 Hey everyone, welcome back. It's Veronica Howard. So this unit is going to be a little bit longer. We're going to talk about the history of behavior analysis in education. And I specifically choose to include this in the section where we're talking about the additional use of prompts, where we're talking about establishing complex repertoire, because that's really what education is. We're trying to teach concepts, we're trying to teach principles, we're trying to teach folks to tell a difference between things. And it's a lot of additional stimuli that we add to that effect. So it fits really nicely here. Now remember that behavior analysts have had a long, long history in education. Essentially, we're trying to teach behavior, right? That's what education is. In fact, Skinner's daughter, his oldest daughter, Julie Vargas, has gone on to be a leading expert in the field of empirically validated educational techniques. She even has this book that's really amazing called Behavior Analysis for Effective Teaching. So if you are interested in education, I would encourage you to check out this particular book. It's very good. And it talks a lot about the ways in which behavior analysis can be used to effectively teach a wide range of learners. The primary subject matter for behavior analysis really is the acquisition and maintenance of lots of different behaviors. And when we talk about behavior analysts in education, we're typically looking at three major ways in which we use that training these behavioral principles to foster new knowledge, new behavior, new skills. So we have something called Programmed Instruction, which is a very careful application. It does involve programming, just like the term says, using those additional scientists in stimuli to foster generalization. We have Direct Instruction, which is a highly scripted form of education. It involves a high rate of response, lots of reinforcement, and then we have something called Generative Instruction, which is kind of a second generation, more advanced, really exceptional form of education that's used to really effective means for the most challenging students or students who need the most support. Let's talk about Programmed Instruction first. This was Skinner's major contribution to the field of education, and he was incredibly dissatisfied with the educational practices that he saw when he was visiting his daughters at their school. So in 1953, he goes, he sees that the school's not really offering lots of opportunities to learn the skills, and he said there's got to be a better way than this. So he comes up with this prototype teaching machine that I'm showing you here. This is one of the very first Skinner Programmed Instruction teaching machines. Now in this particular approach to education, what we have here is there's a bunch of questions that are on a roll of paper, so pre-printed questions, and the questions will be on this big roll, this cartridge that you insert into the machine, you roll the paper up. Questions appear in this window, and then the learner writes their response. They kind of put in their answer there. This was, again, super rudimentary. We didn't have computers in the 50s, so this was one way for folks to do that. They would write their answer, and they would kind of roll up the question. The written response was an active behavior that we could then later measure. We could measure this using permanent product recording, looking to see the results of the behavior, whether or not the question was answered, and the mechanism also didn't allow the learner to roll back, so they could only move forward. They couldn't see the answer and then roll back. So they write down their answer, and then they roll forward to actually see the correct response. So this is fascinating, because it's one of the first opportunities that we have for a kind of automated self-paced instruction. You see that immediate reinforcement, that immediate feedback of whether or not you got the answer correct. If you got the answer correct, whoo, and if you didn't get the answer correct, at least you knew what it was immediately, so that you weren't forgetting it or having to go back and review it. Now, program instruction talks about a lot of these procedures, but you may notice that Miller's textbook does those. So if you've done any of the optional readings for this particular class, then you'll notice there's lots of questions and lots of blanks for folks to write answers. Miller's textbook uses this program instruction paradigm, and there's lots of repetition of these terms. There's lots of use of generalization to promote understanding of concepts, so we're trying to see what are all the things that fit into a category. We're using lots of multiple exemplar training to foster that generalization. And we're also using prompts to help delineate between different concepts. So every time you see that Miller has bolded or underlined certain keywords, he's adding that additional prompt, which then gets faded over time, so that it's easier to tell the difference between the pertinent features of that particular behavioral concept. So that's, you can see the pros and cons here. As I've mentioned before, people either love program instruction, write it self-paced, or it allows you to automate and do your own business. People also kind of hate program instruction because it's a lot of work. It feels like a lot of busy work, but that repetition is all part of the multiple exemplar training. So you can see where this particular approach to behavioral education can either be loved or reviled. Let's switch to direct instruction. When we talk about direct instruction, remember this is highly scripted. It's got these five major components. First, there has to be active learning. In all of these behavioral approaches, you have to have some sort of active response, something that you're doing regularly that you're getting feedback on. In direct instruction, you also have a high speed, fast, paced repetition. So typically in direct instruction sessions, there's going to be like a question, answer, question, answer, question, answer. And what you start to do through this is by presenting lots of different questions. You can extract knowledge from examples and also non-examples. So again, we're building those concepts through lots of practice and learning those different ideas. The challenge with direct instruction is that the teacher's role here is highly scripted. And so the