WEBVTT 1 00:00:03.399 --> 00:00:15.880 You're listening to why we do what we do? Welcome 2 00:00:15.919 --> 00:00:18.719 to why we do what we do. I am your 3 00:00:18.920 --> 00:00:23.160 automatically reinforced host, Abraham, and I'm. 4 00:00:22.960 --> 00:00:25.519 Your alone condition Shane All. 5 00:00:27.239 --> 00:00:29.600 There's so many problems with the way I describe myself. 6 00:00:29.640 --> 00:00:32.320 That's okay, We're moving on. We are a psychology podcast. 7 00:00:32.399 --> 00:00:35.079 We talk about the things that humans and non human 8 00:00:35.119 --> 00:00:37.880 animals do. And this is a part two episode. We 9 00:00:38.039 --> 00:00:41.960 have been talking about something inside of behavior science in 10 00:00:42.000 --> 00:00:45.840 which an assessment is conducted to attempt to determine why 11 00:00:46.240 --> 00:00:50.280 certain behaviors happen, particularly when there are people who can't 12 00:00:50.320 --> 00:00:53.560 just say why those behaviors happen, and so they conduct 13 00:00:53.640 --> 00:00:55.920 some kind of an assessment to try and learn what 14 00:00:56.079 --> 00:00:59.719 is going on for those individuals. It's called a functional analysis. 15 00:00:59.799 --> 00:01:02.039 And so if you did not hear the first part, 16 00:01:02.079 --> 00:01:05.519 we dig a lot into the history, concepts, a little 17 00:01:05.560 --> 00:01:09.640 bit of jargon, and the foundational and seminal study that 18 00:01:09.760 --> 00:01:12.239 kicked off this. So if you do not hear that part, 19 00:01:12.280 --> 00:01:14.120 I do recommend checking it out. We'll do a brief 20 00:01:14.159 --> 00:01:17.040 recap at the beginning of this discussion, so in case 21 00:01:17.079 --> 00:01:19.599 you don't remember or you haven't heard it, you'll at 22 00:01:19.680 --> 00:01:21.439 least have a little bit of context to help set 23 00:01:21.519 --> 00:01:23.879 up for this part because we're going to be building 24 00:01:23.920 --> 00:01:25.480 off of that discussion today. 25 00:01:26.280 --> 00:01:29.000 Yes, yes, yes, yes, yes, yes, so yeah, we'll give 26 00:01:29.000 --> 00:01:30.719 you the ins and out and kind of where we're 27 00:01:30.719 --> 00:01:33.079 at leading up to all the cool new stuff that's 28 00:01:33.079 --> 00:01:33.439 going on. 29 00:01:33.920 --> 00:01:35.840 Right. So, if you're joining us for the first time, 30 00:01:35.879 --> 00:01:37.840 then welcome. I do recommend you go back and check 31 00:01:37.840 --> 00:01:39.400 out that episode and then you'll be joining us for 32 00:01:39.439 --> 00:01:41.879 maybe the second time. But either way, you found yourself 33 00:01:41.920 --> 00:01:44.079 here and I appreciate you being here. If you're a 34 00:01:44.120 --> 00:01:46.319 returning person, thank you for coming back. And if you 35 00:01:46.599 --> 00:01:48.719 like to support us, and you haven't done that already, 36 00:01:48.840 --> 00:01:51.480 you can leave us a rating and review, like subscribe, 37 00:01:51.640 --> 00:01:54.319 tell a friend, join us on Patreon. I'll talk more 38 00:01:54.319 --> 00:01:56.879 about those things at the end of our discussion today. 39 00:01:56.920 --> 00:01:59.120 But this is a part two, but it still comes 40 00:01:59.159 --> 00:02:00.879 out on a different day of the week than Part one. 41 00:02:00.959 --> 00:02:03.200 Matter of fact, this one comes out on August fifth, 42 00:02:03.239 --> 00:02:04.840 So Happy Green Peppers Day. 43 00:02:05.439 --> 00:02:09.759 Yes, of course, love Green Peppers. It is International Alberino Day. 44 00:02:10.159 --> 00:02:10.800 What does that mean? 45 00:02:11.120 --> 00:02:12.639 No idea. We're gonna leave it at what it is, 46 00:02:12.759 --> 00:02:13.919 just like cheese Sacrifice Day. 47 00:02:14.479 --> 00:02:17.879 It's okay, very good. It is International Traffic Light Day. 48 00:02:19.719 --> 00:02:23.960 I have fast. It's National Oyster Day. 49 00:02:24.280 --> 00:02:27.599 M go support your local oyster by giving it. 50 00:02:27.599 --> 00:02:29.159 Go pet your oysters. 51 00:02:29.199 --> 00:02:34.360 You'll plus your oysters. All right. National Psychiatric Technician Appreciation Day. 52 00:02:34.400 --> 00:02:37.280 Well that feels. That feels very relevant to our topic. 53 00:02:38.400 --> 00:02:41.080 Yes it does. So does National Underwear Day. 54 00:02:43.280 --> 00:02:46.599 Nice. And it is work like a dog Day, which 55 00:02:46.639 --> 00:02:49.159 is to say, lay on a couch, chew a bone, 56 00:02:49.439 --> 00:02:52.080 take a nap. Yes, have someone take you out to 57 00:02:52.080 --> 00:02:52.759 go poop in the yard. 58 00:02:53.240 --> 00:02:54.080 Yes, exactly. 