Skip to main content
Loading Explore guides…
Conditioning and Learning · Acedra
Home
Explore
Psychology
Conditioning and Learning
Sign in to save
Chat
If YouTube asks you to sign in, open the video on YouTube.
Watch on YouTube
Reload player
Transcript
156 segments
Copy
Hide video
0:04
Andrea Burgoa: Welcome. Welcome, everyone! Andrea Burgoa: Welcome to our session, our session on the essentials of conditioning and learning. Andrea Burgoa: My name is Andrea Bergoa. I am the higher Ed marketing manager here at Apa publishing. Andrea Burgoa: We welcome you to our session today as everyone is logging in. Please use the chat to tell us where you're joining us from. We'd love to hear from you.
0:28
Andrea Burgoa: I'm going to start with a few housekeeping items, and then I will turn things over to our presenters again. My name is Andrea Brigoa. I'm the higher Ed marketing manager here at Apa, publishing.
0:40
Andrea Burgoa: 1st of all, although all attendees are muted Andrea Burgoa: we have provided multiple modes of communication and interaction for today's sec session. Andrea Burgoa: We do have auto captioning enabled. So if you need closed captions, please use the closed caption button at the bottom of the zoom screen.
1:00
Andrea Burgoa: Chat is open and available for all attendees. If you would like to speak to your fellow attendees, please use chat. However, if you have questions for our panel, please use the Q&A button at the bottom of your screen that allows us to keep track of all the questions that are asked of our panel and find them when our Q. And A. Session at the end of the webinar happens.
1:25
Andrea Burgoa: as soon as it is available, we will be sharing the recording for this webinar via email that will include chat captions and a Chat transcript. So if you feel like, there's something you've missed in chat. It will be there.
1:39
Andrea Burgoa: The recording will also be posted to the Apa Youtube Channel, and we will be sharing the slides and other resources. From this presentation as recognition of your attendance today you will be receiving an attendance certificate and a badge that you can download and share on social media.
1:57
Andrea Burgoa: All right.
2:00
Andrea Burgoa: Now, I'd like to welcome you again to our webinar on teaching complex topics and conditioning and learning. Andrea Burgoa: I'd like to introduce our 2 panelists and the authors of essentials for conditioning and learning. Andrea Burgoa: We have Dr. Michael Domian, a professor of psychology at the University of Texas, at Austin, where he has taught learning for 5 decades. Andrea Burgoa: Dr. Domian, I've lost my notes. Andrea Burgoa: If the recipient of the Do Hub
2:26
Andrea Burgoa: have distinguished scientific contributions award from the Society for Behavioral Neuroscience and Comparative Psychology.
2:34
Andrea Burgoa: We also have Dr. Andrew Delamader, a professor of psychology and neuroscience at Brooklyn College and the graduate center of the City University of New York, where he has taught various courses in learning and behavior theory and in the neurobiology of associative learning. Since 1994, he is the proud outgoing editor for the Journal of Experimental Psychology, animal learning, and Cognition.
2:57
Andrea Burgoa: I will go ahead and turn it over to our presenters. If anyone has any questions? Please let me know.
3:07
Michael Domjan: Well, howdy, folks? And you know, thank you for joining the meeting as Andrea mentioned, I've been teaching
3:18
Michael Domjan: an undergraduate course in conditioning and learning for for a while. Michael Domjan: I'm not particularly good at it, and there are a number of challenges that I would like that I face in teaching this class all the time, and I want to share with you how I've tried to overcome these challenges. And so what
3:49
Michael Domjan: gonna be talking about is
3:56
Michael Domjan: Oh. Michael Domjan: these 3 issues that I face in teaching this class one is that a lot of students think that this is pretty simple, material, and and familiar. After all, Pavlovian conditioning shows up in peanuts, cartoons, and and in some comedy TV shows. Michael Domjan: Another challenge is that students hear about experiments that were done with laboratory animals, and they don't think that the principles are relevant to their personal lives.
4:33
Michael Domjan: And finally, the study of conditioning and learning is rooted in behaviorism which some people in psychology consider to be no longer an important perspective in psychology. So I'm going to talk to you about how I push back against these notions. And let's start with the idea that Michael Domjan: conditioning and learning is pretty simple and everybody is familiar with it.
5:03
Michael Domjan: And so here is a diagram of a basic Pavlovian conditioning experiment that I'm sure talking about people being familiar with this. I'm sure people have seen this diagram lots of times and perhaps use it in their own teaching. The top panel shows a circumstance before conditioning is introduced. A dog is presented with food and salivates on as a Michael Domjan: unconditioned response.
5:30
Michael Domjan: If you ring a bell, the dog does not salivate, then the conditioning procedure is shown down in the bottom left. Here the bell is paired with the presentation of food, and then, after that kind of training procedure. If you then test the dog with the bell, now, all of a sudden he salivates, and lo and behold, you've got conditioned salivation. So that's a familiar story.
5:56
Michael Domjan: The trouble is, this familiar story has a lot of misleading implications. One is that Pavlov didn't actually conduct such an experiment. He didn't discover salivary conditioning, using such a procedure, and he didn't use a bell as a condition stimulus in hardly any of his work.
6:19
Michael Domjan: A second misleading implication, and one that actually Skinner, latched onto and elevated to an axiom, is that Pavlovian conditioning is limited to the conditioning of glandular and physiological responses.
