 Hey everyone, welcome back. It's Veronica Howard. We are going to talk about the larger categories of reinforcement and punishment. So before we get into talking about primary and conditioned and generalized reinforcers and punishers, just want to review some basic vocabulary. Remember that when we're talking about reinforcement and punishment, a reinforcer is any event that follows behavior and changes the future rate of that behavior. Specifically, it increases the rate of behavior. So the event that we're talking about here is something that the environment, or something that the behavior produces in the environment, which then strengthens and drives that behavior again in the future. A punisher is kind of the opposite. It's again something that has changed in the environment, something that the environment produces, but this event is presumed to be aversive because it decreases the future rate of that behavior. Now, we generally, we can evaluate this based on the effects that it produces, but where do these stimuli come from? And this is where we start talking about the classes of stimuli. For this part of the presentation, I'm just going to focus on reinforcers for now, but bear in mind that this also applies to punishers. When we talk about a primary reinforcer or punisher, it's any event that loses its effectiveness only temporarily and only through satiation. So satiation is when you've had a lot of something and then it sort of loses its effectiveness. Primary reinforcers tend to be tied into their evolutionary values. So we might say, for instance, that food or access to warmth or even contact with a sexual partner, all of these could be primary reinforcers. Remember, these are tied to that evolutionary value. And we're kind of programmed biologically to seek these out. A reinforcer is primary reinforcers effective without pairing. You don't need to have any particular learning history for this to be effective. It just sort of is on its own. We have these, this biological predisposition to seek out these stimuli, just like we have the biological predisposition to avoid or escape aversive stimuli like pain. However, remember, these are tied into that biological component. You can, uncondition or you can make a primary reinforcer aversive through pairing, through unpleasant experiences. Now, for example, if you want to determine if it's a primary reinforcer, one good way of asking that is, what would be a reinforcer for a newborn? Well, warmth or comfort or exposure to a caregiver, food, all of these are very powerful reinforcers. And we know that these are kind of tied into survival. We could also do experiments to determine whether a stimulus function is a reinforcer. So even for a newborn, we can do a simple experiment. And we could look to see whether a child is, for instance, reinforced for one response versus another. For instance, imagine that we have a response, suckling is an innate human response. We're born with this particular reflex. Can a reflex be maintained by consequences? Well, yeah, I think we can demonstrate that. So for instance, we have one side of mom's body, maybe one breast produces milk, and the other breast does not produce milk. We then may ask if, if the baby is more likely to choose, if given an option to be held on the right side, on the breast that produces milk, or on the left side, on the breast that produces milk. And if we see that the baby prefers to be held on the right side, maybe they're less fussy, or maybe they will stop crying if they're held on that side, we could demonstrate that the baby has a preference there. And we might assume that the baby will have a preference for the right side, because that's the side that produces milk. If the baby is not full, they are probably more likely to choose that side. And we can actually evaluate whether other innate responses are sensitive to environmental contingencies. So for instance, we know that humans are born with a whole suite, a whole complement of different responses at birth. We have all of these different abilities. For instance, a grasping reflex of someone touches the palm of our hand, we squeeze and hold on. We have a startle reflex where if you drop the baby, they'll kind of put their arms out to protect their head from being damaged. If you rub the cheek of an infant, they'll turn and rotate toward that feeling. It's a rooting response to try to find the breast and search for a nibble for food. Babies have a very rudimentary kind of stepping response or a stepping reflex. And even a basic suckling reflex, all of these responses, we are born with. They're innate. They're coded into us from the point that we are born. They help us survive. But I think as adults, we would find this kind of unusual. If someone were to come over and stroke our cheek, we probably wouldn't rotate over in search of sustenance or food. We might be very offended. And that's evidence that though we have these innate reflexes, they kind of die out over time, especially as they don't contact any kind of environmental reinforcement. So we may have very basic primary rudimentary responses, but they only keep their value for a little bit. And primary reinforcers and punishers remember they have that innate value, but that value can change by environmental experience. And