59 00:02:54.199 --> 00:02:55.280 That is working like a dog. 60 00:02:55.439 --> 00:02:55.840 All right. 61 00:02:56.400 --> 00:02:59.479 Anyway, we don't just talk about holidays. We also talk 62 00:02:59.520 --> 00:03:01.520 about we don't science things, and we don't poop in 63 00:03:01.560 --> 00:03:04.080 you know, well we don't. Maybe some people have listened 64 00:03:04.080 --> 00:03:06.360 to us. Probably not, Yeah, some people do it. We 65 00:03:06.439 --> 00:03:09.080 as a species. Okay, let's do a brief recap. So, 66 00:03:09.120 --> 00:03:12.080 as I mentioned, we've been describing in this discussion so 67 00:03:12.199 --> 00:03:15.520 far something called the functional analysis. What that is, where 68 00:03:15.520 --> 00:03:18.759 it comes from? And so as a brief background here, 69 00:03:20.039 --> 00:03:24.479 long long ago in behavior science, calling something a functional 70 00:03:24.479 --> 00:03:29.080 analysis was more like a conceptual exercise to just try 71 00:03:29.120 --> 00:03:32.680 and like think through what are the circumstances, certain classes 72 00:03:32.719 --> 00:03:35.520 of behaviors tended to occurb. That was kind of all 73 00:03:35.560 --> 00:03:38.479 that it was. But then at the time also you 74 00:03:38.599 --> 00:03:41.120 had people who had disabilities. Some of them were very 75 00:03:41.120 --> 00:03:43.159 impacted by those disabilities that had meant a lot of 76 00:03:43.199 --> 00:03:46.800 limitations for them. They were hospitalized, they were lobottomized, they 77 00:03:46.840 --> 00:03:51.120 were medicated into a coma basically, and along came behavior 78 00:03:51.120 --> 00:03:55.240 analysts behavior as scientists who are doing our behavior modifiers 79 00:03:55.240 --> 00:03:57.439 there was kind of a more popular name at the time, 80 00:03:57.719 --> 00:04:00.280 and they're like, well, we can use rewards and punishments 81 00:04:00.360 --> 00:04:04.360 to direct these behaviors to do these other things. Those 82 00:04:04.439 --> 00:04:07.879 kind of where things started before there was sort of 83 00:04:07.919 --> 00:04:11.080 a more assessment based functional analysis. 84 00:04:11.479 --> 00:04:14.719 Yes, and then around the same time you had Iwata 85 00:04:14.759 --> 00:04:17.560 and friends as well as car and Durand started developing 86 00:04:17.600 --> 00:04:20.720 a technology for us to be able to critically evaluate 87 00:04:21.120 --> 00:04:23.920 and really deep dive into the conditions under which this 88 00:04:24.040 --> 00:04:26.639 certain type of behavior occurred. And they looked at the 89 00:04:26.680 --> 00:04:29.600 conditions before and after a behavior occurred, and they looked 90 00:04:29.639 --> 00:04:33.160 at isolating those as much as possible. Now, Card and 91 00:04:33.199 --> 00:04:35.720 Durand is the nineteen eighty five article that kind of 92 00:04:35.720 --> 00:04:39.199 started to really kind of evaluate this and develop a technology. 93 00:04:39.279 --> 00:04:43.040 But this was also in development in nineteen eighty two 94 00:04:43.240 --> 00:04:46.000 with Brian Iwata and his friends, and so they came 95 00:04:46.079 --> 00:04:47.680 up with kind of what we understand today is the 96 00:04:47.680 --> 00:04:51.199 gold standard of a functional analysis, which has multiple conditions, 97 00:04:51.240 --> 00:04:53.439 and what it does is it looks at setting up 98 00:04:53.439 --> 00:04:57.800 a circumstance for behavior to possibly occur and then delivering 99 00:04:57.839 --> 00:05:01.439 some outcome as a result of that behavior occurring and 100 00:05:01.480 --> 00:05:03.959 seeing if that influences behavior in some ways. And they 101 00:05:03.959 --> 00:05:07.360 did this with really three different types of outcomes. It 102 00:05:07.399 --> 00:05:10.560 was either positive reinforcement, negative reinforcement, or we don't know 103 00:05:10.600 --> 00:05:12.800 what we're doing or what's going on as behavior, so 104 00:05:12.839 --> 00:05:15.879 we're gonna we're gonna see this play out commission and 105 00:05:16.480 --> 00:05:18.160 it changed the world. It changed the world to be 106 00:05:18.279 --> 00:05:21.600 par analysis in ways that we cannot possibly begin to enumerate. 