6:36
Michael Domjan: and Skinnett was not interested in smooth muscle responses and physiological changes. He was interested in ways in which an organism interacts with the environment, using arms and legs and skeletal musculature. And so he pretty much relegated Pavlovian conditioning to the hinterlands. In fact, he tried to build a model of learning that totally ignored Pavlovian conditioning.
7:06
Michael Domjan: Another misleading implication is that the condition response looks to be elicited directly by the condition stimulus, and in many situations that's just wrong. The last of these that I put on here is. Michael Domjan: It's pretty. This is a a misleading implication that actually many practitioners of scientists in the field also have trouble with, and that is, if you don't see a.
7:43
Michael Domjan: The diagram suggests that if you don't see a conditional response, learning didn't happen. Michael Domjan: That turns out to be an incorrect inference in a lot of circumstances. So for those students and faculty colleagues who think that conditioning principles are simple and obvious. It's not so simple and not so obvious. So let's look at the second challenge that is making the study of conditioning personally relevant.
8:15
Michael Domjan: Well, I use a lot of human examples in lectures and discussions, and Andy is going to talk more about that aspect of it. And another thing that I started to do is that for each week's unit of work I pick a technical term or a procedure or phenomenon
8:35
Michael Domjan: that we're studying that week. And I asked students to identify a personal experience and report, write up their short paragraph about a personal experience they've had that illustrates this particular concept or or technical term. And I've been
8:55
Michael Domjan: really surprised with how innovative these are. I mean, usually, when I try to think of human examples, I think of similar kind of things. Students come up with all kinds of stuff that actually fits. Another thing that I do is I have them work on what I call a semester long personal modification project in which I ask them to
9:20
Michael Domjan: at the beginning of the semester, identify some behavior that they perform, that they want to modify, that they're not happy with, or something that they want themselves to to do. Michael Domjan: And then, through several, you know assignments we I asked him to try to develop a procedure to modify this behavior, actually carry out the procedure, collect some data and then review their progress at the end of the semester. And so there's
9:53
Michael Domjan: during this process. I encourage them to look through the book, look for relevant procedures, relevant aspects of the book. So it becomes highly, personally relevant, because, you know, they're trying to stop using their iphone last thing at night, which is interfering with their sleep, for example. So they're invested in the behaviors that they wish to change.
10:18
Michael Domjan: Another thing I do is, I ask them to tell me what their view of free will is, and I do this at the beginning of the semester, and at the end of the semester to see how things their opinions have changed. And so this has been quite instructing for me, and I hope it has involved them a great deal more in seeing the relevance of conditioning principles in their personal.
10:42
Michael Domjan: So let's move on to the 3rd one, and that is the challenge or the fact that conditioning is rooted in behaviorism. Michael Domjan: And some people think behaviorism is kind of not relevant anymore. Of course, you know, in the early part of the 20th century behaviorism was a big deal, and a lot of psychology was devoted to behaviorism, behaviorism, and and Michael Domjan: studies of conditioning and learning.
11:19
Michael Domjan: But as we move towards the end of the 20th century, psychology Michael Domjan: became taken over by the cognitive revolution, the content of journals changes, the title of journals became much more cognitive, and cognition came to dominate, I mean experimental psychology, developmental psychology, social psychology, and even conditioning and learning. And the question then becomes, has behaviorism become irrelevant
11:52
Michael Domjan: to psychology? And I'm here to tell you that not at all. Michael Domjan: And my evidence for for that one.
12:03
Michael Domjan: Let's start with this one. This is the winner of the best title for an article ever. This is an article that I stumbled into in the New York times called Your Brain is not for thinking, and it was written by Lisa Feldman Barrett, who also published a book Michael Domjan: 7 and a half lessons about the brain, and that there is one chapter devoted to your brain is not for thinking. Well, if it's not for thinking, what's it for? And her suggestion is that the brain evolved
12:38
Michael Domjan: to enable organisms to cope more effectively with biological challenges to survival. And the question is, how do you cope with those kinds of challenges more effectively. Well, you do so by learning to predict those figuring out ways of predicting what the challenge will be. And so her basic thesis or suggestion is that the brain evolved to enable Pavlovian conditioning, which I thought was really great.
13:07
Michael Domjan: The general proposition that the brain does a lot of things that don't involve what we casually refer to as thinking that proposition has been around for a while. David Eagleman, who was also a neuroscientist, said that there is a looming chasm
13:29
Michael Domjan: between what your brain knows and what your mind is capable of accessing, so that a lot of what you do you do automatically rather than through conscious intent, and so on. And Daniel Wagner is a prominent social psychologist who wrote a book titled The Illusion of Conscious Will, which the title says it all conscious will is an illusion.
13:57
Michael Domjan: And, interestingly enough, William James, who did you know? A prominent psychologist in the 19th century, wrote a book titled habit, in which Michael Domjan: he said that people act out of habit without any consciously formed purpose or anticipation of results. Okay, so habitual behavior, according to him, him, occurs without consciously formed purpose. And it it occurs without the anticipated anticipation of results. Well, one of the more fascinating things about these issues is
14:39
Michael Domjan: that we are. Michael Domjan: we're still working on these issues. So here's a social, prominent social psychologist, Wendy Wood, who was president of the Association for the Association for Psychological Science in 2,003, and she devoted her entire much of her career to empirical investigations of the nature of habits that people have. So she didn't study animals. She just studied people.
