(The following is basically a response to someone else’s response to a post by a behaviour analyst on the Thinking Person’s Guide To Autism. The post left by Michelle Hecht, MA, BCBA, was a very interesting post and should have awakened more discussion than it did. I suppose that people aren’t thinking about autism so much these days. LizDitz responded with her own take on behaviour analytic ideas, and I set out to provide a similar example of how ubiquitous ABA-type ideas are used in good educational and parental practice. Anyways – my response got too long to post and it has become its own post on my blog. Enjoy!)
Before I got totally into the psychology of teaching and learning, I was a mathematics tutor. Often I’d find students turning up and having serious difficulties with the issue of how to solve linear equations of the form ax+b = cx+d. They couldn’t understand why they couldn’t get these things right, and almost all of them automatically assumed that they weren’t intelligent at all. Rather than try to teach them straight out, I decided to examine what they did.
They would begin by moving things all over the place, without any systematic way of doing it. Why were they doing this? Obviously, to know what to do to help them learn, one has to know why they do what they do. So I asked them.
They’d been told that things moved and signs changed. And they were getting confused. They didn’t get the answers that their teachers at school used to get.
This gave me enough of an idea of what to do: I taught them a formal method, systematic manipulation. Collecting like terms… how to get the terms containing x on one side of the equation and the constants on the other, without moving a thing! I taught them that they would only – in the case of this type of problem – use basic arithmetical operations: add, subtract,, multiply or divide. Simple as that.
I would walk them through a problem: this would involve setting up the problem, and then systematically adding/subtracting/multiplying/dividing … and doing each operation separately. This may be time consuming but it does guarantee that one has a chance of getting a 100% success rate for any test involving this type of problem. I would then set up another, and ask them to tell me what I should do. So I would let them tell me what I should do and, if they had something wrong going on, I could then stop and check with them (and possibly show them what would happen if we went down that wrong road, as it were). The next one, they would get to solve – with me on hand to consult. Then after that, they’d get a handful to solve – and I would oversee the first two or three. This way they could master the technique with someone on hand until they were able to do it, and knew they could do it. And, after doing the remaining problems, and getting them right, they’d get a reward: I’d mark them correct and tell them they’d learned how to solve that sort of problem. And that they were indeed solving the problems on their own!
Leaving out any Vygosktyan/Brunerian concepts of scaffolding and zones of proximal development (and any humanistic concepts such as self-efficacy), the basic processes involved here were:
1- showing the student the nature of the problem, and setting limits on things they can do to solve that sort of problem
2- demonstrating the process of solving the problem, explaining the use of the things being done to solve it
3- letting the student solve other problems of that type, with me on hand to guide, then consult, on how to do it – giving feedback
4- letting the student solve more problems of that type, first overseeing and then leaving alone – and giving feedback at the end of the assignment
Essentially, there’s little difference between this and discrete trial training, in that the student is taken through the process to be learned a little bit at a time and reinforced at every stage. This is a behavioural method of teaching, based on a behavioural analysis of what they were doing that lead to them failing to solve this sort of problem. Sure, as a result of the behavioural analysis, I was able to know what questions to ask, the answers to which were then useful in finding a solution to the problem of them not being able to solve that sort of problem in mathematics. The Vygotskyan/Brunerian and humanistic stuff came as nice side benefits: the main point was the development of a skill set for solving a particular type of problem… self-efficacy is the result of internalisation of the rewarding factors of the learning process.
And nobody had to get slapped across the face for the learning to take place.
Again, no formal training in behaviour analysis. But, as with LizDitz’s situation, assumption that there was a reason why they were doing the wrong things.
I’ve also used behaviour analytic ideas with my daughter, who is Asperger-autistic herself. They work. I was even using them with my nephews and niece when they ‘misbehaved’ (the best one was refusing to pay attention to them until they learned what was wrong about what they did and what they should do instead and that they had to go and apologise – and then, I’d talk to them about what they’d been trying to talk to me about; this is called time out that turns into a negative reinforcement that leads to them behaving ‘appropriately’, which then leads to them getting my attention … the positive reinforcement they desire).
The ABA=doggy-training thing is just a silly way of looking at it (and always has been): humans and dogs have a huge number of things in common, one of which is that they are mammals and another of which is that they have an iron-based oxygen-carrying system as the basis of their blood. Yet another is that they have brains made of the same sort of tissues, and those brains respond to reinforcements in the same way. Sure, it might be more useful and entertaining to dress up the behaviour analytic work in nicer clothes for use on humans, but basically it is being put to work on the same sort of tissue to get the same sort of effect: the learning of something. Above and beyond that, there is no scientific reason to not use behaviour analytic techniques.
Disclaimer: I am not a behaviour analytic psychologist; I am not a behaviour consultant (although I am a behavioural scientist); I am just a guy who trained in applicable psychology and who knows that he has a very wide range of tools with which to work, when his input is needed in some situation.