Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
#5 - Alan Sanfey - How we make decisions and the science of it
14 Plays5 years ago
We most of the time don’t make rational decisions. It feels like feel like we are in control but that’s just an illusion. In this conversation we talk to Prof. Alan Sanfey (alansanfey.com) who is the principal investigator of the decision neuroscience lab (www.decisionneurosciencelab.com/) at the Donders Institute for Brain, Cognition and Behavior. We get a better idea of what drives our decisions, how industry is trying to use these concepts and also where it sometimes goes wrong.
Recommended
Transcript
Introduction to Podcast and Guests
00:00:00
Speaker
Hey, what's up, everyone? This is Ajmar Savary, and welcome to another podcast episode. Today, we talk to Alan Sanfi about how we make decisions and the brain mechanisms involved. We also discuss how this knowledge could be used commercially. Enjoy.
00:00:30
Speaker
Hey everyone, and welcome to another podcast episode. If you're new here, my name's Ajmal. I'm a neuroscientist and entrepreneur. On this podcast, we explore the links between science, technology, business, and the impact they have on all of us. Our guest today is Dr. Alan Sanfi. Alan is a principal investigator at the Donner's Institute for Brain Cognition and Behavior.
00:00:52
Speaker
Previously, he has held positions as Associate Professor of Psychology at the University of Arizona and as a Postdoctoral Research Fellow at Princeton University.
The Science and Application of Decision-Making
00:01:01
Speaker
He currently heads the Decision Neuroscience Group at the Donner's Institute and his research uses a novel approach to the study of both individual and interactive decision-making by combining the methods of behavioral experiments, functional neuroimaging and formal economic models.
00:01:18
Speaker
A further goal of his group is to use the knowledge gleaned from these studies to inform public policy debates. In this conversation, we talk about the science of decision-making and how it could be used in industry. We discuss various aspects of decision-making theory and how industry has tried to incorporate them and where things sometimes go wrong. You can find Alan on his personal website at www.allonsanfi.com. Enough background. Let's get into it, shall we?
00:01:49
Speaker
Alan, thanks so much for taking the time. Um, it's actually a beautiful day outside and for the past few days, been raining quite a lot. So it's again, sad that we sit inside, but I hope the listeners can appreciate the audio quality at least. Yes. I can imagine the sun shining down. Yeah. Well, they can sit outside listening to this. I hope that will be a good day too. Well, actually I'm certain it will be now.
00:02:14
Speaker
We have actually met each other. I don't actually think you remember. We met and I don't blame you for not remembering plenty of students in that class, but it was 2010.
00:02:28
Speaker
And you gave a guest lecture in, I think, what the course was, social neurocognition. Yes. Is that possible? Still giving that. You're still giving that. The guest lecture has changed. Oh, right. The students change and the lecture changes. OK. That's still in my schedule every year. Yeah. But I remember that we were, I think, we were just four students or five from the certain theme of the studies.
00:02:50
Speaker
And it was taught by Ivan Tony. And yeah, you gave a guest lecture on decision-making. And everyone was super fascinated. And I still remember it so well because you explained findings from the trust game.
00:03:05
Speaker
which we'll talk about in a little bit, hopefully. So listeners, you then understand what we're talking about. But in essence, you have to play with a game partner. And you propose the scenario. You put a picture up on the screen, and you ask, how much would you give or invest in this person? And it was like a number between 0 and 10.
00:03:27
Speaker
And everybody said something like seven or nine. And I was the only one that said zero. And you were very upset about that because it was a friend of yours that was on the screen, which of course I didn't know. But, well, that definitely made it very memorable to me.
00:03:43
Speaker
Perhaps it said more about your trust than my friend's face. It definitely does. But there's always one or two who are conservative in their investments. Yeah, well, I think conservative is still very nicely put. Because I don't think ever anyone invested zero on the first go. Well, in these games, typically that does happen. And probably 10% of people are
00:04:09
Speaker
would like to keep what they have and not engage in risky behavior with trusting other people, which can't go badly wrong, of course. Oh, yeah. I mean, I can see it in the industry all the time. When it comes to collaboration and there's no trust, it just goes very, very wrong. Yes. And even if the economic incentives are aligned. Yeah. And that's interesting how trust can be then separated from risk. So people obviously, when you're trusting someone,
00:04:37
Speaker
There's some element of risk assessment, but there's I think an extra component of this feeling of is this person or entity or organization trustworthy. One interesting thing I think is how quickly that can go. If trust falls apart, it takes a long time to repair that again. I think it's an interesting phenomenon to study in the context of decision-making.
