 Hey everyone, it's Professor Howard. So this time around what I want to talk about is when we're examining human behavior as a phenomenon of interest, what does that mean for us as scientists? And we have kind of a difficult time because when we're talking about understanding human behavior, when we're trying to understand people, we come steeped in this history, we have cultures where we learn that people behave a certain way because of certain reasons, and it can sometimes be difficult to approach that from an empirical or scientific perspective. So what I want to talk about this time around is what does it mean to be a scientist of people, to be a scientist who understands the behavior of people from an epistemological perspective. So I want to begin here by talking about the levels of scientific understanding so that we can then see how psychology and behavior analysis fit within that larger domain of sciences. When we're talking about science, we're talking about the stuff of the scientific method we're talking about a principle. And if we are considering psychology as a whole as a field of study about people of behavioral processes and we want to understand that, then we know that psychology is a science because it shares a lot of characteristics with other types of sciences. So sciences in general have very precise and careful measurement. We want to understand the phenomenon of interest. We as scientists are committed to repeated experimentation, which means we find a way to ask questions. We formulate hypotheses. We want to understand what's going on with that particular phenomenon and we do that experiment again and again and again under different experiences, under slightly different circumstances to see how universal is a principle. How universal does that rule hold true? We also want to make sure that as scientists we're always exposing ourselves to peer review where we actually show other people our methodology, show them our results, and invite them to give us feedback about what we've done in a way to improve the way that we ask questions in the future. The whole reason that we have this scientific method, the whole reason we approach understanding people from a scientific perspective, is because we know that we have very personal biases about people and the way that people function. So we want to make sure that as a science, we're trying to minimize our personal biases about a principle, about the phenomenon of interest, then we want to make sure that we are correcting any earlier mistakes that we have in our field. There will be a few times throughout the semester where you really have to question and challenge some things that you believe to know about content to make sure that you can approach it as a scientist with an open mind and that's very difficult. Now there's a question of whether behavior analysis is psychology. There are some behavior analysts who are very hard-line and don't believe that we should actually fit within that category of psychology. If psychology is the science of behavior and mental processes some behavior analysts say that we're not really in that camp they say. They say well we really only are interested in the behavior part right and that's a kind of methodological behaviorist approach. So I'm going to move forward understanding that we as behavior analysts we do align very closely with psychology as a mainstream field of study. The difference being that we are just an extremely rigorous form of psychology. So remember we mentioned before that behavior analysts do not say that private events like thoughts and feelings don't exist. We don't say that mental processes don't exist. We simply do not use those as causal variables for the phenomenon that we do see happen. So as I mentioned behavior analysis is very very rigorous. We have a heavy emphasis on directly observable phenomenon that's why we focus on behavior rather than private events like feelings and thoughts as our dependent variable or as our independent variable or even as mitigating features. We have very very tight methods of achieving experimental control. We use something called within subjects design and we're going to come back to that in a future topic but for now we want you to understand that we are very very interested in making sure that we are demonstrating experimental control when we're conducting our studies. And then finally we want to make sure that we have this ongoing critical evaluation of our independent variable and our dependent variable so that we can be sure and confident in the data that we're collecting. Now when we talk about the goals of science any science it doesn't matter what it is. We're talking about four different levels of scientific understanding starting from the top starting from least rigorous. There's description. Description is really just about telling people what's happening what is the phenomenon. So I wake up in the morning and I say the sky is blue. That's a description it doesn't tell me why it doesn't tell me how it just says what is and that's a beautiful place to start the experimental process but it really tells us very little. It's telling us what's happening not why not how not when is it going to happen again just what is. Next we come up with a potential explanation for why it's happening so I could wake up in the morning and say the sky is blue that must mean that God is happy. Now I have come up with a theory right I've come up with a potential hypothesis for why this is happening and this is an explanation of a particular kind of value I can't say it's true I'm just positing a reason for why this should be. Next we have prediction so that's the next most rigorous right it's not perfect but it's pretty good because with prediction we have this way of saying like if you see that this thing is present then you should also see the other thing is present. So if you know one variable you can predict the presence or absence of another variable prediction is pretty pretty good and remember we talked a little bit ago about the fact that as a science behavior analysis has those two goals we want to be able to predict when behavior will happen and if we're meeting our gold standard we're going to get to the fourth level of scientific understanding control but not like a 1984 like twirling your mustache mad science control we're talking about can we create the circumstances to produce the effect we want right so for instance with prediction I could say that I predict if a mother smokes when she's pregnant then her child is likely to be born underweight right that's a prediction there are some mothers who smoke when they're pregnant and their babies are born real healthy there are some mothers who don't smoke when they're pregnant and their babies are born real small that's a prediction it doesn't necessarily hold true but as a scientist if I want to be able to demonstrate that there's a strong causal relation between smoking when pregnant and underweight births I would need to demonstrate that only when or when and only when mothers smoke when they're pregnant do we see underweight births now of course there's some ethical concerns there you definitely couldn't run this study very easily which is why we do sometimes run our experiments on non-human animals to see if those relationships exist in this particular case maybe that level of scientific control is not necessary maybe we're comfortable with a kind of strong predictive value but there are definitely circumstances under which we want to be able to demonstrate with rigor we want to be able to demonstrate that there's control between two variables if I'm working with a client who has some very challenging behavior I want to be able to demonstrate that when I teach that client an alternative skill that allows them to contact reinforcement and they're earning reinforcement for that alternative skill that the problem behavior is going to decrease in frequency so depending on what your goal is in terms of your level of understanding you're going to look at these four different goals remember from least rigorous to most rigorous description what's going on explanation a possible reason why it's happening prediction allows you to determine when it might happen again because if you know this then you could potentially predict the next and then finally control being able to actually produce the phenomenon of interest be sure to check out the next video where we're going to be talking about what correlations tell us and how they're an example of prediction