we also see that when they lose their effectiveness, they can lose effectiveness temporarily. They could continue to be pretty effective over time. The value of them can also change. So primary reinforcers and punishers, just one type of reinforcer. I mentioned pairing before, and that's because a conditioned reinforcer is a type of reinforcer that loses effectiveness by being unpaired with a stimulus, which then means it gains its power through its association with another stimulus. This type of reinforcer is typically associated with one backup reinforcer. So you have one thing that this particular stimulus predicts or can be traded for, things like that. I'm going to post a link in the description of this video so you can see some examples of how to establish conditioned reinforcers. Essentially what you need to do to establish a conditioned reinforcer is to make sure that a neutral stimulus reliably proceeds and predicts the delivery of another stimulus. It could be another conditioned stimulus. It could be a primary stimulus. It could be something else. Essentially a neutral stimulus has to proceed and predict that a backup reinforcer will be available. And a backup reinforcer is anything that makes a conditioned or generalized reinforcer effective. Now they may be, like I said, primary reinforcers. They may also be other conditioned or generalized reinforcers. Clickers are a perfect example of that. I have one right here on my desk. A clicker is a tool that's often used to mark or communicate to a learner that the response is the one that will contact reinforcements. So if I go and then I deliver a backup reinforcer, I do that enough times that eventually the click itself will predict that the backup reinforcer is coming. That's how you establish a conditioned reinforcer. So conditioned reinforcers, the way you do this, you start with a neutral stimulus, you then you provide the neutral stimulus and then you deliver the backup reinforcer. Your textbook, your reading describes the neutral stimulus like pairing it with a backup reinforcer, but pairing is not enough. This is something that I want you to bear in mind. Pairing is not enough. You have to make sure that the stimulus occurs first. The neutral stimulus must always precede the backup reinforcer because if the order is not right, that conditioned stimulus will not take on value. And I'm going to post some more resources about clicker training so that you get a better idea of how to. But clicker training, if you have a stimulus, if you have a clicking noise and a treat, if they occur at the same time, the learner is going to pay attention to the treat. But if you have the clicking noise and then you deliver the treat, timing matters, then the clicker will acquire value. If you just deliver them at the same time, learner is not going to pay attention to the neutral stimulus. Timing matters. Conditioned reinforcers can also gain their value through verbal rules. And I want to warn you that this is kind of advanced. This is highly theoretical and don't rely on this working. But we do know that a verbal rule can be effective. So imagine I'm conducting a parent training workshop and I say to parents who are having some difficulty, you know what, we're going to use extinction. We're going to decrease this problem behavior, but it's going to get worse before it gets better, right? Because we know that with extinction, you get an extinction burst and a lot of undesirable behavior occurs. Then those parents go back into their homes and they're working with their kids who are having some behavior problems. And they say, you know what, I am not going to reinforce that response that has previously contacted reinforcement. The kid just has a meltdown. Now, previously, this tantrum, this meltdown has been really aversive. But remember, I was training them and I said it's going to get worse before it gets better. And in this case, perhaps seeing that kid have a little bit of a moment, perhaps seeing that situation means that the tantrum itself could be a conditioned reinforcer because the tantrum could communicate that you're on the right track. You have identified the reinforcer and you are effectively withholding it. Now, when we talk about conditioned reinforcers, the reason that we typically use them is because they are very, very effective. They're very quick. They're very portable. They're very useful. They're handy. Imagine, for instance, that if I'm going to post a link to this beautiful tag teach video, if you had to deliver a reinforcer, like a primary reinforcer, like a bite of food, every time a target behavior occurs, but it's a really fast-paced, really challenging behavior. The delivery of the reinforcer is actually going to be a challenge. It's going to be a barrier to effective delivery of reinforcement. So, having a conditioned or generalized reinforcer, something that's quick and easy to handle, like the clicker, keeps the delivery of reinforcement from being a barrier to the teaching. We also use other kinds of conditioned or generalized reinforcers. So, for instance, praise or seeing other people being happy. Is there anything innately valuable in seeing other people happy? Well, I would say no. I would say seeing other people happy is not itself really that valuable. It's kind of