Alan's Research Journey and Inspirations
00:05:01
Speaker
That was nine years ago. Now, nine years ago, you came to the Dawners Institute, if I remember correctly, right? Yes, that's correct. And basically, you're a professor in decision making of neuroscience. Now, my question to you is so that the listeners can understand a bit more because they already heard the bio in the intro, but what made decision making so interesting to you?
00:05:31
Speaker
Well, for me, it's always been the most fascinating aspect of psychology. I think it's in some way the essence of what makes us human, that we have to make decisions. We have to decide our courses of action. And these can be very trivial decisions as to what we have for lunch or more important decisions, like do we engage in a podcast? Are really serious decisions. Do we buy a house? Do we get married? What job do we take?
00:05:58
Speaker
And of all the different facets of psychology, this was the one that kind of resonated with me immediately once I started studying psychology way back as an undergraduate, because this seemed to encompass all of those other
00:06:13
Speaker
aspects, memory, cognition, language, perception, all of those seem to really boil down to be distilled to how we make decisions and choices. One thing that always struck me was that we didn't really know that much about it. There was a lot of research in different fields, for example, economics,
00:06:32
Speaker
some in neuroscience, but the psychology of decision-making was rather limited and limited to a particular type of studies, which were these studies by Kahneman and Tversky. People probably know Danny Kahneman as a Nobel Prize winner and an author of Thinking Fast and Slow, which is one of the most successful popular science books. And Danny Kahneman with Amos Tversky led a very influential research program
00:06:59
Speaker
which tried to understand decision making, but which tried to understand it from the perspective of when it goes wrong. And so they had wonderful, vivid examples of people making kind of quote, irrational decisions, which are tremendously entertaining. And it's really nice to see these examples, which are a little bit equivalent to looking at perceptual illusions when you're trying to study vision. So you look at where the system kind of breaks down.
00:07:28
Speaker
I mean, it tells you something about the system, right? It does indeed. And of course, it's very informative, but I always felt it was a little bit limited in that there's a wide variety of cases where there isn't a rational or irrational response. There isn't an optimal solution. And most of the time we actually do pretty well with our decisions. We successfully navigate the world to the most part. We've all
00:07:51
Speaker
arrived here thanks to careful decision making by our generations of ancestors. So we must be doing something right. So I became interested in studying decision making, but not just from the kind of where it fails part. And so could we kind of better understand how we make decisions without necessarily the judgment of whether they're good or bad, but just understand the process by which we arrive at choices in everyday life.
00:08:16
Speaker
And that has then led you from the like more studies that were more focused on figuring out when does it break down more into the neuroscience aspects of it.
00:08:27
Speaker
Yeah, that was the trajectory. Initially, my PhD was in cognitive psychology and there we were inspired by many of these Kahneman-Tversky studies to look at how memory plays a role in decision-making and how memory sometimes helps and sometimes hinders our decision-making, depending on the accuracy of our memories.
00:08:49
Speaker
This was back in the late 90s and so the study of neuroscience was really starting to come into psychology then. So the availability of things like fMRI brain imaging started to be available to psychologists and that just seemed to me really interesting development because one
00:09:09
Speaker
One downside of the psychological models is that there were largely a black box where the brain was, and it's like some magic happens here and the decision is output. But it wasn't very clear what kind of processes actually went on to arrive at the decision. Neuroscience and fMRI in particular seem to offer a
00:09:29
Speaker
a way of constraining our models by the biology. And so I had an opportunity to do some fMRI studies as a final year PhD student. And that really, I think, set me on a path of trying to integrate the psychological and neuroscientific aspects to build these models. Right. I mean, you had one
00:09:52
Speaker
in the field, you had one very super influential paper very early on. Okay, I would say early on, if I look back, can you tell us a bit about that study? Because it was, yeah, I thought it was pretty awesome. Yeah, it turned out to be quite influential. We didn't think that at the time. And it largely came about, I think, one of the
00:10:13
Speaker
very positive aspects of academia is occasionally people are freed up just to think and do stuff, which may seem to be a little bit crazy or a little bit not immediately relevant. And I was lucky enough in my postdoctoral fellowship.
00:10:28
Speaker
At Princeton in the US, I had a situation where I already had a faculty position, which I deferred. I had a two-year postdoc, which was risk-free. I could do whatever I wanted. I wasn't worried about a job marketer strategically positioning myself to get a job. I could just do what I wanted.
00:10:47
Speaker
It doesn't happen often. That doesn't happen often. It was a very nice circumstance. And an ally to that was I was in a lab where I was a little bit of a peripheral member. So the lab was really interested in a process called cognitive control or how we exert control about our cognition. But the PI of the lab had a kind of a side interest in decision making. And he just thought it'd be fun to have a decision making researcher there. So I got that position and I happened to be in an office with an anthropologist.