107 00:05:22.000 --> 00:05:24.120 I mean, we could possibly begin to, but that's as 108 00:05:24.120 --> 00:05:25.759 far as we make it. Before we were a live 109 00:05:25.839 --> 00:05:29.439 well and that's another episode in the bank. So yeah, yeah, yeah, 110 00:05:29.680 --> 00:05:31.879 it was very profound. And yeah, just as you said, 111 00:05:31.920 --> 00:05:33.720 I think an important thing then maybe a lot of 112 00:05:33.759 --> 00:05:37.759 people don't necessarily think about with respect to the functional 113 00:05:37.759 --> 00:05:41.160 analysis is it was essentially these behaviors that are happening, 114 00:05:41.839 --> 00:05:44.959 which in the original study was was primarily self injury, 115 00:05:45.079 --> 00:05:48.079 is like self harm behaviors. They're either happening because there's 116 00:05:48.120 --> 00:05:51.240 some positive reinforcement for it, there's some negative reinforcement for it, 117 00:05:51.319 --> 00:05:54.399 or there's some self sensory stimulation that is happening as 118 00:05:54.399 --> 00:05:56.800 an outcome for it. That was the hypothesis that drove 119 00:05:56.879 --> 00:05:59.519 the development of the framework for that assessment. It's like, 120 00:05:59.600 --> 00:06:02.199 let's create those conditions see when the behavior occurs, so 121 00:06:02.199 --> 00:06:05.360 we now understand why the behavior occurs, right and with 122 00:06:05.480 --> 00:06:09.879 the technology developed by car and Durand who conceptualized these 123 00:06:10.519 --> 00:06:13.720 behaviors as a form of communication, essentially that these are 124 00:06:13.720 --> 00:06:16.360 people who are saying, they were trying to communicate, we 125 00:06:16.399 --> 00:06:18.240 need something. We don't know how to ask for what 126 00:06:18.279 --> 00:06:20.439 we need, but we do need it, and right now 127 00:06:20.480 --> 00:06:22.759 the way that we've learned to get it is doing 128 00:06:22.800 --> 00:06:25.759 these types of behaviors, and they're saying, well, okay, once 129 00:06:25.800 --> 00:06:27.800 we know what you need, we can now teach you 130 00:06:27.839 --> 00:06:29.040 to ask for that in a different way. And so 131 00:06:29.079 --> 00:06:32.279 that was the intervention that was born out of the assessment, 132 00:06:32.279 --> 00:06:34.600 and we'll get into those a bit more as later 133 00:06:34.680 --> 00:06:36.959 parts of this discussion. But essentially what we did in 134 00:06:37.000 --> 00:06:40.319 the last episode is describe in fairly great detail how 135 00:06:40.360 --> 00:06:44.079 that experiment was set up, the different conditions, how they worked, 136 00:06:44.639 --> 00:06:47.600 and because it'll be relevant for today, I'll just briefly 137 00:06:47.600 --> 00:06:50.519 describe that in the original assessment, these were fifteen minute 138 00:06:50.519 --> 00:06:54.839 conditions where they rotated between whether they got social disapproval 139 00:06:54.879 --> 00:06:58.480 as a form of positive attention and reinforcement. They got 140 00:06:58.720 --> 00:07:01.959 removal of demands, which was escape from a condition that 141 00:07:02.079 --> 00:07:05.560 was the negative reinforcement or demand condition. They had a 142 00:07:05.600 --> 00:07:08.319 toy play condition, which was essentially a neutral everything is 143 00:07:08.399 --> 00:07:11.199 chill vibes session, and then there was like, yeah, all 144 00:07:11.240 --> 00:07:15.000 by yourself, do you do these self injury behaviors when 145 00:07:15.279 --> 00:07:17.519 just there's no one around to do anything with you 146 00:07:17.600 --> 00:07:19.319 type of thing to try and figure out if that's 147 00:07:19.360 --> 00:07:22.000 the automatic piece. They were fifteen minutes long. They were 148 00:07:22.079 --> 00:07:26.879 usually randomly rotated. Every time that the challenging behavior was observed, 149 00:07:26.959 --> 00:07:30.519 there was a specific type of outcome that was provided 150 00:07:30.600 --> 00:07:33.279 by the person in the room, unless it was two 151 00:07:33.279 --> 00:07:35.879 of those sort of control conditions, the toy player alone, 152 00:07:36.480 --> 00:07:41.759 and that when they observed a higher percentage of intervals 153 00:07:42.439 --> 00:07:45.079 with that self injury, then that indicated that that was 154 00:07:45.160 --> 00:07:47.439 the condition that under which the behavior was most likely 155 00:07:47.480 --> 00:07:49.639 to occur, that was the most likely outcome that that 