 I think we'll start this out with a demo. We got the Hilbert chain as you're going to have it. You grab by the function. We might be wrong. We're funny. Ladies and gentlemen, today, Psychcore brings you multiple baselines. Well, maybe it's multiple. So, the big concern with multiple baselines is that some people have a hard time understanding it. How can you establish experimental control using three A-B designs? A-B. There's a B over here and another one down here. There's an A and there's a B. So, A-B designs, as we all know, have zero experimental control. But yet, when you do a multiple baseline, you get lots of it. It's pretty cool. Anyway, how does that happen? It's really simple. It has to do with prediction, verification, and control. So, all of it becomes magical. We'll walk you through these examples. Let's start with a multiple baseline across individuals. All right? I don't know why, because why not? Anyway, so, multiple baseline across individuals. So, we start off capturing a baseline with one person. So, whatever that person's name is, it is in the video at this point. So, there's the baseline. Then we're going to start a phase change. That's what we do for phase changes here at Psychcore. And then we continue on with the intervention. Whatever that may be. So, since we have a particular person here that we're talking about, and in fact, a couple more people, we should pick a behavior that's kind of fun. I don't know. Maybe nose picking. So, nose picking is a good behavior. So, let's pick noses. I'm sorry. Maybe we should not. You can't even tell if I'm picking mine. Anyway, we're going to watch nose picking behavior. We're going to find out how many times somebody puts their fingers in there and starts digging. So, we've got a particular pattern of responding for the first baseline. We start our intervention, and then we have responding after the intervention. We have nothing yet. But we have established a little bit of prediction. How do we do that? Over here. How did that happen? We got our baseline. And we predict that something is going to change based on our intervention. Right in here. So, subsequently, we can then verify whether or not we got the changes we expected. However, because this is an AB design, we do not have any experimental control yet. We're going to add a second baseline for that. Okay? In fact, this time we're going to cross people. So, we have our second individual that we're working with. I don't know. It might be a little lower. I can't quite tell from behind the screen. Anyway, so we have our second person. We're going to run that baseline for longer than the intervention. So, this baseline down here is going to continue past the start of the first intervention. In fact, we're going to start the intervention over here. Alright? Why do we do that? That way, we can see if the behavior changes at the moment that the intervention starts for both people. Now, the person, the second person in the baseline, their behavior should not change at all. Nothing whatsoever at the time that the first person starts their intervention right there. Their behavior should change here when they start their second intervention. Okay? Or when the second person starts their intervention, their behavior should change accordingly here. This allows us to test that effect of that intervention. We are now establishing, again, more prediction and verification. Again, we predicted that the intervention would work based on the we have our baseline and we predict that our intervention is going to have an effect on behavior. And as you can see in the data, it had the effect for this person, then it had the effect for this person down over here. So, you see those things? So, everything is working absolutely perfectly. So, we have our prediction, we have our verification and then we double-checked it with the next one, more prediction and verification and again, more prediction and verification down here for the third person. Now, you'll note with the third person that we extended their baseline well past the first one and the second one, we're into their third position now. So, their intervention didn't start till over here. This allows us to make comparisons in each person's baseline up. So, we go from this person's baseline here and we look up and we see what's going on. The other person's behavior is changing is this person's. If it's not then we can rule out things like history effects and environmental effects that we may not have control over, all evaluating whether or not our intervention is having the effect it was having. So, again you're going to look carefully at the screen, you're going to see a series of overlapping data points or some overlapping bars in our graphs and those bars will show you exactly what it is you need to compare when you're looking at a multiple baseline design like this one. So, we've talked about prediction. We've talked about verification. So, how did we get control? Well, control was established by the fact that the behavior for each person changed at the moment the intervention started. So, for this person and for this person and for this person the behavior started to change after the intervention, not before. We do not want the behaviors to change before the intervention, otherwise we do not or we have not established the effect of that intervention. We have a problem with control in our experiment. So, all of these things combined make one beautiful, really cool design based on some really, really simple rudimentary designs, the AB. When we combine them with an AB and another AB and another AB in a very specific pattern as you can see here then you will note that we have a very powerful, internally valid design called the multiple baseline. We could have done this across individuals, but we did. We could have done it across settings for one individual and we could have done it across behaviors. So, we could have multiple behaviors or multiple settings for one individual. There's really no limit as to how you could do these things. You could do more. You could do reversals inside of this too, but that's for another situation at another time if you'd like to cover it. So, again, we have some prediction. We're predicting what's coming next based on each individual baseline for each person. We're also predicting whether or not the effect is going to continue for if it started here. Well, it can be predicted it will continue here. We verify that by comparing up above and that multiple comparisons allows us to establish control. If you have any other questions about a multiple baseline, please see us.