15:10
Michael Domjan: and she collected a lot of information, tried to figure out what it all means and what causes habitual behavior. And her conclusion just blew my mind. Her conclusion was that habitual behavior basically occurs because, or is mediated by Thorndyke's law of effect. Now we all know about Edward Thorndyke right and his puzzle box experiments. Michael Domjan: and Thorndyke not only did collect the data, but he also had a theoretical explanation, for why it is that the
15:48
Michael Domjan: kittens that were put in a puzzle box and made an escape response, operated a latch in such a way that they could get out and get a piece of food. Why did they subsequently make that response more quickly on subsequent trials? Well, he said, the reason that the response increased is not because it became associated with the reinforcer. Michael Domjan: It wasn't because they were looking for the reinforcer, or that the behavior was goal directed.
16:20
Michael Domjan: His suggestion was that what was learned was an association between the instrumental response and the stimuli in the presence of which those responses occurred. So these stimuli were typically the cues of being inside the particular puzzle box. So this is the origin of Sr learning.
16:41
Michael Domjan: Okay, it is the origin of Sr learning, which is the foundation of behaviorism. And what I find remarkable is that more than 100 years after Thorndyke formulated this idea, we are reaching the same conclusions, not based on puzzle box experiments, but based on social psychology experiments conducted by
17:08
Michael Domjan: people like Wendy Wood. Okay, so what's another piece of evidence that behaviorism is still very much relevant? Well, this is, I'm going to talk about the relationship between cognition and language. And this is from a review paper that was published in Nature in 2024, Michael Domjan: and
17:33
Michael Domjan: philosophers and scientists have been interested in a relationship between cognition and language. For a long time Plato said, that thought is a silent inner conversation of the soul with itself. Don't you just love that language? I wish we wrote in such poetic terms these days. Chomsky was not so poetic. He said. If there is a severe deficit in language, there will be a severe deficit in thought.
18:02
Michael Domjan: The notion here is that language is necessary, and is a mechanism for thinking Michael Domjan: well what these investigators did is a review of the literature on functional magnetic resonance, image and brain, imaging essentially of human participants under various circumstances, in some involving language, some not, and so on. Michael Domjan: And here are the results. So they combined the neuroimaging data across experiments.
18:39
Michael Domjan: And here's the parts of the brain that are involved in the language network. So those are indicated in red here, and these support language comprehension production. In this middle panel the areas of the brain indicated in blue are part of what's called a multiple demand network, which is what mediates what
19:04
Michael Domjan: we normally think about as cognition, such as executive function problem solving mathematics, some forms of reasoning, and so on. And what's interesting here is that the parts of the brain are that are involved in cognition are not the same as the ones that are involved in language.
19:25
Michael Domjan: Another form of cognition is what's immediate by the theory of mind network, and this involves social reasoning and mentalizing, and those areas also are not the same as the ones that are involved in language. So based on these kinds of data, the authors concluded that language plausibly co-evolved with our thinking and reasoning
19:49
Michael Domjan: capacities, but it only reflects rather than gives rise to the signature. Sophistication of human cognition. So, and that's why their title is language is primary, a tool for communication rather than thinking.
20:06
Michael Domjan: Well, what's interesting about this? This is exactly the Skinnerian behavioristic view of language. So you tell me that behaviorism is dead, and it's certainly alive and well in brain imaging data. Okay, another interesting thing that I ran across is this book by Barbara Tversky, who's prominent cognitive psychology at Princeton University. Michael Domjan: and she the book is titled Mine in motion, and the book is about
20:37
Michael Domjan: how action shapes thinking. So it's a behavior that's causing is shaping cognition, not cognition, generating behavior. She points out that constant motion in space is a given. No wonder it is the foundation of thought. Michael Domjan: Constant motion is the foundation of thought. Action in space came along before language, as did thinking based on action in space.
21:11
Michael Domjan: Another quotation I pulled as spatial thinking is the core of our very existence. And that's nonverbal. And it's based on behavior. It's based on behavior. And you see that in these pictures, you know, babies are trying to figure out where their body ends and the environment starts, and and what's afoot for, and so forth, and so on, and they do so by interacting
21:35
Michael Domjan: behaviorally with different parts of their body. And of course, here's an example of a kid learning that the chair doesn't disappear when you turn around. It's an idea that the kid learns, not from by from language or thinking about it, but but from doing something. Okay. Another piece of evidence that I want to share with you is that. And I think we are living in the
22:02
Michael Domjan: golden age of measuring behavior. We have ways to measure behavior today that Skinner would be really excited about. We have ways of measuring behavior. The behavior is 100 years ago you didn't even dream about. Michael Domjan: Here's a review article by Kaya Barbaro that was published in developmental psychobiology, in which she talks about how technological advances make capturing and analyzing large volumes of daily activity in natural environments possible
22:34
Michael Domjan: from wearable cameras, step trackers, smart homes, smart thread sensors to capture daily activity are literally being woven into the fabrics of our daily lives. The widespread adoption and presence of these sensors allows unprecedented access to the dynamics of activity Michael Domjan: in daily context, in which development happens. Of course, she's real interested in using these wearable devices
23:04
Michael Domjan: to advance the science of developmental psychology. But you can use these to measure anything and businesses. Successful businesses in this day and age are virtuosas in the measurement of behavior retailers and online vendors. They do nothing but collect behavior. They keep track of what you buy when you buy it.