156 00:07:50.000 --> 00:07:52.879 individual was seeking in those moments, and that was it. 157 00:07:52.959 --> 00:07:55.519 So fifteen in intervals. They were divided up in these ten 158 00:07:55.680 --> 00:07:58.800 second intervals per session, and they repeated each of those 159 00:07:58.800 --> 00:08:03.560 conditions until they either had clear differentiation between the conditions 160 00:08:03.600 --> 00:08:07.839 meaning that the self injury only occurred in one condition, 161 00:08:08.879 --> 00:08:12.639 or they had twelve days of sessions that they had 162 00:08:12.680 --> 00:08:15.399 experienced those conditions. The last part of this then, is 163 00:08:15.439 --> 00:08:18.920 that this study was later republished in nineteen ninety four, 164 00:08:18.959 --> 00:08:22.399 and that was sort of like really kicked off a 165 00:08:22.439 --> 00:08:24.120 lot of new research in this field. 166 00:08:24.800 --> 00:08:27.120 Yeah, yeah, yeah. So that's why the I WANTA study 167 00:08:27.160 --> 00:08:29.920 is often cited as nineteen eighty two slash nineteen ninety four, 168 00:08:30.160 --> 00:08:33.200 because it was published, it was published, it was republished, 169 00:08:33.559 --> 00:08:36.240 and really, I mean, like you said, kicked off a 170 00:08:36.399 --> 00:08:40.759 whole set of research that was specifically focused on functional 171 00:08:41.120 --> 00:08:44.799 equivalents and making sure that we were teaching skills based 172 00:08:44.799 --> 00:08:45.759 on circumstance. 173 00:08:46.320 --> 00:08:49.120 Okay, so that is as much I think of a 174 00:08:49.200 --> 00:08:52.120 nutshell as I can do about the last hour long 175 00:08:52.159 --> 00:08:54.399 discussion that we did Yeah, is there anything that I 176 00:08:54.399 --> 00:08:56.639 missed you felt like was important that we communicate before 177 00:08:56.639 --> 00:08:57.559 we dive into part two. 178 00:08:59.000 --> 00:09:02.240 Let's see, we had some recommendations at the end of it, 179 00:09:02.279 --> 00:09:06.200 which was really nice. We talked about how the current 180 00:09:06.200 --> 00:09:10.360 administration is trying to institutionalize people and we don't want 181 00:09:10.360 --> 00:09:12.279 to go back to that because it was bad. No, 182 00:09:12.360 --> 00:09:13.639 I think you covered it. I think we did it. 183 00:09:13.600 --> 00:09:18.639 Great, perfect, Thank you all right? Perfect? So let's go 184 00:09:18.679 --> 00:09:21.879 ahead then, and essentially the next part here is understanding 185 00:09:21.879 --> 00:09:24.559 what happened after that big study sort of came out 186 00:09:24.559 --> 00:09:28.000 because there was a lot left to learn and new 187 00:09:28.120 --> 00:09:31.080 challenges faced by people who were at tempting to implement 188 00:09:31.200 --> 00:09:34.840 these types of strategies. But whole what a groundbreaking and 189 00:09:35.039 --> 00:09:38.840 like foundational seminal article this was. It really changed the 190 00:09:38.919 --> 00:09:41.480 name of the game. In this discussion, we'll essentially be 191 00:09:41.559 --> 00:09:45.600 expanding on what came after. So when technology is when 192 00:09:45.600 --> 00:09:48.840 new technologies such as this one emerge, there's often a 193 00:09:48.879 --> 00:09:51.840 substantial interest in determining what we do with this, like 194 00:09:51.879 --> 00:09:54.240 how do we improve it, how do we refine it, 195 00:09:54.639 --> 00:09:56.720 what do we need, what do we not need? How 196 00:09:56.759 --> 00:09:59.440 do we make it more efficient? And they're a hardly 197 00:09:59.480 --> 00:10:03.000 better exams this than there are with functional analyzes, which, 198 00:10:03.039 --> 00:10:04.559 by the way, I forgot to mention this, but we 199 00:10:05.320 --> 00:10:08.600 might call them fays because that's a very common way 200 00:10:08.639 --> 00:10:11.440 that they're talked about, is just to abbreviate functional analysis 201 00:10:11.480 --> 00:10:15.279 as FA FA, so you might hear that just to know. 202 00:10:15.399 --> 00:10:17.759 So anyway, there are hardly better examples of the refinement 203 00:10:17.799 --> 00:10:20.759 process than is what happened with FAS. So will briefly 204 00:10:20.879 --> 00:10:23.799 enumerate some of these refinements and experimental arrangements that have 205 00:10:23.879 --> 00:10:29.679 evolved the implementation standards of this experimental FA and that's 206 00:10:29.720 --> 