapist is like, "I don't know, you had really hot soup or something?" He's like, "No, it was a strawberry, because it's red and lava's red, you know what I mean?" Right, and it's not a very good, you know, it's not the best performance, but they're getting the basic.
The basic thing there is it shares properties.
And that's what a metaphor is.
It's a different object that shares properties with the thing that you're actually describing.
And so, and we never even taught them that, never prompted it, never tried to establish the expressive component at all.
It just started happening on its own.
And now Anna Ramon Cortez is a PhD student at the University of Almeria in Spain.
And she's doing her dissertation now on teaching kids with autism to create their own metaphors expressively with as opposed to the basically the listener piece of the repertoire is what we taught.
Now she's extending that to getting kids to create their own novel metaphors.
Okay, let's keep steam rolling through non-literal language here.
We did a study on teaching kids with autism to identify and respond appropriately to sarcasm.
Sarcasm is the lowest form of humor.
It's where you say the opposite of what you literally mean.
And there is research showing a lot of folks on the spectrum have difficulty with this.
Again, it's non-literal language.
It doesn't make sense.
Why would you say the opposite?
Why wouldn't you just say what it is, right?
And so the kids in this study were kids where if someone made a sarcastic comment, they disagree with it, right?
So like let's say they're trying to play basketball and they miss and a peer says, "Wow, nice shot." They would say, "No, it wasn't a nice shot, I missed." Which doesn't work socially.
It makes them stand out, look silly.
It's an opportunity for them to get made fun of.
One of the participants, even his dad was super sarcastic, always cracking sarcastic jokes, and he didn't get it.
So mere exposure was not enough.
He needed actual training, like practice with feedback on the behavior not just exposure to the stimulator.
So this we published in 2013 and Angela Persicke was the lead author on this and basically same approach.
We came up with a bunch of examples of sarcastic comments, and we alternated them with sincere comments with the same stimuli.
And we would just prompt that student to respond in a reasonable way.
And what we defined as a correct response to a sarcastic comment is to not challenge it.
So challenging a sarcastic comment is almost always an incorrect response, 'cause it shows that you don't understand it.
So agreeing with a sarcastic comment works, or like a wink and a smile works, or just laughing works, or even saying nothing at all is pretty reasonable.
But you can't challenge a sarcastic comment, that doesn't work socially.
And so anything other than that, and I don't think we allowed them to just say nothing.
So any response other than that made any sense at all and that was not challenging it, and it was a reasonable response, we counted as a correct response.
So here's some examples of sarcastic comments.
Like let's say we're looking out the window, it's a beautiful, warm sunny day.
A sarcastic comment might be, "Oh, it's definitely gonna snow today." Whereas a sincere comment might be, "It's so warm outside, it's nice.
It's a beautiful day." Non-preferred food item is present.
We might say like, "Oh, you'd love to eat that all day every day, wouldn't you, yum?" Whereas a sincere comment might be, "Yeah, I know you don't like broccoli, it's gross," whatever.
I think one of the exemplars for one of the kids was like, there's a squished worm on the sidewalk.
And the sarcastic comment was, "Mm, yummy." And in baseline the kid was like, "No, no, don't eat it." You know?
(audience laughs) And then a sincere comment would be like, "Ooh, that's gross, look at the gross worm." Does that make sense?
So we're randomly alternating back and forth between just saying something sincere and saying something sarcastic.
And we had baseline same as before, low data points show low accuracy in responding to sarcastic comments.
The first few, the first three sessions we showed videos of stuff and then we made comments based on those videos.
I'm not sure why we even did that.
We should have just went straight to NVivo training, but Angela thought it'd be a good idea.
So we did it and it worked fine but I'm not sure that it was really necessary.
I think we probably could have gone just straight to stuff in real life.
But in any case, we did that.
And we went to Nvivo.
Again, we're looking for accurate responding to novel exemplars on the first trial.
And we saw that for all kids.
And then in the post-training phase, we went back to the baseline exemplar, same ones here.
Now they're responding accurately.
We added additional novel probes that had never been, additional novel sarcastic comments that had never been done anywhere else just to make sure they did great on that.
And then we did one, two and three month followup and they did great on that too.