23:29
Michael Domjan: how much you're willing to pay? What's the sequence of your choices as you search through Amazon, whether you put something in your basket to purchase, or you take it out of the basket. If you take it out, it's likely to show up the next time you log into Amazon and so forth. Search engines collect all kinds of data about you because social media
23:54
Michael Domjan: computer made games. And what's interesting about this businesses don't care about what you think. Amazon doesn't care about what you think. Amazon cares about what you do Michael Domjan: and what you do also goes into product design. And one of the more fascinating areas of product design is the iphone, the iphone. I could spend an hour talking about the iphone. But the iphone was designed based on behavioral observations about how
24:33
Michael Domjan: children interacted with prototypes and the behavioral, how they interacted with that follows the principles of conditioning and learning such that you could take each chapter of this book. Michael Domjan: and I'll have a section on what function of the iphone is made possible and is possible to operate through these principles. So my conclusion today for you is behaviorism is alive and well.
25:09
Michael Domjan: and and so is the conditioning and learning and students should. There are lots of reasons why students. Students should be excited about about this. Not all of them are, but I'm doing my best to get them there. Thank you very much. Folks.
25:30
Andrew Delamater: Thanks very much, Mike. Andrew Delamater: Now it's my turn to step in and and and offer some of my thoughts about challenges and teaching this course. But but before I do that, let me see if I can get this thing going. Andrew Delamater: Can everybody see? Actually, I can't see you guys. But Andrew Delamater: hey, Mike, can you see this before I start? Michael Domjan: I can't. I cannot see your slides yet.
25:56
Andrew Delamater: Oh, you cannot. Okay. Oh, alright! I have to share. Let's see. Andrew Delamater: So now I'm sharing. Andrew Delamater: Is this visible to everybody? Michael Domjan: Yeah. Now we go. Andrea Burgoa: Good to go.
26:08
Andrew Delamater: Great. Okay? Good. So one of the things I truly enjoy about studying learning processes is that it's a lifelong activity. And and another thing is that psychologists who study learning processes over the years have really devised very interesting ways of asking questions about the mechanics, of how learning processes work. And before I start my presentation, I just like to
26:36
Andrew Delamater: acknowledge that several years ago, when Mike 1st contacted me about, maybe joining efforts with him and producing this next edition of his book. I was absolutely thrilled because I myself have learned quite a lot about how to write, how to write effectively from Mike, who I regard as one of the masters of this in this discipline. So so when when we got together and we
27:02
Andrew Delamater: started a brainstorm on how we were going to develop this book. We had a couple of important goals in writing this book in in this edition, and and one was to present Andrew Delamater: research findings and explanatory concepts in this field that we take to be very essential or very critically important. But we wanted to present this stuff in a very highly accessible way.
27:26
Andrew Delamater: So so I think we've accomplished that quite well. But at the same time we also wanted to update the reader on some of the latest discoveries, and we wanted to provide the interested reader and lecturer with additional directions in which they could take their students if they wanted to, to further explore particular topics in the book. So
27:52
Andrew Delamater: so this is not a comprehensive evaluation of conditioning and learning principles, but it's rather a presentation of the essentials, with some directions to where to go next. Andrew Delamater: So those were. Those are the 2 main goals now, like Mike. I've been teaching this class for many years. I started off, actually using Mike's earlier textbook on on this topic and it and it it
28:18
Andrew Delamater: created a good framework for me to think about how to teach. But I've encountered many challenges like Mike over the years, and how to teach this class effectively to students and and what like Mike, one of the one of the main challenges I faced is is how to convince students that this, that studying how non-humans learn relates at all to the human condition, because
28:40
Andrew Delamater: people tend not to see that link so immediately. So that's 1 challenge that I'll talk some more about. Another challenge is Andrew Delamater: that I find to be actually quite
28:53
Andrew Delamater: important as as higher education higher educators.
29:00
Andrew Delamater: and that is how to teach students to appreciate the scientific process. I think this is very fundamentally important as as students develop a toolkit for developing critical thinking skills. So in particular, what I aim to do is I aim to teach people how to think analytically and how to think quantitatively.
29:20
Andrew Delamater: and if they can do that through the vehicle of learning about conditioning and learning principles, I think I've done a great job in teaching them and bringing them along because they could use those critical thinking skills in all sorts of domains. So that's 1 of the things one of the challenges that I continue to stress and try to address in my teaching. I also
29:42
Andrew Delamater: I also love this stuff. So I like to stimulate at the end of the day some small subset of my students to themselves also be interested in pursuing some of the intellectual traditions that have developed over the field, and and I'll depart a little bit from Mike in saying that it's not all about behaviorism, but it's also about associationism, too, and comparative cognition has has a role to play in in animal learning and conditioning as well. But
30:12
Andrew Delamater: but we try to. We try to cover many of these different traditions within our within our textbook as well.
30:19
Andrew Delamater: So, okay, so how do I try to address some of these challenges? Well, okay, so first, st we're dealing with a question of relating learning processes that we study in other and other animals to humans. One simple question that I give is is that the basic learning mechanisms that we're studying are really involved in a whole lot of different psychological processes. Andrew Delamater: And I like to give a lot of examples that emphasize examples with everyday life.