00:10:32.600 actually that is a qualifier that is sometimes attached to 207 00:10:32.639 --> 00:10:35.639 functional analysis as to call it experimental because it is 208 00:10:35.679 --> 00:10:38.120 set up like an experiment. You have essentially test and 209 00:10:38.159 --> 00:10:42.039 control conditions that are alternated to help establish a pattern 210 00:10:42.480 --> 00:10:46.759 that differentiate data trends in only certain conditions to illustrate 211 00:10:46.759 --> 00:10:49.759 and a numer like help understand what's going on as 212 00:10:49.759 --> 00:10:52.960 opposed to just functional analysis, which as we mentioned KITE 213 00:10:53.039 --> 00:10:56.360 maybe could refer to something more conceptual in nature and 214 00:10:56.360 --> 00:10:59.240 not necessarily in experiment. So this is sometimes called an 215 00:10:59.279 --> 00:11:02.480 experimental functional analysis or experimental EFFA. 216 00:11:03.080 --> 00:11:05.080 Yeah. I think another term too that we want to 217 00:11:05.120 --> 00:11:06.720 make sure that everybody is aware of because I think 218 00:11:06.799 --> 00:11:10.039 you said this just now, which was new technologies. When 219 00:11:10.080 --> 00:11:12.679 we talk about technologies, I think people automatically assume that 220 00:11:12.720 --> 00:11:14.639 what we're talking about is something like electronic like the 221 00:11:14.639 --> 00:11:17.799 advance Secures advice that look a device, And it's really 222 00:11:17.840 --> 00:11:21.320 important understand that a technology is also something like something 223 00:11:21.399 --> 00:11:24.559 that we do. Yeah, it's basically like some something new emergency, 224 00:11:24.559 --> 00:11:27.080 the sign something that is useful and pragmatic that can 225 00:11:27.080 --> 00:11:30.480 be used to you know, improve somebody's life, change behavior, 226 00:11:30.519 --> 00:11:32.759 stuff like that that's also a technology. Like a token 227 00:11:32.799 --> 00:11:34.720 economy is an example of a technology that we use 228 00:11:34.960 --> 00:11:38.039 in behavior analysis because we can take this intervention and 229 00:11:38.080 --> 00:11:40.039 we can apply it in a in a different place. 230 00:11:40.080 --> 00:11:41.799 So like I think that, I think that's really valuable 231 00:11:41.799 --> 00:11:44.200 to remind everybody about too, because technology does have its 232 00:11:44.200 --> 00:11:45.240 own connotations. 233 00:11:45.279 --> 00:11:47.240 Such a good catch. Yeah, I wasn't thinking about that 234 00:11:47.240 --> 00:11:50.320 beer exactly right, that like a strategy is a type 235 00:11:50.320 --> 00:11:52.559 of technology. There's it's a it's a different way of 236 00:11:52.600 --> 00:11:54.440 doing it, Like there's this whole form of expertise that 237 00:11:54.480 --> 00:11:57.440 can go along with a new strategy, a new way 238 00:11:57.480 --> 00:12:01.240 of doing things. The antithesis of technolo logical development as ADS. 239 00:12:09.240 --> 00:12:11.480 All right, we're back, So we barely started to really 240 00:12:11.480 --> 00:12:13.399 get into the new discussion here, but we felt like 241 00:12:13.440 --> 00:12:16.399 it was really important to ensure that we had recapped 242 00:12:16.440 --> 00:12:19.519 the foundation. It's been a week since that episode came out. 243 00:12:19.919 --> 00:12:22.240 Maybe you're already familiar with this, maybe this is new 244 00:12:22.279 --> 00:12:23.759 to you, but we just wanted to make sure we 245 00:12:23.879 --> 00:12:27.080 had enough sort of foundation that we could proceed with 246 00:12:27.159 --> 00:12:29.200 the next part of this and have that makes sense 247 00:12:29.759 --> 00:12:32.480 because it's building on what we've talked about last time. 248 00:12:32.519 --> 00:12:34.440 So that took the majority of our initial part here 249 00:12:34.519 --> 00:12:37.399 before we had to take our first AD break. But 250 00:12:37.559 --> 00:12:39.759 here we are. We're getting into now the new stuff. 