[BLANK_AUDIO]. Okay, so now you have the basics of classical conditioning and operating conditioning. So, what I want to do for the rest of the lectures of this week is talk about some other principles related to conditioning and learning. and relate them to issues that are part of real life, and, and ones that I think you'll find interesting. So, we're going to begin with flirting. Why not? and we're going to use flirting to describe a, a concept called schedules of reinforcement. So that sounds very scientific and boring. And I, I thought putting it in the context of flirting would, would help alleviate some of that. so let's go flirt. All right. week 4, lecture 5. Again, in my, in my quest to give you, useful little tidbits from this course "How to Flirt, and Other Forms of Gambling." weird mix? Not really. Flirting's gambling. Think about it. Alright, so. Imagine, you're in some situation. there is somebody that you might like to know better. And, and let's do a little different situation than the one depicted here. Let's just imagine you're in some context whereby, every now and then, you pass by somebody that you might be interested in getting to know better. And that person kind of maybe looks and smiles at your way quite consistently. And so the question I have is, for you to consider, is what should you do? Should you smile back? Should you smile back every time that he smiles at you? If you really wanted to attract that person, what's the best method? Well it turns out the answer can be found in operant conditioning. and in the behavior of rats. So, specifically let me get to that, let me get to the flirting. Lemme start with just giving you a really concrete idea of, of this sort of vanilla, basic, operant conditioning kind of experiment. And a lot of operant conditioning. Used one of these things, called a Skinner box. I told you Skinner invented a bunch of neat things. This is also called, by the way, an operant conditioning chamber, or an operant chamber, most of us know it as a Skinner box. it would work in the following manner it would have some aspects. So specifically a speaker and some lights that you could use as stimuli for the environment. Remember we talked about discriminant stimuli. Well you can turn on a certain light, maybe a green light. And that might tell the rat, oops. That might tell the rat that certain rules are in play. And the red light might say those rules are no longer in play. Or you can present sounds that sort of that thing. So, this is to present some stimulation to the rat. This is a lever, so imagine this as something he can push down on. and that his way of responding and then we have two ways of presenting consequences. one is to deliver food pellets from some food dispenser. So, a little food pellet comes down the other. However a little nastier, is that the floor of these Skinner boxes were actually, little met-, electric grids. Metallic grids that, you could turn on and that would present a mild electric current. all animals, humans included, hate the feeling of electricity. Even when it's not real strong, we hated it. It really, really bothers us. and the rat is no different. So, this is an aversive stimulus, a negative stimulus and so it's way we can deliver punishment. Alright. So, in, just to give you a sense of the, the typical experiment, one of the things you can easily teach the rat to do is to press the bar to get food. and the way it, this would typically be done is first of all this rat would be fed less food than normal. In fact it would be fed less food for a while until it was something like 75 percent of its normal body weight. So, they call this a motivated rat, hungry rat. This is a rat that wants food, and therefore is very motivated to find ways of getting food. And so if you just put the rat in this condition, it being a rat, it will just sniff around. And even if we don't use any stimulation at all, it'll just sniff around and if we just set the rules like this, hey if he presses that level, a food pellet will drop. He will figure that out eventually, just by chance. He'll sniff around, he'll touch things, he'll, you know, do whatever he does, scratch himself for a while over in the corner. And at some point he'll be in this corner of the cage. And he'll happen to touch this lever, maybe trying to sniff the speaker for example. Imagine him sitting up to sniff the speaker. He pushes the lever down, and a food pellet drops, and he'll go right over and eat that food pellet. that single experience will kind of bias him to kind of be around the area where the food pellet dropped. so now he's going to hang around here a little bit more. Which increases the chance that once again, sort of by chance, he will touch this lever and a food pellet will drop. And it wont take very long at all, before the rat has figured out the relationship. And he will literally come over like this rat is demonstrating, push the lever. And go and eat the fruit pellet. Come back, push the lever, go and eat the fruit pellet. So a rat can very quickly learn that contingency. you could also as indicated here add a discriminative stimulus. So, you could tell the rat, hey well you don't have to tell the rat. You could turn the green light on and when the green light's on when the rat presses the lever he gets a food pellet. Then you could turn that off and turn the red light on. The rat will still press the level hoping for food, but it won't get any. And after a few trials, it will learn, wow when the red light's on, I don't get the food. So, it will stop pressing the bar when the red