30:48
Andrew Delamater: So I point out that learning processes can influence perception, our perceptual experiences, our emotional reactions to events. It can influence how we reason about cause and effect in the world, and there are also a variety of interesting clinical applications that we could talk about. So just to give one sort of example that I think a lot of people can relate to is that perceptions can be strongly influenced
31:13
Andrew Delamater: by your expectations. So if you go into into a situation thinking that what you're about to consume is is very sweet or very dry, or very fruity. It can shape the way in which you actually perceive that event. Now Andrew Delamater: there was a recent paper published one of many types of papers that were published on this topic, called The Sonic Seasoning Effect, and what they, the investigators did in this experiment is, they asked people to rate
31:46
Andrew Delamater: along sweetness and bitterness, dimensions chocolate that they gave them to consume. But under 2 conditions they were either listening to a background purely instrumental. Soundtrack that had independently be rate, was rated as being very bitter in its quality, or a different soundtrack that was independently rated as being very sweet in its qualities. And so just listening to these auditory soundtracks
32:13
Andrew Delamater: influence very dramatically how people rated the actual chocolate. And so so it kind of shows an interesting example of how a learning process can interfere or interact with an underlying perceptual process in a in a top down kind of manner.
32:34
Andrew Delamater: so I thought that was kind of an intriguing sort of finding in the literature, but also to highlight how emotional reactions can be can come under the control of learning processes, you know. Imagine war veterans, people who have
32:50
Andrew Delamater: have been unfortunate to be in a position where there's extremely traumatic events that they experience during wartime. Well, when those when those people come back to, let's say the civilized back to the home country in a more non-war type situation. They might encounter a stimulus Andrew Delamater: such as 4th of July fireworks, and react to those stimuli in very different ways from the ways in which people who have not had those prior wartime experiences would react.
33:20
Andrew Delamater: Okay. Many of us Andrew Delamater: see those fireworks as being very pleasant and interesting, but for somebody who's gone through the trauma of wartime. They, I imagine they would react very differently. So so here's another situation where? Where? And people might be totally unaware of what's going on here. But they just feel differently. So emotional reactivity is also something very strongly impacted by learning processes.
33:46
Andrew Delamater: And then, to take my favorite example of the way I teach causal reasoning or simple forms of Pavlovian conditioning is to think of it as as a reasoning kind of process where people are trying to uncover cause, effect relationships between events in the world. And so I invite people to think about what somebody might learn when they encounter the consumption of strawberries for the 1st time in their lives, and then they break out into a rash.
34:14
Andrew Delamater: Well, they're forming an association between these 2 events. Strawberries predicts rash, or is associated with the aversive consequences of a rash developing. And so, if people learn this connection, this association very readily, they kind of developed a cause, effect, understanding. But then, subsequently, if if the person now consumes those same strawberries that previously produced the rash along with a new food like oysters.
34:41
Andrew Delamater: and they lo and behold, end up developing this rash. The question of interest here is whether people think there's a causal attribution to be made to the oysters like, does? Is it the oysters that cause the rash? Or is it the strawberry? And most people would agree that. Oh, yeah, it looks like it's the strawberries. It's not the oysters, because I've already learned that strawberries produce the rash.
35:01
Andrew Delamater: So this, of course, is a classic implementation of a phenomenon. A famous phenomenon known as the Cayman Blocking Effect. And it's a much easier to grasp way of Andrew Delamater: of teaching this very fundamentally important learning phenomenon. Most people can see immediately that. Oh, yeah, the the causal mechanism here is is related to this blocking effect. So I think a strong case can be made that that causal reasoning in humans has an important learning component to it.
35:39
Andrew Delamater: Now. Andrew Delamater: So so that's how I try to address these problems of relating the non-human animal stuff to the human stuff. I also talk about evolutionary theory as another vehicle for getting into describing potential similarities and differences among animals of different types.
36:00
Andrew Delamater: Now, in terms of developing these critical thinking skills. This is probably the most difficult part of difficult challenge that I face in teaching, but I think it's probably one of the most important ones to try and grapple with. And here I want students to come away thinking more analytically than they have previously, and also to be able to think more quantitatively. And so how do you do that? And the way I try to approach this problem is, I want to emphasize
36:30
Andrew Delamater: the importance of doing experimental research. So I want students to understand the logic of the experiment, to see how we could start with a hypothesis and then run some sort of experiment and talk about the data, and then show how the data relates to those hypotheses that we're grappling with. This is a very analytical procedure that students have a lot of difficulty understanding. But one thing that's the beauty of learning
37:00
Andrew Delamater: studies. Learning research is that it's very fundamentally steeped in this
37:05
Andrew Delamater: important component. And so, just to choose one example, I'll discuss something that Mike also touched upon, which is the known in the literature as the learning performance distinction. So how many of us have ever given an exam. And students come up to us afterwards? And they say, Professor Professor, I knew that material. But I just do poorly on tests. I don't do well on tests, but the knowledge is in there.
37:32
Andrew Delamater: And so this is a classic sort of circumstance in which there is learning in that brain, but expressing it into performance sometimes is very difficult to do. Andrew Delamater: And so how do learning theorists or learning psychologists grapple with that important issue, that learning is a reflection or sorry that performance is a reflection of an underlying learning process which may or may not be expressed in behavior.
37:57
Andrew Delamater: Well, consider a classic sort of learning psychology, example, and this example is known as auto shaping, where the bird, if the bird experiences a relationship between a visual cue presented on the screen and food to follow it. What the bird ends up doing is they automatically start pressing, pecking at that key light. That's the predictor of the food.
38:21
Andrew Delamater: So we know that that's happened. The actual Kepec response is the readout that learning has occurred. But we can ask the question, What happens if we take that bird? And instead of presenting a visual cue, paired with food, we present an auditory signal paired with food. Well, the answer, there is nothing the animal doesn't do anything that's detectable in its overt behavior.