251 00:12:40.080 --> 00:12:43.840 Yes, so there are a few major methodological shifts in 252 00:12:43.919 --> 00:12:47.039 that PAS that have become essentially standard practice in several 253 00:12:47.080 --> 00:12:49.320 other versions of that FAS that have been attempted. So 254 00:12:49.360 --> 00:12:51.720 I think it's really important though, Like just because this 255 00:12:51.879 --> 00:12:54.559 was done and it was developed and it was incredibly 256 00:12:54.639 --> 00:12:56.960 useful at the time that it was developed, doesn't mean 257 00:12:57.000 --> 00:12:59.399 it stayed like that, because remember, science does not have 258 00:12:59.399 --> 00:13:01.399 a finish line. So we're going to start with some 259 00:13:01.440 --> 00:13:04.480 of the major procedural shifts that occurred. So first that 260 00:13:04.519 --> 00:13:06.919 we talked about was, and this is really important you 261 00:13:06.960 --> 00:13:10.039 brought this up in the recap, is the session length. 262 00:13:10.480 --> 00:13:14.000 So in the original study, each condition or sometimes you'll 263 00:13:14.080 --> 00:13:17.399 see the term condition and session used interchangeably, but condition 264 00:13:17.480 --> 00:13:21.360 is really a better term. But each condition lasted fifteen minutes. 265 00:13:21.720 --> 00:13:23.519 And there was an early attempt in nineteen ninety one 266 00:13:23.519 --> 00:13:26.559 by Northrop and friends to streamline the FA process by 267 00:13:26.559 --> 00:13:30.120 conducting shorter sessions that were only five minutes long, or 268 00:13:30.159 --> 00:13:33.600 shorter conditions that were only five minutes long. However, several 269 00:13:33.639 --> 00:13:36.799 different studies attempted shorter session durations, and in nineteen ninety nine, 270 00:13:36.879 --> 00:13:39.600 Whillas and i Wata published a study demonstrating that reducing 271 00:13:39.639 --> 00:13:45.279 sessions to ten minutes produced identical data to fifteen minutes. Further, 272 00:13:45.679 --> 00:13:49.039 reducing sessions to only five minutes yielded only minor discrepancies 273 00:13:49.200 --> 00:13:53.279 from the fifteen minute observations. So thus I sow, thus 274 00:13:53.320 --> 00:13:55.759 I don't need to say both of those. Thus, most 275 00:13:55.840 --> 00:13:59.200 contemporary studies use five minute sessions for each condition, yielding 276 00:13:59.200 --> 00:14:02.919 a more efficient procedure and analysis. So basically what this 277 00:14:03.000 --> 00:14:05.399 is saying is originally fifteen minutes, we don't need to 278 00:14:05.399 --> 00:14:07.000 do fifteen minutes we were able to shorten it. 279 00:14:07.399 --> 00:14:09.159 Yeah. Yeah, it's like we got it down to ten, 280 00:14:09.200 --> 00:14:11.320 that already lopped off thirty three percent of the time 281 00:14:11.360 --> 00:14:13.799 we were spending, and it turns out we can actually 282 00:14:13.799 --> 00:14:16.080 get it down to five. And because that was like, 283 00:14:16.759 --> 00:14:20.120 your data were likely to show similar patterns at five 284 00:14:20.159 --> 00:14:22.799 minutes as they were at fifteen minutes, I was like 285 00:14:22.840 --> 00:14:24.960 that that's a lot more efficient. We can get a 286 00:14:25.000 --> 00:14:27.679 lot more done in a lot less time because we 287 00:14:27.720 --> 00:14:31.080 apparently don't actually need that much time to do this. 288 00:14:31.120 --> 00:14:33.720 So that it was like a really important refinement to 289 00:14:33.799 --> 00:14:36.399 determine that they could do that. So it's not that 290 00:14:36.480 --> 00:14:38.159 every single study is going to do this, but it 291 00:14:38.240 --> 00:14:41.000 is fairly common that if you were to look at 292 00:14:41.039 --> 00:14:45.360 a relatively recent study that used a functional analysis as 293 00:14:45.480 --> 00:14:49.759 part of their methods, which you will very undoubtedly find 294 00:14:49.799 --> 00:14:54.759 if you're looking in behavior analytic research, then it probably 295 00:14:54.879 --> 00:14:57.759 used a five minute interval for their functional analysis. Not 296 00:14:57.799 --> 00:15:01.000 always right, but a huge amount of the time. Right now, 297 00:15:01.080 --> 00:15:04.799 So that was one of the first major changes is 298 00:15:05.080 --> 00:15:09.480 that the standard setup was fifteen minutes. Now the standard 299 00:15:09.519 --> 00:15:12.039 set up, regardless of anything else you're doing with that 300 00:15:12.080 --> 00:15:16.240 function analysis, is probably five minutes right, major caveats in there, 301 00:15:16.240 --> 00:15:18.559 we'll get to some of the other refinements that have happened. 