And somewhere in there, I'm forgetting, it's in the article, but somewhere in there towards the end, we started going to novel locations too, just to make sure we're getting generalization.
So we'd go to Starbucks and the mom would joke to the kid that she's gonna get him a quadruple latte or something that he's not allowed to have, you know, and that was the sarcastic comment to see how he would respond.
And by the end, doing great, getting them all right.
Oh, and this one we were really concerned about over-generalization, 'cause nothing would be more annoying than if we got one of our students to be sarcastic all the time, constant sarcasm, right?
Be awful, so we were really concerned about that, really looking out for it and it didn't happen.
Okay, let's talk rule-governed behavior.
I promised that we would talk about it early on in the problem solving bit and in the planning bit and the inhibition bit.
So talking about the future is a huge part of all of those EF skills, talking to yourself about the future and what's gonna work and what isn't, big part.
Rule-governed behavior is by definition is behavior that occurs in response to a rule as if the person had contacted the contingencies described in that rule in the past, even though they have not.
So in more technical terms a rule is an antecedent description of contingencies that controls behavior as if the behavior had contacted those contingencies.
So for example, you're never going to drink fluid out of a bottle with a skull and crossbones on it, right?
That says poison.
Did you have to do that behavior in order to learn that discrimination?
And I could teach you a new rule right now.
That's whoever raises their hand first gets an extra continuing education unit.
Ready, set go.
And that, oh, it works, look at that.
Does that make sense?
So you've definitely never done that exact behavior in this presence in response to me saying that before, right?
Sure looks like it.
I mean, that looked like a perfectly trained response to an SD, but it can't be because by definition and a trained response to an SD is a behavior that's been reinforced in the presence of an SD in the past.
That is not that, rule-governed behavior is not that, it's emergent behavior.
It's novel behavior.
It's behavior that occurs as though you've contacted the contingencies described in the rule.
Cognitively speaking it's behavior that occurs because you know, you understand, you comprehend what you need to do because of the consequences that it's going to produce.
All the way back to the 19, I don't know, Skinner's writings in the 1950s or so or 60s, it was clear this is a really important repertoire behavior, super important, but we didn't know what the heck it was.
It didn't make sense.
It's not an SD, it's different from an SD.
Super relevant to everything we do obviously.
It's how knowledge is passed on from generation to generation.
Imagine if you yourself had to learn how to teach people with autism through pure direct contact with the consequences of your behavior, right?
Like what if you couldn't go and take a class and hear rules about what works and what doesn't work.
What if you had to discover the principle of positive reinforcement on your own just by rent, right?
That's what life would be like if we didn't have rule-governed behavior, super critical.
Before 2011 there were no studies on teaching people the ability to follow rules, like establishing that repertoire of rule-governed behavior in people who didn't have it.
There was lots of research on verbally able adults who have rule-governed behavior looking at the properties of rule-governed behavior.
But no research saying like, "Okay, how about folks with developmental disabilities who they can't actually talk about the future and have their behavior changed on the basis of that?" They just don't have that ability.
Can we establish that ability?
So we designed what we thought was the simplest possible first step in that program of research where we taught children with autism to follow rules that described an antecedent and a behavior, what is this?
This is the wrong, this study is wrong.
It says following novel rules describing antecedents and consequences, that's wrong.
Novel rules describing antecedents and behaviors.
So for example, so we made a big list of rules and each rule describe an antecedent and a behavior.
So here's an example of ones that we had in baseline, and for generalization, if this is orange, then touch your head.
So we'd hold up a picture card.
Sometimes it was orange, sometimes it wasn't.
The correct response when it's orange is, touch your head, right?
What if I hold up an apple and I say, "If this is orange, touch your head." The correct response would be, don't touch your head.
Okay, so it's like conditionality, right?
If then essentially.
And that's really what rule-governed behavior is about, is that relating, that conditional relating.
If this circumstance happens then I should do that behavior.
Or if I do this behavior, then I'll get that reinforcer.
So what we did was multiple exemplar training across dozens and dozens and dozens of these if then rules until we got generalization to untrained if then rules.
So these are kids that could already tell you the objects on the picture cards.
And they could already do all of the receptive instructions.
They could touch their head.
They could stomp their feet.
They could clap their hands but they couldn't do it conditionally, if this, then that.