38:41
Andrew Delamater: But so we can ask, did that bird learn anything about the toned food relationship. And it's not expressing it into performance or or is it just not learning anything.
38:54
Andrew Delamater: maybe can't learn about tone food relationships? Well, the answer to this question is revealed when one considers an experiment in which you train the bird to associate tone with food, and then what you do is you pair a light, a localized key light with the tone, not with food, but with the tone itself. And what happens is that the bird starts pecking at that key light.
39:17
Andrew Delamater: So what this shows is that the inference here is that the animal must have learned something about tone and its relationship to food. It must have acquired some knowledge, because, after all, that knowledge would be required to support the second order. Learning of the key, peck response to that key leg.
39:36
Andrew Delamater: So another example of where, if you walk methodically, the student through an interesting experimental result with an interesting theoretical question, then hopefully, the students can understand the logic of how the scientific process works.
39:54
Andrew Delamater: and one interesting sort of conclusion from all this is that you really need specialized tests to reveal that learning process, because fundamentally, if what we mean by learning is something that's occurring at a mechanistic level somewhere in the brain, it may or may not have a direct readout. So there's a separation between the actual over performance or readout of the behavior of the learning process and the learning process itself.
40:22
Andrew Delamater: This, this is something that experimental psychologists have spent many, many years trying to finesse. And this discipline is no different from that.
40:30
Andrew Delamater: Another very interesting topic which I'll just touch upon because I won't get into. But it's the question of is the behavior goal directed like Mike's allusion to William James's comment. Sometimes behavior is not goal directed. It turns out that learning psychologists have a very good way of separating between a behavior that is governed by a goal oriented process, or an active representation of what might happen in the future
41:00
Andrew Delamater: versus when behavior is not at all governed by by these expectations or goal Andrew Delamater: representations. And so it's interesting to incorporate that into one's lectures, I believe as well, and we do that we touch upon this in the book, too.
41:20
Andrew Delamater: Now.
41:24
Andrew Delamater: that 3rd process, how do we? Or the 3rd challenge, how do we teach? How do we really get students interested in the intellectual traditions that define the field? Another way that I try to approach that question is to talk about some applications of learning principles, and there are a variety of interesting applications that one could point to, and some of these are practical applications, and others are theoretical applications, and so in terms of
41:52
Andrew Delamater: in terms of practical applications, I often weave into my lectures discussions of how substance abuse models are actually, some of them come from learning theory, type models. We also talk about anxiety disorders, and how there are practical applications and considerations based on learning principles. And I also touch to some degree upon learning theories applied to neuroscience. So I'll just touch upon Andrew Delamater: 2 examples here? In in closing. So so when talking about
42:28
Andrew Delamater: potential anxiety disorders, it's interesting to consider phobias. And and some people have intense phobic reactions to spiders.
42:38
Andrew Delamater: And just just the sight of looking at a spider on a hand is very nerve wracking. But to overcome anxiety responses what therapists have one approach to this that therapists have adopted is known as systematic desensitization and exposure therapy. Well, what are these things? Well, these things are very intimately related to something that in the learning lab we call extinction
43:01
Andrew Delamater: learning. And so what we try to do is we try to understand how the mechanisms of extinction works. That is, we want to extinguish the aversive reactions that people might Andrew Delamater: display when encountering a stimulus like this, and one of the interesting findings and discoveries important discoveries that has come out of this literature largely promoted or due to the work and ideas of Mark Bowton is that learning theories
43:33
Andrew Delamater: have shown us that extinction, learning itself is a highly context specific sort of process. So although the therapist, for example, might be able to help somebody overcome their phobic reaction in the therapist's office Andrew Delamater: generalizing that outside of the office becomes very difficult because the extinction learning is context dependent.
43:57
Andrew Delamater: So that's 1 interesting avenue for discussing applications and further interest generating further interest in the field. Another one that I like to think about is is this notion about the fundamental question of when does learning occur and and learning.
44:13
Andrew Delamater: Researchers have pointed out to a very interesting and important discovery, which is that learning best is accomplished when there is a prediction error, that is, when there is a failure to predict the event that's to be learned about.
44:27
Andrew Delamater: Okay, that's a reward prediction error, and that has given rise to a very fundamentally important concept, which is that unexpected rewards are more fully processed by the machinery than expected rewards. And so there are plenty of examples of Andrew Delamater: trying to understand, at the neuroscience level of analysis, how this works. Andrew Delamater: But just to choose what 2 examples.
44:54
Andrew Delamater: Joshua Johansson published an interesting paper back in 2010, where he was recording electrophysiologically from neurons in the amygdala, and he presented foot shocks to rats while he's doing these recordings. And what he found was that if those foot shocks occurred unexpectedly, that is, they were not signaled.
45:16
Andrew Delamater: Then the electrophysiological response was quite substantial to those events. Those neurons processed the shock very effectively in the amygdala when the shocks occurred unexpectedly, but when they occurred signal, that is, when they occurred right after a cue that predicts that shock. Then notice that that response, the electrophysiological response of those individual neurons was very much blunted.