a weird response for seeing a lot of their teeth, but it has value because it tends to be a very early conditioned reinforcer. We learned very, very early on that when other people are happy, they're more likely to do good things for us. They are in a better mood, which means we're maybe going to do some fun things. That stimulus gets associated with fun times, good times, and it tends to be associated with lots of backup reinforcers. So, we're smiling when we're happy and when we're happy, we're loving our family. We're taking care of people. So, we can see that that stimulus, smiling, or other stimuli that we come to think of as being very social indicators, things like praise, these become very effective generalized reinforcers because they're associated with many backup reinforcers, but they are not in and of themselves naturally reinforcing. Okay. So, moving on, the last one that we want to talk about is a generalized reinforcer. Some people will also call this a conditioned generalized reinforcer because a generalized reinforcer is just like a conditioned reinforcer. It's a neutral stimulus that takes on value through its association with a backup, but what makes generalized reinforcers different and special is because they're associated with many, many backup reinforcers. So, remember, in the praise and smiling example, this is associated with lots and lots of backup reinforcers. So, praise or smiling or good times would be a generalized reinforcer. If it was associated with just one thing, we would call it a conditioned reinforcer. We also see that there are certain kinds of conditioned punishers and we're probably very familiar with these. So, a conditioned punisher is any stimulus that reliably proceeds and predicts the delivery of a primary punisher. So, something that is present immediately prior to when punishment will occur can itself become a conditioned punisher. For instance, if your parent shouts and reprimands and this reliably proceeds and predicts that there's going to be some form of verbal punishment, then we're talking about the yelling itself being a conditioned punisher and the spanking, the pain that's delivered on the learner would be the primary punisher. The conditioned punisher takes on its value through association with the primary punisher. We could even go one step further to say that conditioned punisher, if it's associated with lots of backup punishers, for instance, criticism or anger might not be just a conditioned punisher. It might be a generalized punisher because criticism, anger, yelling, they tend to be associated with lots of backup punishers that we've talked about in the class before. Anger tends to be associated with the loss of attention or being ignored. It could be associated with being spanked. It could be associated with grounding or the losing of privileges. It could be associated with the loss of a job or resources. So, we see that when we have lots and lots of backup punishers, that's when we develop a generalized punisher. Now, another way to think about generalized reinforcers is what does it get you, right? And there's one stimulus that is almost a universal generalized reinforcer. Money is the ultimate generalized reinforcer. Money in and of itself, it's not that great. It's kind of smelly. It's covered in germs. It's got weird textures. Money itself is gross, but money is amazing. We love money because it gets us all kinds of stuff. There's almost nothing that money can't get you. So, how then did we take on? How did we learn that money is a good thing? Well, because we have that experience. We have a system. We have an economy, if you will, where we've learned that by taking this weird piece of green paper and giving it to people, we can in turn receive goods. We can receive services. We can get treats and rewards and whatever. Speaking of economy, money really beautifully describes this idea of a token economy. And a token economy is a system where learners can earn conditioned or generalized consequences, conditioned or generalized reinforcers. And often in a token economy, we'll do things like tokens, like poker chips, like coins. And those are delivered immediately contingent on a response. Participants can then accumulate those tokens or those points or those, whatever the conditioned reinforcer is, and exchange them for other things. If they can earn the points and exchange them for one other thing, then you're looking at a conditioned reinforcer. If they can take those points or those tokens or those chips and they can exchange them for lots of other stuff, like access to video games or trips out into the community or, you know, lots of different things. And we're looking at a generalized reinforcer. Token economies typically have that neutral stimulus that acquires the value because it can be exchanged. It typically has a number of responses that you can earn points or possibly even lose points for. And then ultimately it must, must have a menu of backup reinforcers. That menu of backup reinforcers is what makes a token economy effective. So make sure that you're checking out the resources on token economies, which are fantastic in a wonderful way, a wonderful clinical application of some of these procedures. And let me know if you guys have any questions. I'll see you next time.