light's on. But as soon as you turn the green light on, it'll come over and press the bar and get the food. Okay so this is your basic learning experiment. And just about any animal, can, can pick that up very quickly. okay so lets start kind of there. That's, that's something we call contingent conditioning. So, we're giving rewards, or could of been punishments. But in this case we're giving rewards. Contingent on their behavior. Now, how do we reward them? So, what I'd described so far, let me go back here for a second. What I'd described so far is that every time the rat pressed the lever it got a food pellet. But you don't have to do it that way. You can use a ratio system. You could say, even when the green light's on you have to press the lever five times, and on the fifth press you get a food pellet. Okay. Now the, the twist we're going to add here is you can make it exactly every five times or you could make what we're going to call a variable ratio. So approximately every five presses gets you a reward. But sometimes it may be as quick as two presses. And sometimes it may take eight presses. So on average it will be five. But you're never exactly sure when you're going to be rewarded. And so this is going to bring us back to flirting. So, let me show you this graph. This is what we are going to call a variable ratio, that's when that approximately five. This is a fixed ratio when it's exactly five, every fifth press you get one. And what I want to point out is the difference of these curves and they're different in a couple of ways. If it's, a fixed ratio where it's every 5. Where in fact in this case, where I look a little closely at this graph it looks like 1,2,3,4,5,6,7,8,9,10. So, about 10 presses because this is going to show where the reward is. And so, let's say he has to do 10 and then he gets a reward. Well, what you will see is this kind of behavior, he will go 1, 2, 3, 4, 5, 6, 7, 8, 9, 10. Food pellet drops. He gets it, eats it, chews, hangs around for a little while, relaxes. That's this flat period. So over this time, nothing's happening. And then at some point he'll walk over to the bar, and he'll go 1, 2, 3, 4, 5, 6, 7, 8, 9, 10. Food pellet drops. Munches, relaxes, chills out, so you get these little plateaus. And then he'll hit it again. One, two, three, four, five, six, seven, eight, nine, ten. and then he'll relax. And so you get these plateaus followed by bursts, what we call them. Behavior bursts, okay? Typical sort of pattern. But if instead you use that variable ratio and so maybe its on average every ten presses. But again we could say but it could only be two presses and it could only be 18 presses. Now that little bit of mystery, that little bit of uncertainty about when the reward is coming radically changes the behavior. The rat will in fact press a lot more constantly. You wont' see these plateaus. It will press press press press, get's a reward. Eats it, starts right back to pressing again, gets a reward. Eats it, starts right back to pressing, and let's say a reward's not coming. It keeps pressing, keeps pressing, keeps pressing, keeps pressing, until the reward comes. And then it eats it. And then it goes right back. Now why is it going back so quickly? Well it's going back so quickly because it knows sometimes rewards come real quick. Sometimes I only have to press it once or twice. Why is it continuing to press when it's been a long time since it's been rewarded? Well because it has this feeling the thought is, that man, a reward is overdue, it could be here any minute. And so, throughout, what this variable thing does, is it gives the rate the impression that at any point in time, a reward could come if he behaves. And so, he keeps behaving, he keeps pressing the bar. So, these variable ratios produce strong, consistent, constant behavior. What's this mean for flirting? Well, it means if you're now trying to decide how often to smile back, you should adopt a variable ratio. You should tell yourself in your head, "well, he's going to have to smile at me five times before I smile back.". But it won't be exactly five times. Because if it's exactly five times he'll smile exactly one, two, three, four, five and then I'll have to smile back and then he won't smile at me for awhile. Until he's kind of ready to do it again and then he'll smile again real quick. I don't want that. I want him to always be interested in me. And so if that's the case, then you say no, no. He has to smile five but plus or minus four. Sometimes I'll, I'll, if he just smiles once I'll smile back at him. Sometimes he's going to have to smile at me nine times before I smile back. And now there will be that mystery. You'll keep thinking he's going to get a smile, and because he keeps thinking that's possible, he will keep being interested. He will keep smiling, he or she, let me not be presumptuous here. Alright? now that's, that will work. That will produce the strongest behavior. But there's a dark side, there's always a dark side isn't there? And that's the notion of extinction. What extinction means is okay let's say, let's go back to our little rat here. Let's say, he's been doing fine, learning that little reward. But now we suddenly decide where going to change the rules on. That even when the green lights on, we're no longer going to reward them. we're not going to reward them at all, no more food pellets. what will happen once he's been trained that, oh when the green lights on I press the bar, I get a food pellet. Well, when we stop that it's going to take him awhile to figure out that we stopped