So like they couldn't play Simon Says for example, right?
Simon Says is all about that.
If the person says, Simon Says, do the behavior, right?
If they don't say Simon Says, don't do the behavior.
They couldn't do that.
So here's the graph.
We actually did two experiments in this study across six kids.
And I'm not gonna belabor the details.
But basically what it showed was, if we had to tweak the procedure a few ways here and there for different participants, but across all six kids, all six of them were able to learn that basic sort of component repertoire of rule following.
And they were able to generalize to completely novel and untrained rules.
It just took a lot of exemplars.
This kid was fast.
That's accurate responding to untrained rules after only learning I think about six rules.
These kids took, I don't know, they had to learn 30 or something or 40 exemplars before they finally got it, but they did get it.
So in the end you could hold up, you could describe any antecedent and any behavior conditionally and they could follow it.
And then we did it again and with some minor tweaks.
Okay, Sarah Weimer at, where was she at?
Georgia State University I think, extended what we did to rules that involved behaviors and consequences.
And so what she wanted to see was, could we get kids who do not have the ability to relate their own behavior to a consequence, either preferred or non-preferred consequence, can we teach them that ability?
The same thing.
If I do a behavior, I'll get a particular consequence.
So for example, let's say you hate broccoli.
If you clap your hands, you're gonna get broccoli.
What's the correct response?
Don't clap your hands, right?
What if you love broccoli?
Correct response would be clap your hands if you want the broccoli, right?
Okay, so she did same old stuff, multiple exemplar training.
And she did preference assessments at the beginning of each session to identify what the kid did actually want and not want in that session.
So she put out a bunch of stimuli, identify stuff the kid wanted, identify stuff the kids didn't want.
And then those were the stimuli that she used.
Here's the results.
These kids, the first two kids learned like too fast.
It looks like they almost had this already but kind of weren't able to demonstrate it maybe in baseline.
I'm not sure.
Or maybe they were just that ready to learn that repertoire.
All it needed was repeated practice and big powerful reinforcers.
In this child required two sets of exemplars, so it took a little bit longer.
But all three kids were able to get these repertoires pretty fast.
So just a few baby steps into this literature but, and it might seem kind of obvious, like, yeah okay, basically taught them Simon Says, who cares?
We basically taught them like what every kid should know, like do your homework first, then you get your- Eat your vegetables, then you get your dessert, right?
Like behavior consequence, behavior consequence.
That is kind of obvious.
It is kind of simple.
And these kids couldn't do it before.
And if they don't get specific intervention, they might not pick up those repertoires just through natural contact with the verbal community.
It might not happen on its own.
And of course it also matters very much that we have research that actually shows what works and what doesn't work.
So that's why we're doing the studies.
It is kind of overly simplistic first steps.
What we're really interested in is, can kids figure out and follow, derive their own rules and follow other people's rules that specify really long delays.
So for example, you and I would be doing a lot better if we would follow rules such as don't eat X, Y, and Z because you could have a heart attack in 20 years.
Some of those long-term rules, people are doing better with, like smoking for example Has smoking gone way down in Canada?
It's gone way down in the U.S.
thank goodness, it's great.
And actually it probably has nothing to do with rule-governed behavior, it's probably more the cost of cigarettes is the number one thing in the U.S.
that decreased smoking.
But in any case, long-term delays matter a lot.
Like wouldn't it be great if your students were thinking more about what's gonna happen next week or next semester based on their behavior right now, rather than I don't want this in this moment or I do want this in this moment.
How about if I do this behavior overall, I'm gonna do better in school.
Or overall I'll have more friends.
All of us behave with respect to consequences that may never ever happen.
A lot of folks play the lottery constantly, even though there's almost no chance they're ever gonna win.
But their behavior is oriented towards that consequence.
A lot of folks' behavior is affected by what they believe is gonna happen to them in the afterlife, which I'm not even commenting on whether or not that's real or fake.
It doesn't matter.
The point is, that consequence can never affect their behavior.
By definition that consequence can never affect their behavior as a consequence.
It can only affect their behavior as a perceived future consequence, right?
Which is what all of this rule-governed behavior stuff is about.
So very relevant to everything we do.
So there's a lot more need for research in this area.
I'm gonna wrap things up here.