302 00:15:18.559 --> 00:15:22.080 But a second thing that changed and changed fairly early 303 00:15:22.120 --> 00:15:25.960 on in this was the measurement strategy. Now, as we 304 00:15:26.039 --> 00:15:29.799 described in the original study, they divided this fifteen minute 305 00:15:29.799 --> 00:15:33.639 session into intervals, specifically ten second intervals, and then they 306 00:15:33.679 --> 00:15:38.360 calculated the percentage of intervals in which the target behavior occurred. So, 307 00:15:38.519 --> 00:15:40.879 just as a quick description of what that meant is like, 308 00:15:41.320 --> 00:15:45.399 across fifteen minutes, there was a timer running and every 309 00:15:45.440 --> 00:15:47.919 ten seconds they entered a new interval, and if that 310 00:15:47.960 --> 00:15:50.240 behavior occurred during the interval, the would essentially fill in 311 00:15:50.240 --> 00:15:52.720 the box to say yes, the behavior occurred, And then 312 00:15:52.720 --> 00:15:55.600 they could calculate, of all those ten second intervals in 313 00:15:55.639 --> 00:16:00.519 a fifteen minute window, what percentage of those had album 314 00:16:00.600 --> 00:16:05.600 that challenging behavior the self harm in the original study occurred. Okay, However, 315 00:16:06.000 --> 00:16:09.000 it quickly became apparent that this is not super sensitive 316 00:16:09.080 --> 00:16:14.000 to changes in actual frequency. So, for example, let's assume 317 00:16:14.080 --> 00:16:17.519 that we are measuring something like self harm and that 318 00:16:17.519 --> 00:16:22.840 that behavior occurred once every single second. That would be 319 00:16:23.039 --> 00:16:27.159 ten instances of that behavior in a ten second interval, 320 00:16:27.240 --> 00:16:31.440 But it would only be recorded as one interval. So 321 00:16:31.519 --> 00:16:34.480 imagine we're going to do a five minute session. There 322 00:16:34.519 --> 00:16:37.759 are three hundred seconds in a five minute session, meaning 323 00:16:37.759 --> 00:16:40.919 that there would be three hundred instances of a behavior 324 00:16:40.919 --> 00:16:44.919 that occurred once every second. But what that would show 325 00:16:45.039 --> 00:16:48.879 on a interval recording would be one hundred percent of intervals, 326 00:16:48.879 --> 00:16:51.120 which is a lot. We understand that's very high. But 327 00:16:51.240 --> 00:16:54.440 take this other example. What if that challenging behavior had 328 00:16:54.440 --> 00:16:58.919 occurred once every ten seconds, but it occurred once every 329 00:16:58.919 --> 00:17:03.919 ten seconds, that would be thirty instances of that behavior. 330 00:17:04.359 --> 00:17:08.960 It would also show one hundred percent of intervals. Three 331 00:17:09.119 --> 00:17:12.920 hundred and thirty are pretty different numbers. 332 00:17:13.319 --> 00:17:14.240 Those are different numbers. 333 00:17:14.279 --> 00:17:17.559 Yeah, that's a wide range for your percentage to show 334 00:17:17.599 --> 00:17:21.799 the exact same amount of challenging behavior occur. From thirty 335 00:17:21.880 --> 00:17:25.799 times in an observation to three hundred times in observation, 336 00:17:26.319 --> 00:17:31.160 that's not very sensitive, right, So essentially, researchers found that 337 00:17:31.240 --> 00:17:35.039 counting each instance of behavior that occurred across the entire 338 00:17:35.119 --> 00:17:39.480 five minutes and then instead converting the number of instances 339 00:17:39.480 --> 00:17:42.559 to a rate per minute, that was a much more 340 00:17:42.599 --> 00:17:48.799 sensitive way to measure behavior. So essentially, when they sort 341 00:17:48.839 --> 00:17:51.319 of made this discovery of like, look, we can really 342 00:17:51.359 --> 00:17:53.519 measure moment to moment changes here that it's going to 343 00:17:53.559 --> 00:17:56.960 be a lot more sensitive to trend if we just 344 00:17:57.079 --> 00:18:00.480 count every instance and then convert it to account permit it. 345 00:18:01.160 --> 00:18:05.279 That became the standard for how people generally conducted these. 346 00:18:05.319 --> 00:18:10.000 The overwhelming majority of FA studies published now use a 347 00:18:10.079 --> 00:18:13.480 rate permittent instead of the interval recording method because it's 348 00:18:13.480 --> 00:18:16.519 so much more sensitive at capturing the changes and rates 349 00:18:16.519 --> 00:18:19.200 of those behaviors and the magnitude of the changes in 350 00:18:19.240 --> 00:18:19.960 those rates. 351 00:18:20.440 --> 00:18:24.960 Yeah, makes sense. Now. The third change is the condition ordering. 