it. That's a process we call extinction. So, slowly over time, he'll start by pressing the bar and coming over here, no pellet. Dang. Go back, press the bar, you know maybe press it five times if he thinks he has to. Come back. No food pellet. Eventually he'll give up. He'll say well, no sense pressing the bar, no food's coming. But here's the point. If it's a variable ratio we used, it's going to take him a lot longer to give up. He's going to keep thinking no, no, it's, it's a random sort of number of presses, so I better keep pressing because that reward is probably just around the corner. and so it takes him a lot longer to realize that the reward will never come. It becomes much more persistent, the behavior, even under extinction. So what that means, so, so, this is meant to be extinction by the the way. So, let's say you are trying to get this person to smile at you, well now you smile and they give you this. Or the, or if you're more attracted to this kind of human being, then you try to smile and they give you this. So, they're giving you extinction, they're not smiling back. But, how many times do you have to see that before you get the hint? Well, if, if they had been using a variable ratio on you, notice I've kind of switched roles now, you are now the smiler all the time. But if they'd used a variable ratio on you, it's going to take you a lot longer to figure out that, okay, this isn't just one of these trials where they're not smiling back. They're never going to smile back. it's much more resistant to extinction. So, if you choose to use that variable ratio and you're flirting, that's cool. You'll be more likely to get the person interested. But, if you suddenly decide you're not interested and you're never going to smile back. Don't be surprised if it takes that person a whole lot longer to get the point. Okay? Variable ratios are more immune to extinction. Now I've cast this all in terms of flirting but I promised other forms of gambling. I mentioned other forms of gambling because in fact within the gambling system they use variable ratios all the time as their reward system. for example, these sort of one-armed bandits. We have a good example of a, a guy over here. You put in some money. You pull the handle. Well the way these things work is it, it's exactly. This is a Skinner box, this is a human version of a Skinner box. he's pulling the lever. He's pulling the lever. After so many pulls, plus or minus so many. A reward will occur, a jackpot of some sort, of various sizes. but you never know how many lever pulls you have to pull. And you always feel like the next one could be the one that pays off. And in fact there's been sort of funny behaviors where some people who've been sitting at machines for a long, long time. And it hasn't paid off, refuses to leave a machine. They feel so strongly that it's about to pay off any minute. And they have this though that, jeez. I, if I go to use the washroom over here. Then this woman's going to come over and take my spot. She's going to put one coin in, and she's going to get the payoff. I can't let that happen. And so they sit there. Some have even been know to literally used the washroom in their seat, because they refused to leave the one arm bandit. you know that's how strong that resistance to believing the payoff isn't coming is. even things like poker, you know why do we always have that random part of it. You know poker there's a lot of skill involved, but it's also about the cards that you get. That randomness means that somebody isn't always going to win. and it gives you the impression that you could win sometimes and your turn to win might be just around the corner. So, all through gambling they use this. This is why gambling is often described as psychologically addictive. you can literally have trouble not behaving, you know, playing cards, playing the one-armed bandit. Once you've been rewarded a few times at that variable ratio. This is why so many people see it as a dangerous thing, alright? So, if you want more of this here's a few things, so, so gambling and conditioning, is a video that's more specifically about that. This one as well Play to Extinction, a little more provocative but, but it kind of the cool video. I was trying to find you more flirting things. This is a so-called Flirting Expert it's a podcast. It's kind of, it's a little strange, but I think ultimately there's some interesting stuff there. So, hey, you know, I' m trying I'm trying help you with your social life as well. So, if you're interested, check that out. here I decided to get a little creative again. I like to throw different things at you. This is somebody's PhD on gambling and conditioning. I thought it might be interesting for you to see what a PhD actually looks like. You've heard of people writing a PhD, I thought you might like to see one. So there's an actual copy of one student's PhD there you can check out. We'll all have a bunch of hits on his PhD which would be cool. and here's another reading about slot machine specifically and how they incorporate those schedules of reinforcement. So, within these gambling ones, they're not just about gambling. They're really about offering conditioning and schedules of reinforcement. They're just presented in that context, which makes it a little bit more interesting. Okay, cool. So, check those out. And then, when you're ready, come on back, because our next topic is going to be superstition. And why humans show superstitious behavior and how this all relates to, good ol' B F Skinner again. Have a great day. Bye-bye.
