 Okay, so returning to our description of circular reasoning, I want to make sure that I'm very clear that when we're talking about circular reasoning or circular logic, what I mean is that it's the situation where the hypothetical construct or the private event is considered both the cause of and evidence of an observed phenomenon. Now keep in mind that is exactly why most behavior analysts, even modern radical behaviorists tend not to use things like self-report measurements or private events like attitudes or thoughts or feelings as the primary explanatory variable in a lot of our studies. We tend to look at the environment in ways in which the environment changes around the organism to better understand why they're making those choices. So this then highlights some of the really big differences between behavior analysis and related fields. For instance, I'm going to just give you a quick breakdown of the differences between most mainstream psychology and that even itself is misleading because I'm making a broad generalization about an entire field of study, ignoring the fact that there are many different subcategories and subdomains of psychology. So take this with a grain of salt. Some of the big differences between behavior analysis and mainstream psychology include things like the fact that psychology uses a lot of indirect self-report measures like surveys or direct observation in their studies, meaning that they consider both direct observation and self-report surveys to be worthy independent variables, whereas we'll see some differences in behavior analysis. Mainstream psychology self-report tends to be used as a proxy of actual behavior. So when I say it's a proxy, what I mean is I might ask you, are you likely to attend a big party if you're going alone? I don't actually go and watch you observing the big party. I ask you whether you would or not. And your reporting of whether or not you would is taken at face value. We don't always question in the same way that we should whether a person would actually do that. And then also that some studies in psychology can be published without any kind of self-report observation. What I mean by that is a lot of studies in psychology tend to focus on relationships between two self-report variables. What you say that you would do in this circumstance and what you report on a different metric like your personality. As opposed to behavior analysis where for the most part our published studies only use direct observation. It's one of the core kind of foundational principles of our field that we only focus on direct observation. That is changing a little bit. We're starting to see a few more studies being published but for the most part when we're talking about clinical work, we're talking about only looking at observable client behavior. We also see that we do use self-report but we tend to use it as a tertiary or a secondary measure to the direct observation so I can see how the intervention is working, like actually observe the learner's response but then I might ask the learner about their experiences and what caused those responses. The primary and most important variable that we're looking at is the observation but we supplement that with self-report. Then we do this a lot when we're talking about social validity and when we're looking at quality measurements, when we start talking about reliability and social validity and so on and so forth. That's where I want you to keep in mind where we use self-report and then finally many behavior analytic journals will not publish a research study if it doesn't have direct observation so you can see that there's this really big difference between most mainstream psychology and behavior analysis. In this topic, I gave a few resources from other sources about why people choose to be behavior analysts, like what is it that behavior analysis offers to them that other fields don't and what I think this description helps offer, taking into account the private events and the way in which we interact with the learner, I see this idea of believing self-reports and the fact that self-reports don't always predict public behavior as being a big weakness in social science where we're trying to really understand what's happening to the learner. For the most part, we've gotten to this point where when the learner doesn't behave in the way that we need them to and when the learner maybe doesn't complete assigned tasks or if they thinking like of myself, I don't go to the gym as often as I should. There's two ways that you could explain that behavior. You could see it either as a sign of weak willpower or some sort of private event that is lacking for that learner or you could see it as a product of the environment. And for me, the reason that I choose to be a behavior analyst is because I want to try to first understand and explain and change the environment before I put blame on the learner because for the most part, it's kind of like this really beautiful quote here that I have on screen. When a flower doesn't grow, you fix the environment in which the flower grows, not the flower itself. We understand that for the most part, people will do the best that they can under the circumstances that they find themselves. And I consider behavior analysis to be perhaps one of the most hopeful sciences because we understand that even if behavior is not where we want it to be, even if it seems as if the learner is beyond hope, what's important is that we can help, even in a very small way, we can help change the environment to better support the learner. Even if we can't get them from here to 100, we can change the environment maybe to get them from here to 60 or here to 40. And doing this rather than trying to explore what variables exist inside the organism, what reason that they're broken, focusing on ways the environment can better support the learner, to me seems like the most helpful approach to solving a lot of the problems that we have for us as individuals and as a culture. So I hope that this helps clarify a little bit about what behavior analysis is and what our opinions about private events are. If you have any questions, please, of course, make sure that you post them in the great question form. But otherwise, I look forward to hearing what you have to say, either here or in the comments. I'll see you guys next time. Hey, everyone, it's Professor Howard. So I want to talk a little bit about what makes behavior analysis different than some related dimensions or some related domains of psychology as a whole. The way in which behavior analysts approach understanding behaviors is very different. And I think the hardest part about this particular course is the fact that we're going to have to talk about things differently. We, all of us, myself, yourself, anyone watching this video has grown up in a culture where, for the most part, we tend to explain the decisions that we make using a lot of internal variables. Like, why did I get up this morning and make a cup of coffee so I could come in here and record videos with you guys? Because I wanted to. You know, why did I choose to become an instructor and teach behavior analysis to people? Because I wanted to. And the way in which behavior analysts approach that is, well, then we have to account for why you want to, right? The way a psychologist might approach that could be looking at things like, what about my personal learning history? What about my personality? What about certain, if we're talking about biopsychology, are there any particular ways in which my physiology prepare me for that and make me well suited for that? But behavior analysts, we want to tie it back to the environment and we want to tie it back to what we've experienced in that environment to help explain that. So when we say that Skinner was different from earlier researchers, remember that Watson and all of the methodological behaviors were very, very clear that private events did not matter. They were explanatory fictions and we should not use them in science. That particular opinion gets ascribed or attributed to Skinner a lot more than it should and Skinner did not believe