352 00:18:25.240 --> 00:18:28.119 So in the original study and for a while, the 353 00:18:28.240 --> 00:18:30.599 order in which conditions were presented was either random or 354 00:18:30.640 --> 00:18:33.119 quasi random. And there were some exceptions with this, and 355 00:18:33.119 --> 00:18:34.960 we'll talk about that in a little bit, but basically 356 00:18:35.000 --> 00:18:36.440 it was like, you know, when we talk about the 357 00:18:36.599 --> 00:18:40.079 escape alone, play and control conditions, what would happen is 358 00:18:40.079 --> 00:18:43.440 they would randomize that. Yeah, so it wouldn't be escape alone, 359 00:18:43.559 --> 00:18:47.119 play control, escape, you know, whatever. It would go escape, escape, 360 00:18:47.400 --> 00:18:50.640 attention alone. You know, like, it would randomize it in 361 00:18:50.680 --> 00:18:53.640 lots of different ways. But in nineteen eighty four, it 362 00:18:53.640 --> 00:18:56.079 would suggested that the order in which sequences were presented 363 00:18:56.119 --> 00:18:59.680 might be important for obtaining clear results, And finally, in 364 00:18:59.720 --> 00:19:04.160 twenty thirteen, a study by Hammond and friends clearly empirically 365 00:19:04.200 --> 00:19:07.559 showed that results were clear and obtained more quickly if 366 00:19:07.599 --> 00:19:12.000 they were presented in the order ignore, attention, play and demand, 367 00:19:12.359 --> 00:19:15.759 So ignore being the alone condition, attention being the social disapproval, 368 00:19:16.000 --> 00:19:17.960 play being the toy play that we talked about earlier, 369 00:19:18.079 --> 00:19:20.559 and then the band being the escape or demand condition. 370 00:19:20.880 --> 00:19:23.920 This has become the generally accepted standard procedure for conducting 371 00:19:23.920 --> 00:19:24.720 an FA. 372 00:19:25.160 --> 00:19:28.319 Yes, with some exceptions, but yes. Essentially, what they sort 373 00:19:28.359 --> 00:19:32.680 of realized was how you end one condition potentially set 374 00:19:32.799 --> 00:19:35.480 up how that person was kind of activated to go 375 00:19:35.519 --> 00:19:38.480 into the next one, and that you would have some 376 00:19:38.559 --> 00:19:41.480 carryover effect if you weren't actually planning these out very 377 00:19:41.519 --> 00:19:44.920 strategically right, like what was the mo for one became 378 00:19:44.960 --> 00:19:47.119 the setting event for the next one, and so you 379 00:19:47.200 --> 00:19:51.759 might accidentally create some artifacts by having the sequence set 380 00:19:51.839 --> 00:19:57.799 up to accidentally promote or inhibit the likelihood of some 381 00:19:57.839 --> 00:20:00.920 of those behaviors occurring. And so they did an analysis 382 00:20:00.960 --> 00:20:05.480 of these fixed versus random patterns and empirically showed, yes, 383 00:20:05.559 --> 00:20:08.079 you actually if you do this fixed pattern you are 384 00:20:08.160 --> 00:20:13.119 more likely to obtain differentiated results where data were primarily 385 00:20:13.160 --> 00:20:16.119 observed in only one condition, and you are more likely 386 00:20:16.160 --> 00:20:19.759 to obtain that outcome more quickly than if you had 387 00:20:19.799 --> 00:20:22.759 those randomized sequences. And so yeah, it just became like 388 00:20:22.799 --> 00:20:25.279 everyone's like, oh great, we'll do that from now on, 389 00:20:25.400 --> 00:20:29.319 Like that sequence moving forward makes sense to me perfect 390 00:20:29.359 --> 00:20:32.079 And so that's generally the one of the refinements. So 391 00:20:32.359 --> 00:20:34.319 essentially the three of you we've numerated so far from 392 00:20:34.359 --> 00:20:36.680 the original study. That changed were that they reduced from 393 00:20:36.680 --> 00:20:39.480 fifteen minutes to five minutes, They started counting every instance 394 00:20:39.480 --> 00:20:41.880 of the behavior and converting that to a rate per minute, 395 00:20:41.920 --> 00:20:44.759 and then finally that the order in which they presented 396 00:20:44.799 --> 00:20:48.880 the provocative situations became fixed because it allowed for clearer 397 00:20:48.920 --> 00:20:51.759 differentiating the results. And that became essentially if you pick 398 00:20:51.839 --> 00:20:55.319 up a journal article in the last fifteen years or so, 399 00:20:55.359 --> 00:20:58.359 maybe ten years since all these happened, like those will 400 00:20:58.359 --> 00:21:01.440 be essentially what you'll see most of the time when 401 00:21:01.480 --> 00:21:04.920