 I think we'll start this out with a demo of the Hilbert chain as you're going to have eight pairs. You grabbed by the pumpkin full of her. We might be wrong. We're funny, but not always a joke. As you might have guessed, it's another episode of Behavior Beast. Surprise, surprise, we're kind of going back to basics a little bit today because we're moving back to the beginning of some of the stuff that we probably should have covered a long time ago, but don't ask why. There was actually a plan as to why we didn't cover this early on, but we're going to get into it now. Why don't we talk a little bit about experimental design, experimental concept, talk a little bit about methodologies and things like that. Welcome out here and let's talk about what it is we need to know in order to get where we're getting. I have a hard time talking about this topic, not because I'm confused by it. This is actually my area that I really love the most or have the most training in, so to speak. But because there's like two different pieces to experimentation, and I don't think it's well delineated in the various textbooks, one of the things that you need to know is that the applied world versus the experimental world are two different things. You can do experiments in the applied world. That's a lot of what I've done throughout my history as an academician and done my work in, but you lose control, right? And you lose experimental control. We'll talk about experimental control in a minute. So that's the applied setting. In the laboratory and experimental setting, you might have a basic question that you're trying to answer. You don't have to answer that question with animals and rats and pigeons or other non-humans. You can answer it with human animals as well. Some people think that all work in the laboratory is done just with rats and pigeons, and that's just a bunch of hooey. I mean, I've done plenty of work in the laboratory with humans, and not in a sort of naughty way, which leads us to kind of that distinction, right? It's really about the research question that you're asking. A lot of times when we're in an experimental setting, in a laboratory type setting, or when we're designing an experiment per se, we are really worried about experimental control. Experimental control is literally the ability to minimize the influence of extraneous variables. So what's an extraneous variable? Well, there's a lot of names for them. Confounds, concomitant variables, extraneous variables, whatever. There's anything that can come and mess with your experiment. So what do I mean by mess with your experiment? I mean this. I mean a mess with the ability, or interfere with the ability for you to determine what is causing the effects that you see. So your experimental design and your experimental control that is developed out of the... Sorry, there's bugs out here. I'm like, I'm getting eaten alive. Anyway, so that has a bit of a confound on my ability to deliver a crunch lecture. I'm going to scratch on my face off, or my neck off, so you get the idea, right? So that is a confound, and the irony is that it really does itch like crazy, and normally they don't bug me. So when we're talking about experimental control, what we're really talking about is the ability to minimize confounds. And what do the confounds influence? They influence your ability to determine what is causing the results that you see. So if you have a really nice design, like an ABAB design, you come up with some really good experimental control, some high internal validity, but because it minimizes confounds. But an ABAB design doesn't do squat for minimization of confounds. Nothing, like, well, maybe a little bit, but nothing worth speaking of. It's a garbage design when it comes to the ability to detect effects, right? It's a cool design that you might use in research or in practice. So we got our baseline and we implement a change, or so we have a baseline and we implement change. Great. Congratulations. You implemented it. You don't know if that changes the result of your independent variable or not, but at least you got the change in the real world, right? In an experiment, that's not good enough. It's not going to get you published just to show an ABAB design. It's not going to get you the level of control that you need to determine if that intervention at that phase change was actually what's causing whatever change you see during that B condition. And this is why we go to things like ABAB designs or BAB designs or multiple baseline designs or changing criteria. All these other design types out there allow us to draw some more conclusions about cause and effect. So another note. Sorry, this video is going to be full of notes. Another note is about cause and effect. In our field, we don't use the term cause and effect. We use the term functional relation. I tend to use cause and effect when I'm like teaching about it initially because it helps you thinking about what a functional relation means. And here's why. Is event A functionally related to event B? There you go. In other words, does event A cause event B? The functional relation language is more accurate, believe it or not, than cause and effect because we can't ever completely rule out every single possible thing out there. So we just worry about whether or not something is functionally related. So experimental control comes about through the type of design you choose. There's lots of different design types. We're going to get into a few of those. But we have to worry about a couple other things here. So the setting that you're in, obviously this thing. So we have different experimental settings. It could be in a laboratory. It could be in the real world. We talk about subjects, right? So I prefer the term subjects as opposed to participant. But you can think about it as whatever you want. And then you can think about the single subject logic. The logic that we use in our experimentation is single subject. That does not mean we're studying one person only. You could study 50 people using a single subject approach. It's about an individual level of analysis. So you analyze people by themselves. You have five people on a study. You analyze each one of them behavior for their own sake. Each person's behavior for its own sake. We don't aggregate it. We don't average. We don't do those sorts of things. We don't do single subject logic. Let's see. So minimization of confidence. Dependent variables. This is the thing that you're measuring. What's it dependent upon? It's dependent upon the independent variable. That's your intervention. Independent variables, interventions. Dependent variables, that's the behavior you're seeking change in. So those are some of the basics, the core pieces of experimentation. I'm going to take you back in another video in a fair minute and talk about parametric analysis because that seems to be an issue in our field with people understanding what parametric analysis is. We'll find a better spot to do that in and we'll come back shortly. Thank you. If you like the way it is and you want it to continue what it is, what it was and what it shall be if you want that then like, subscribe, and share please because that's the only way it's going to happen. I promise. That's it. That's all you get. More. You want videos? Like, subscribe, share. Please. Please.