 I think we'll start this off with an example, the Hilbert chain is you're going to have eight passes. 105-bill-ous! You grab by the pumpkin! Bullock-fire! Yeah, I'm a biker! We're funny, but not always a joke. Interest. It's a baseline, okay? So we're gonna talk about the baseline logic more. There's a lot to a baseline, it all sorts of things. That was a really confusing baseline. It was one of those ones that's all over the place. The baseline logic is simple. We need to have a way to assess whether or not the independent variable intervention that we're using has been effective. We're not doing large group analysis, so we're not going to take a pre-test and lump of 47 responses together and get an average and compare that using the deep tester and ANOVA or something else to the intervention, right? To the intervention group or the post-test or whatever, right? We're not gonna do that because we're using single-subject logic, which simply means that we're gonna compare one person to themselves and analyze one person at a time, not necessarily compare to themselves. We're just gonna analyze one person at a time, that single-subject logic. So in order to pull that off, we need a baseline, okay? So we gotta figure out where people are at. Well, the first thing you say is, how do you get one? It's simple. You measure, right? You measure, you count, you get frequencies, latency, striations, you know? So maybe in those other videos, you've cut maybe some responses, you know, that have to do with things like, you know, maybe it's a baseline. Maybe if you've watched enough of our videos, you understand that we use behavioral principles to teach you behavioral principles. So meow. One of those things was this meow thing that we were doing in these other videos, meow. And you could argue that that other video, meow, was a baseline of this video about baselines is not necessarily another baseline meow for the meow behavior, but it could be an intervention and we could be assessing it. Anyway, beside the point. So let's just go ahead and take the logic first. We're gonna track behavior for a while without doing anything. A while. What does that mean, right? Usually three, four, five, ten, 12, 47 data points. I don't know. Lots and lots of data points in order to be able to predict what's coming next, right? So what do we need to worry about first? We need to worry about steady-state logic, right? So steady-state logic is the behavior stable, right? So stable, I mean, horizontal. Is it relatively stable? It's gonna go up and down a little bit. So you gotta up and down. So it's gonna go up and down a lot, right? Or versus a lot. We want nice stable stuff, okay? We wanna be able to predict what's coming next. There's some mathematical things that you can use to figure out stability, but really it's a visual analysis. It's an ocular analysis. So just take a look at your data if it feels stable it is. You also have trends that you can worry about in steady-state responding, right? So is our trend in our baseline increasing or is it decreasing or is it stable, right? So we've got those different things. As long as everything's nice and clear and predictable, then we can move on. So once we've got our good baseline, now we're gonna start our intervention phase. Boom, right? So we start that intervention phase and then we figure out what's happening, right? So we do the new independent variable. We see if we can get a modification of behavior out of it. And we're gonna compare that the intervention phase to the baseline, right? So what's the point, right? This is the core logic that we use to establish experimental control and internal validity. We use this comparison, right? Now it gets really complex when we get into the methodology of how you do it, ABAB and ABACA, CBACA, all these different things that we'll get into later. But right now we're just looking at AB, right? So the baseline is the A condition and B is the intervention condition. So what we wanna know during baseline is, do we have predict? Can we predict what's coming, right? And then can we replicate? And then what are they? Prediction verification. Oh, prediction verification and replication. It's kind of funny. So anyway, prediction. Can we predict what's coming next based on what we have in our baseline verification? Can we verify that the independent variable is having an effect by comparing it back to our baseline and in replication? Can we remove that independent variable or whatever the case may be and then bring things back to base? Can we verify that the independent variable had an effect by bringing the behavior back to baseline condition when you put those baseline conditions back in place, right? And then, so that's the replication. So the baseline logic is really how our entire field was built, right? So when Skinner got going on this stuff, he really needed a wit to develop a new methodology that wasn't based on statistics, that was based on actual control of responses. The baseline logic allowed for that. So again, just through quick recap. Baseline, you want stability, you're doing trend analyses, and then you're switching conditions, and then you're making comparisons across from that condition back to the other condition. There's a lot more to it than that and we'll come back to it when we get into all the methodology stuff. So folks, I think, I absolutely think it's time for a beer. We're tired, we're wet, we're cold, and I think we've got to go there to Odorty. I think we're going to shoot a video in Odorty. I don't know if they're going to let us, but we're going to find out. So anyway, enough of this baseline meow, being out here in a damn rain meow, and let's get out of here and go have a beer. All right? I'll tell you. Put the beers over there. Oh, sorry. That's what happens when you're in film.