45:40
Andrew Delamater: So here's a very clear case where where the processing devoted in single neurons devoted to unexpected versus expected shocks is very different, and of course a lot of investigators have also been interested in determining whether the same is true in the human brain. And there's fmri data supporting this very notion as well that certain regions ventral striatum, the midbrain areas care very much about whether an event
46:10
Andrew Delamater: and expected reward is predicted or unpredicted. So there's a lot of interesting ways to tie a lot of research together between animal learning, research, and research from other domains, as Mike had pointed out, and I hope I have pointed out as well. But these are some of the strategies that I've used. There are plenty of other examples, and the discipline is ripe for additional discoveries. Andrew Delamater: So with that I will
46:37
Andrew Delamater: stop my sharing, and I think we have some time to entertain any kinds of questions from folks. Andrea Burgoa: We do. We do have some questions. Thank you to you both for your presentations, and thank you to everyone for your great questions and your conversation in the chat.
46:57
Andrea Burgoa: I'd like to start Andrea Burgoa: with one question here from Victor. Victor says, Thanks, Michael and Andrew, for the new book. I always struggle the most when teaching the more technical portions of any book on fundamentals of conditioning. Andrea Burgoa: I see that you've kept mechanisms and theories around. So do you have any tried and tested tips to teach those chapters while keeping the students engaged.
47:23
Andrew Delamater: Great question, Mike. You want to handle that one.
47:27
Michael Domjan: I don't. I don't have any novel tricks well, it you know the the
47:47
Michael Domjan: encouraging students to come up with personal examples that can that that strategy can. And I use that just as much when I'm talking about theories as I do with basic phenomenon. Michael Domjan: So, for example, the response deprivation theory of reinforcement
48:07
Michael Domjan: students can come up with personal examples that illustrate that in their own experience, talking about the Roscolo-wagner model and the over prediction effect. You can find personal examples that match that over prediction effect. So the general Michael Domjan: strategy of of sending students out there to figure this out Michael Domjan: is applicable to the theory part just as much as the basic phenomena.
48:42
Michael Domjan: And and I think it does make make certainly predictions of theories more easy to for students, at least it motivates them to grapple with the issue more seriously than they would if they just read it in the book, you know, read it in a paragraph.
49:04
Andrea Burgoa: Great. Thank you, Andrew. Anything to add to that.
49:08
Andrew Delamater: Yeah, I can add just a little bit. One thing I found really doesn't work is to focus too much on on the original experimental situation, that is, you know, with the rat in the box, or the bird in the box. And and so, if you could, if you could find a hook with the students, with a real world sort of everyday example that people can relate to. That's a great place to start. So, for example, the strawberries, rash example that I gave.
49:35
Andrew Delamater: and then the strawberry oyster rash example. I very frequently make reference to that, and once they understand that basic effect, and they could relate to it. Then you could say, How do you think this works? Andrew Delamater: I mean, let's try to understand this theoretically. Andrew Delamater: and that's where you know you get into things like processing interfere processing problems. The Riscorda Wagner discussion can come out of that. But but it's easier to.
50:04
Andrew Delamater: I found that it's easier to think about theoretical concepts only when people have a firm understanding of the phenomenon of interest, and if you could describe that phenomenon with simple language and simple, everyday examples. Then then that goes a long way.
50:23
Andrea Burgoa: Thank you. Andrea Burgoa: Alright. Our next question. Andrea Burgoa: It's about artificial intelligence. Andrea Burgoa: I'm interested in your thoughts on reinforced learning as used in AI. I saw this morning that this topic received this year's Turing award. Do either of you have any thoughts on reinforced learning as used in AI.
50:47
Andrew Delamater: Yeah, I guess I'll start off with his with this one. Mike. Andrew Delamater: one thing I don't do in this class is, is I? I don't go beyond are Andrew Delamater: theoretical notions based on the riscordel-wagner thinking, which is, which is a 2 layer neural network. Andrew Delamater: But one thing I do in my research is, I have multiple layers to the networks that I that I use to simulate behavior.
51:16
Andrew Delamater: And so an AI network is really just a multi-layered connectionist network Andrew Delamater: and reinforcement learning is one version of that. Andrew Delamater: So so for the really interested student, I'll bring that, you know, during office hours I'll get into this a little bit more with them. But yeah, absolutely. I do think that these neural net models are
51:38
Andrew Delamater: are very much intimately related to AI. I mean Jeffrey Hinton was one of the grandfathers of this whole movement, and he came right out of the Connectionist movement in the 19 eighties with David Rummelhart and Jay Mcclelland. So it's all based on these simple networks. But it goes Andrew Delamater: beyond rescortle Wagner, because Rescorda Wagner was simply 2 layers
52:00
Andrew Delamater: deep. Deep. AI networks are multiple layers. I see them as categorically, very, very related to one another. And indeed, I think it's it's a great question to ask about the power of these models Andrew Delamater: and and sort, yeah, yes, absolutely. Sutton and Bardo were right up there as well.
52:23
Andrea Burgoa: My goal here to add any thoughts. Would you like me to move on to the next question. Michael Domjan: I don't have any anything useful to add, except that on a number of occasions I've had computer science undergraduates or computer science graduate students who chose to take this class, and they were fascinated. They loved it. And I know some of the graduate students went on to do projects in computer science that
52:53
Michael Domjan: imported some of the ideas from the basic learning paradigms into issues that they were working on in computer science. So there's definitely there's a huge connection. Andrea Burgoa: Yeah. Michael Domjan: Okay. Michael Domjan: I might not notice that using artificial intelligence in a different way, less mechanistic way. But everybody's talking about how you have to integrate AI in your in, in your teaching. So I've started using Chat Gpt.