teacher is presented with a script of how to deliver the instruction. They're given lots of questions. They're given the prompts, the specific prompts to use. And finally, there's frequent reinforcement. If the people are getting it right, yes, do it. And if they're not getting it right, you give that immediate correction you said, no, not that this instead. And then you practice it with the correct answer. There is a wonderful video of this that I'll put into the study guide of how this fast-paced, really fun direct instruction works. And direct instruction was developed actually as a result of some federal intervention. There was something called project follow through. And it was this huge study in sort of the 1960s and 70s. In this particular study, there were 700,000 children in kindergarten through third grade. And through 1980, this study was run for quite a long time. It was very, very expensive. But the purpose here is that we're trying to find something that's going to work effectively for folks, even those who have a significant delay, people who have been significantly challenged by education, either because the access to education that they've had is really bad, the quality of the education that they've had access to is really bad, or because they're delayed in some other way. Now, in project follow through, this huge 13-year study, the purpose of this was to try to figure out which of the different educational strategies are most effective. Three different approaches were evaluated. First, there was an affective skill approach, affective mean, like, how do you feel about your learnings? This is another way of saying a kind of self-esteem approach. The idea of an affective skill approach is that if you're confident in your learning, you'll perform better. We'll see what the data say. There's also a cognitive skills approach. So cognitive skills approach refers to this idea that if you can describe the skill that you're performing, it will make it more effective for you to be able to do that skill. Again, we'll go back to the data in a second. And then finally, the basic skills approach. And these are programmed instruction, which we talked about before, the applied behavior analysis teaching, and this new direct instruction that was developed and then was evaluated. And to make sure that there was a limitation on the bias from the researchers, the government contracted with Stanford Research Institute and the Stanford Research Institute analyzed and partnered with these ABT associates, these neutral third parties for reliability observers to make sure that there was agreement between there and that the quality of the study was not jeopardized. Because again, we're talking about billions of dollars. We're trying to find the most effective approach to education. We don't want to mess this up. Let's talk about the results of those different approaches. So the first one is that that self esteem or that affective approach. And what I'm showing you on the y axis is the amount of progress per year. So essentially, you want to see zero or above because it indicates that you're making a full year of progress. If you see less than zero, it means that you had poor learning outcomes. And what they were evaluating were three major areas, basic skills, like the ability to actually perform it, cognitive skills, how people could describe it and affective skills, how the learner felt or how they reported they felt about their competence and their fluency in that area. The problem with the self esteem approaches is that we see that across the board, it resulted in poor learning outcomes. The lowest outcomes were in basic skills across the board and Bank Street, Responsive Education and Open Education. What was most odd about this was that these were specifically designed to increase folks' confidence about their learning. And we actually see that it seems to have decreased self esteem. We see that that's one of the areas where there wasn't as much or wasn't as good performance, although it's a little bit better than the basic skills and the cognitive ability to describe those skills. Almost as if our self esteem comes not from building up self esteem or fostering the skill of talking well about ourselves, but almost as if self esteem is a side effect of our ability to perform well. I'll put a pin in that. We'll come back to that later. The second approach that they evaluated, the second approach to education was that cognitive approach. Remember, the cognitive approach is your ability to describe your skills as a way of mediating additional skill development. The three different approaches here that are cognitive skill development are parent education, which is where you teach kids to be taught by their parents. Their parents are teaching them to kind of metacognitively analyze what they're doing and describe what they're doing. The team approach is something called cognitive curriculum. You see for team and cognitive curriculum significant reductions in learning, so it's really not necessarily about how you're able to describe what you're doing. It's about your ability to perform, but parent education was an incredibly effective intervention and increased, interestingly enough, folks confidence in their ability in themselves. Having parents involved is an incredibly important component of education. Oddly enough, these cognitive approaches decreased higher order cognitive understanding, so people were not able to as well describe what it was that they were doing with their particular learning approach, and so that's a big challenge. Finally, the behavior analytic approaches to education, which included program instruction that's behavior analysis and direct instruction that highly scripted, fast-paced, teacher-led education strategy. We see that the behavior analysis approach improved basic skills and improved people's confidence in their knowledge, but it didn't increase their ability to describe their performance or be able to higher order reason through and explain to us how they came to those. So if you want to foster your ability to talk about your performance, it's almost like that's a separate response class. Your ability to perform and your ability to describe your performance are different behaviors, but we see that direct instruction saw amazing improvements. We see increases, dramatic increases in basic skills, cognitive functioning, and people were more confident about their learning. Let me just give you