that. Skinner was a radical behaviorist. He believed that we have to account for all behavior, even private behavior. But what Skinner did say is that private events like thoughts and feelings, et cetera, while they exist, we have to still explain them. The problem is that those private events, like whether or not you want to do something, whether you have an attitude towards something, you still have to explain it. You have to explain where it came from, you have to account for it because ultimately, we're deterministic, we believe that things have causes. But we also know that when you look at the conditional probability between when a private event occurs and when the behavior occurs, the probability can be pretty low. So Skinner advocated for something we call the principle of public events, in this idea that while we could use the relationship between a private event, like whether I wanted to get up and make coffee this morning, and whether I actually made coffee this morning, whether we could look at that conditional probability, yes we could, but I can also look at the environment and see if there are environmental variables that are observable and objective to figure out whether or not that would have happened. So for instance, imagine for a moment that because you can't possibly know, right? You're watching the video, you don't know where I am or what time it is here when I'm recording the video. What if it's one in the morning, right? What if when I say I got up, I took a late nap and it was eight or nine PM, you look to see what time of day is it? So it's one in the morning. We could look at the conditional probability of me making coffee at 1 AM versus say 8 AM when most people, most Americans get up and get ready and start their day or six or seven AM or whatever. That's one way that we can look at public events to help explain behavior. We could look at things like, what's my relative level of energy before drinking coffee versus after drinking coffee, right? So if I'm waking up and I'm like, going into the kitchen to make that coffee and I drink it and then suddenly I'm chipper and et cetera, you can see the change in my behavior, which is, right, I am a part of my environment. You could see if that change, that environmental variable is added following the drinking coffee. And we start looking at the ways in which our environment changes as a function of the behavior. And that's how we want to explain behavior. Now, this one is particularly important. This idea of not using private events like self-report feelings, that idea of avoiding what we might call a mentalistic explanation is actually quite important from Skinner's perspective. Here's why. Skinner says that we have to really be careful about mentalistic explanations because they feel very self-satisfying for us. We grew up in a culture where we use a lot of mentalistic explanations, but they're flawed. Let me give you an example. Imagine that you're sitting in class and you suddenly hear the fire alarm go off. If you believe that there's a mentalistic mediator for that, then hearing the fire alarm should elicit from the organism. Remember, pulling out or drawing out this reflex of reaction of anxiety. And then that anxiety should set the occasion for us to leave the room. But if you've ever lived through a fire alarm situation, then you know that sometimes a person can hear a fire alarm and have no reaction. A person can also leave the room, right? The ultimate product is leaving the room and you can leave the room under different circumstances. So Skinner says rather than seeing this mentalistic explanation as a mediating stimulus, what if instead that fire alarm sets the occasion not only for the organism to experience anxiety, which is one response, but also to leave the room to escape from the fire alarm situation, which is a different response. And what if these responses can occur independently of one another, right? So then that gives us the power to see that. There are situations under which the fire alarm can go off and we can experience anxiety and not actually leave the room. And there are situations in which the fire alarm can go off and we experience no anxiety and we leave the room. If the fire alarm were to go on the fritz and people were to send out an announcement that says, oh, by the way, there's gonna be some problems this week. If you hear the fire alarm, just wait two minutes before actually evacuating. You might hear the fire alarm and everyone go, oh, okay, well, I'll just get my bags and go, right? And I'm taking my coffee with me because coffee's very important on my researcher. You can have one without the other. And we can look at this a different way, right? So we can think about this relationship that we get into as being problematic because then, sorry, let me go back a second here. This idea that you can have different responses is very important. It's important because if we always have to have the anxiety for the other thing to happen, right? If we always have to experience anxiety for us to leave the situation, then it makes it less strong. You can have a very weak relationship, a very weak correlation between anxiety and escape. And then when you have that weak correlation, the whole explanatory variable breaks down. Now, from this, you can also start getting into what we call circular reasoning. Let me give you a different example. You have a young person, you're dropping them off for school, they're new to school, so we're going to preschool. They take Steve to preschool. When you drop Steve off at preschool, they cry, right? And we might say, well, why is Steve crying? Well, because Steve's feeling separation anxiety, right? This young baby is having some trouble. It's very difficult. You can see the crying, but you can't actually see or touch or feel or measure the anxiety. The anxiety is deep within the organism. You can't get in there, right? But how do we know that the anxiety is happening? Well, you know that the anxiety is happening because the crying is happening, right? And we tend to explain the things that we see publicly with variables that we can't see, things that we see privately. And we have this tautology here. We have this problem because when you use a private event to explain something public, not only is there a problem with the conditional probability, there's every probability that Steve is crying for some other reason, right? Not related to separation anxiety, but that circular reasoning means that we can never break them apart because the public event will always be considered proof of the private event and the private event will always be considered the cause of the public event. And if you can't get at and measure and confirm that the private event is happening 100% of the time that the public event is occurring, then your explanation is ultimately flawed. Put another way, if a person in a meeting with me is always talking over me, right? They're always interrupting me and repeating the thing back to me. I could leave the meeting and be very frustrated and just say, oh, that person's a misogynist, right? They're just so mansplaining at me all the time. So them talking over me is seen as proof of their misogynistic attitude. And their misogynistic attitude is a cause of them talking over me. But what if it's something else? What if there's a third variable like this person has some hearing loss and my particular voice is just at a range where they can't hear it when they're talking over me? There's a third variable which breaks apart my entire proof and makes it more difficult to believe it. So when we're talking about the way in which behavior analysts approach understanding behavior, we always want to be looking at other concrete environmental variables, not because we don't believe that private events exist but because they're so flawed in a scientific explanation of behavior that we can't rely on them fully. And we want to be confident and strong in our explanations and we want the highest conditional probability between independent variables and dependent variables between environmental variables and their products as possible.