53:31
Michael Domjan: and and the way that I use Chat Gpt, I ask it questions about conditioning and learning, you know. Like I asked Chat Gpt, what is the Thorndyke's law of effect? And I see what the response is Michael Domjan: in. Invariably the response that chat, Gpt generates, has some mistakes in it. Michael Domjan: And so that's a beautiful exercise to bring to class and talk about
53:58
Michael Domjan: common misconceptions, because Chat Gpt is gonna report things as true which are common misconceptions. Andrea Burgoa: Okay. Michael Domjan: Also, I think, tempers students enthusiasm for immediately going to an AI program to try to do their homework. Michael Domjan: I cautioned them as a if you do that, you could just as well come up with the wrong answer.
54:28
Andrea Burgoa: Next question for Matt. Given you refer to animal research in the book. How do you address concerns about animal welfare and research?
54:40
Andrea Burgoa: That's not a tough one. Andrew Delamater: I always talk about Iacuc Iacuc committees as as playing a fundamentally important role in in this whole scientific enterprise with non-human animals.
54:54
Michael Domjan: Oh,
54:57
Michael Domjan: I have a lot of a lot of things to say about that. But I have pulled back from spending too much class time on that. Instead of you know, I pulled back on Michael Domjan: how much animal work I present, and so corresponding. I mean, I try to talk about human situations, and you know, so it and students haven't brought it up. Michael Domjan: But certainly animal welfare and animal rights is a huge issue, and I'm happy to talk about it. And and I think I
55:44
Michael Domjan: but it it doesn't come. Thank goodness, it doesn't come up a lot. In my personal teaching.
55:51
Michael Domjan: We we mentioned those issues in Chapter one, in the 1st chapter of the the book, to some extent.
56:00
Andrew Delamater: I see, Mike there. There's a question in the chat about whether whether we devote coverage to Tolman, Spence and Hull, and I think the answer is yes, we do.
56:15
Andrew Delamater: Given the scope of the of the book. We don't delve deeply into many of those topics, but in one section, having to do with the structure of instrumental learning. We do distinguish between stimulus, response, learning, response, outcome, learning, which was more of a Ptolemanian sort of concept and hierarchical stimulus, response, outcome, learning. So we touch upon
56:40
Andrew Delamater: upon how modern thinking has evolved from the earlier ideas that that the fathers of the field have have really laid out. Andrew Delamater: I mean, that's 1 place. Michael Domjan: But I must admit we don't talk about those individuals and in much detail. I mean, we. We reflect their intellectual contributions, but we don't specifically discuss them
57:10
Michael Domjan: in detail. After all, this is just the essentials, and one of the one of the advantages of the essentials is that it gives provides freedom for instructors to fill in details that they are more interested in, or they think is more important.
57:35
Andrea Burgoa: We have a question here.
57:40
Andrea Burgoa: Could human emotions and emotional emotional regulation interfere in conditioning.
57:50
Andrew Delamater: Yeah, you know, we don't go into that very much in the book, but but it's an interesting. It's a very interesting question. And and 1 1 area of research that I'm aware of that has addressed that question is is whether stressed out individuals Andrew Delamater: can learn specific about specific consequences of, let's say their actions
58:14
Andrew Delamater: and and and I think the verdict is still out there in terms of the literature. There are some suggestions that stress might suppress learning Andrew Delamater: about cognitive events, for example. But there are other studies showing that not always. It's still possible to to process cognitively while while you're stressed. And my suspicion is that the Yerkes-dodson law is going to play a role here in all likelihood. But no, it's a it's a fascinating topic. I think.
58:47
Andrea Burgoa: So christy mentions that she is adding some information about gamification to her class, and she's looking for Andrea Burgoa: your thoughts on. This is a good real world application of operant conditioning. And then she went on to define gamification for those of us who may not be in the know, applying elements of game, playing to online marketing or motivational apps, earning points, competition, rules of play, etc. Michael Domjan: And online games.
59:18
Michael Domjan: It's all about schedules of reinforcement. It's all about operant behavior. And it's all about keeping kids tuned in. That is the goal of the game designer is for the kid never to put the game down. Michael Domjan: So I mean, and that, interestingly enough, that's also the goal of Facebook and Amazon and all the online. You know, major vendors, if you will. They don't. You know. They don't want you ever to leave? And and what
59:59
Michael Domjan: traps kids into the game is the particular schedule of reinforcement that's used so. Andrew Delamater: And gambling casinos. Mike, don't forget gambling casinos. I was recently in the Bahamas, and I spent some time in the gambling Casino, and you could see it everywhere, you
1:00:17
Andrew Delamater: everywhere, behavior, principles at work. Michael Domjan: It's another environment where the last thing they want you to do is to leave. Michael Domjan: I want you to stay there and keep working it.
1:00:32
Andrea Burgoa: All right. I think we have reached our end. I'm sorry if we didn't get to any other questions, but thank you all for your participation and your insightful questions and your interest in our presentation. Today, we're happy to have you here and thank you. Thank you so much to our wonderful authors. Thank you, Andrew and Michael. We appreciate you so much, and we appreciate. Appreciate all of your wonderful insights.
1:00:57
Andrea Burgoa: Please join us for another webinar again. We hope to see you soon. Andrew Delamater: And everyone have a great day. Thank you, Andrew. Thank you. Andrew Delamater: Everybody. Andrew Delamater: Bye. Andrea Burgoa: And recordings and resources, and all of that will be sent to you. Just be patient with us. Okay, thank you so much. Bye, bye. Michael Domjan: Thank you. Bye-bye.
Here to help you learn
Sign in to keep a private conversation with your course tutor.
Ask me anything about the material…
Sign in to chat