another example here. This is the same graph. These are the same data presented in a more colorful format. You can see that the approach to education here, the effective approaches to education were direct instruction, followed by involving parents, getting parents involved, followed by that programmed instruction behavior analytic approach. So, great. It should mean that all of us have experienced direct instruction in our classes, right? Well, we didn't, and it's because there is a lot of pushback to direct instruction, even though there's a lot of support here. So there were subsequent trials to try to figure out, okay, why is it? What's going on? Why is this not being adopted? And here's what we know. So direct instruction was never widely adopted. Here's why. Critics of direct instruction see it as a kind of betrayal of a humanistic egalitarian foundation of public education. There's a lot of resistance to the idea of teacher-proofing curriculum, the idea that teachers are not unskilled workers that you can just script and send in like robots, that teachers are highly skilled, highly professional individuals, and they know their craft best, which I'm a teacher. I get where people are coming from with that. Direct instruction has also been criticized as being incredibly inflexible. So this idea that it's going to handcuff a teacher, a teacher doesn't have the freedom to do what they want in their own classroom and teach the content in the way that they know their students are going to learn best. And despite the evidence that we see that direct instruction increases affective skills, it increases a person's metacognitive abilities to actually talk about their learning, there's some belief that direct instruction produces rote learning or tiny robot learners, which is really heartbreaking. There's a belief that direct instruction doesn't foster higher order cognitive understanding, even though the data we just reviewed suggests that that's not the case. So again, we're doing some follow-ups and see what they found. When you look at the different approaches, when you look at those project follow-up through data, we see that cognitive, higher order cognitive understanding was best with direct instruction followed by the behavior analytic instruction. We see that self-esteem was highest with direct instruction. So this is the best approach to education, but folks don't use it. And this is actually really common in behavior analysis. I'm using this as an example of the fact that though behavior analysis is incredibly powerful and very effective, it's often not supported. It's often not seen as very socially valid. And it tells us that we have a little bit of a reputation problem. One approach that helps solve this is the generative instruction model. In the generative instruction model, this approach to education uses a combination of direct instruction and programmed instruction techniques. And what they're doing here is we have small classes and there's an emphasis on fluency. We're increasing the rate of correct answers over time. And fluency is a foundation of higher order cognitive skills. Your ability to describe or do something fast permits you to think about that skill that you're doing. You're not struggling with trying to produce the answer because you're so fluent in it. You can focus on ways in which that information connects to other pieces. You can synthesize lots of information together. You can see where it connects and you're not focusing on just oh, what's the answer here? You're focusing on how is this applicable in lots of different ways. That's part of the reason why I continue to use Miller's book is because fluency is so important. And this is used at a place called the Morningside Academy. They have one of the best programs and it's been used for kids who traditionally struggle in mainstream education. Kids come in they use this generative instruction model and it's incredibly effective. So effective in fact that the director of the Morningside Academy, Dr. Kent Johnson says he's got a money back guarantee. We tell families that their children will gain at least two grade levels in their skill of greatest deficit or their money back. And we've returned less than 1% of our tuition in 25 years. It's an incredibly effective model. And they show that they demonstrate that with students learning on standardized achievement scores. So what they have here with their displaying here is year by year by year how many grade levels of progress are kids making per year. So on the y-axis we're showing the number of years of grade progress along the x-axis along the bottom I'm showing you the school year the cohort of the kids. In the generative instruction used by Morningside Academy we see that kids are making on average two years of progress. Some are making as much as two and a half and you see that the progress only gets better over time. They're refining that implementation. We see that here toward the end in 1992 kids were making three or three and a half years of progress in their language art skills in a single year of education. Generative instruction is pretty amazing. Again it's that combination of scripted programmed instruction. We're talking about direct instruction strategies and we're talking about building that fluency as a foundation to other skills. And there's no problem with doing those skills and practicing often and getting feedback. Kids are incredibly motivated by it. It's a really great approach and I wish that more schools knew about this. If you ever get an opportunity to check out the Morningside Academy I hope that you'll take it. I'll see if I can find some resources to share with you about their program. And just to quickly summarize remember behavioral approaches to education are incredibly effective but they're unpopular. And for the most part behavior analysts we have a little bit of a reputation problem. Part of it is because we need to improve the way in which we communicate with our stakeholders. And part of it is because systems are well established and it's very hard to change an established system. If you're interested in behavioral approaches to education let me know. If you want to know more about systems change you might consider taking a class called organizational behavior management or actually talk about how do you adjust the behavior of behavior change agents like teachers and managers and business owners. Thanks so much for joining me. I'll see you next time.