I highly recommend checking out the PEAK curriculum series.
It's four curriculum books.
I mentioned them already, I wanna mention them again.
Four curriculum books published by Mark Dixon and his graduate students.
They've put in just a Herculean effort over the last three or four years to create the biggest most comprehensive curriculum for autism in existence that is based on the latest research in stimulus equivalence and relational frame theory.
So it's very relevant to a lot of these complex repertoires, a lot of these complex higher order conceptual or cognitive skills that we've been talking about today.
Definitely worth checking out.
It also happens to be the curriculum for which there's the most research supporting it.
So he has been working his graduate students like slaves to publish so much research.
This slide's already outdated, but when I made this slide six months ago there were 12 studies already published on the validity of the PEAK curriculum.
And 15 studies already published on the effectiveness for teaching the skills, using the procedures in the curriculum.
That's more than the VB-MAPP.
That's more than any of the other existing curricula out there for autism.
And there's, I don't know, I think another 15 or so that are impressed, that are being reviewed.
So that'll probably be published in the next year.
So tons of research supporting the use of the PEAK.
So check it out.
I'm not saying it's a replacement for everything else you already do but it'll definitely expand your horizons in terms of the types of skills and the complexity of skills to target.
Especially the third and fourth volume.
Volume one and volume two, the direct training volume and the second volume is called the generalization volumes.
So the direct training volume and the generalization volume are more what you're already familiar with, the more sort of standard ABA stuff.
The third volume, which is the stimulus equivalence volume and the fourth volume, which is the transformation of function volume are very very, forward-thinking complex, crazy stuff that's definitely worth checking out.
And again, I have no monetary connection to Dixon or his stuff at all.
It's just good stuff that's worth looking at.
We don't have time for this, but I highly recommend checking out mindfulness, it'll make your life better and it'll make your students' lives better.
Is it becoming popular in education now in Canada?
Good, I hope that it's being done well.
It's actually kind of hard to screw up.
It's actually a pretty simple thing.
Mindfulness training has been practiced for literally thousands of years in Asia.
It's actually not that hard to do.
It's pretty straightforward.
I recommend checking it out.
Don't have time to talk about it today any more than that, other than to say, there's been several studies now published by Nirbhay Singh and his group showing that self-injury, aggression and non-compliance decrease, aggression goes down, social behavior goes up when you do mindfulness-based parent training for parents of kids with autism.
And check this out.
In 2004, they published a study where they did mindfulness training for staff in group homes for individuals with developmental disabilities.
But the outcome variable they were interested in wasn't what the staff did.
It was happiness demonstrated by the clients who lived in the group homes.
And what they found was when they did mindfulness training for the staff, the residents with disabilities started to smile more often and laugh more often.
So that's pretty cool.
I mean, what that looks like to me is, the staff were spending less time worrying about their own stuff and more time being present and engaged with the residents, which made them all have more fun.
So really cool stuff.
And acceptance and commitment therapy.
Definitely no time to go into this.
Very, very cool stuff.
I had a cool graph to show you.
This is good behaviors on the part of parents of kids with autism going up after acceptance and commitment training-based parent training.
And there's some resources out there.
I highly recommend you checking out acceptance and commitment therapy.
There's webinars and there's books and things like that.
I think I'm gonna wrap this up.
So to conclude, the big, big ticket recommendation here is focusing on generalization.
Always doing multiple exemplar training until you get generalization to untrained examples of the skill, that is the mastery criteria.
Generalization to untrained examples of the skill.
Whatever the skill is that you're teaching, if it's one of these higher order complex skills.
It's not an afterthought, it's the main focus.
So, I mean, I don't want to overstress it, but so far literally every single area of cognitive functioning that we've tried to take an ABA approach to improving has worked, every single one.
And I'm not saying that every person is capable of anything, not everyone's gonna be Einstein.
You're never gonna make me be a good dancer, that's for sure.
But everyone's behavior is capable of getting a little bit better.
So wherever your students are at in terms of their perspective taking abilities, their executive function skills and their non-literal language skills, they can improve a little bit.
They can be pushed a little bit.
And so far it works pretty much every time to make some measurable improvement.
Really, regardless of how complex these skills are.
So thank you very much.
And I included some resources here to check out if you're interested.