00:11:15
Speaker
was also interested in these kind of broader questions. And our projects really came out of kind of spitballing over beers and come up with ideas that we thought, well, we have the opportunity. We were literally beside the scanner. Nobody was
Exploring Fairness in Decision-Making
00:11:30
Speaker
using the scanner. And so we could just go in and play around. So it was really, yeah, it was really playtime. I mean, when I read sometimes about
00:11:38
Speaker
some of these tech companies who let their employees devote certain percentage of their time to play. It reminds me of that time, except I had 100% of my time. We got interested in the notion of how people make decisions when they're interacting with other people.
00:11:58
Speaker
concepts like cooperation and coordination and competition. So how do decisions change when other people are involved? And one thing I've always been very interested in is the concept of inequality and unfairness that always seem to be a
00:12:13
Speaker
a remarkably strong motivating force in people's behavior. If people feel they're being treated unfairly, they're often willing to go to great lengths to try and address that or punish the people who are treating them badly, often way beyond what the actual resource they're being treated unfairly about. You see it in children. If children feel their brother or sister gets more or less than them, you can really see how strong and aversive a feeling it is. Very visceral.
00:12:42
Speaker
It is. Yeah. And so we were interested in, could we study that in the lab? Could we make people feel unfair, unfairly treated? And what would that do in the brain? Could we make sense of that neurally? And we presented some ideas at one of our lab meetings and Danny Kahneman. It was nice to meet one of your heroes. He happened to be sitting in on the lab meeting. Wow.
00:13:05
Speaker
And one day he just said, you know, you should look at this game called the ultimatum game that really gets at this kind of thing. We had never heard of the ultimatum game, which turned out to be a game proposed in experimental economics, which wasn't super well known at the time. But we went off and investigated and it turned out it was a
00:13:23
Speaker
quite nice task to enable us to put people into these situations where they felt unfairly treated. In short, the ultimatum game is a task where you and your partner, who you don't know, so another person, are asked to share $10 between you. Your partner can make any proposal as to how this money can be shared.
00:13:44
Speaker
And then you have to decide whether this is a reasonable thing to do or not. So we were interested in situations where the partner offered you a very low amount. So he or she said, I'm going to keep $9 and I'm going to give you $1. So you're faced with this situation now where this very unfair division has been proposed and you can decide to either accept this offer, you get $1, they get $9 and the game is over, or you can reject the offer in which case nobody gets anything.
00:14:14
Speaker
to every model of economics that exists, this is a trivial choice. It's not even a choice, right? Because if you accept, you get $1. If you reject, you get $0. So you would always accept. But as it turns out, and as we've seen now thousands and thousands of times, people don't behave this way. So when people are confronted with this choice, more than half the time they say, no, I don't want any part of this, right? I'd rather have nothing and you get nothing than we have this unfair division.
00:14:44
Speaker
And so I think that shows the profound effect of this feeling of unfairness. So we were interested in studying this. We did many experiments. And in particular, we looked at what happens in the brain when people experience this moment of unfairness. And we found a nice network of brain areas that we've since replicated many, many times that started to give us some kind of toehold into how the brain encodes.
00:15:10
Speaker
complex, I would say, moral emotions, such as kind of unfairness, injustice, and so on. I think we were lucky when that paper came out. It happened at just the right time. People were really interested in neuroscience and fMRI. People were really interested in decision making. We managed to immerse these two, and the papers turned out to be at least well-sighted over the years. Right.
00:15:34
Speaker
And it took me off in a whole research direction, which I certainly hadn't anticipated when I showed up at Princeton. But when I left, I had a whole research path looking at these social influences, so going beyond unfairness to other social motivations. But I guess that's what kickstarted and
00:15:51
Speaker
Yeah, it was largely happenstance. I happened to be in the right place at the right time and was able to take advantage of that. So that was a good start. And you were allowed to play. And that's turned out to be extremely important. Other friends who went off to postdocs who didn't have the resources I had or the freedom, or they didn't allow themselves the freedom, I think, sometimes to just study what's inherently interesting to you because that's, of course,
00:16:19
Speaker
It's hard to beat the motivation of studying something you're just really interested in. Right. And just to follow up question, when people thought they play against some type of machine, they would accept low offers.
00:16:35
Speaker
Yeah, so that turned out to be interesting. So initially, we also had people play against the machine because we kind of had the idea that what's driving this feeling is that somebody else is treating you badly. Not that I'm being treated unfairly per se, but somebody is kind of trying to screw me here. And we thought that was the real reason why people got so upset.
00:16:57
Speaker
And so almost as a test, we had people also play against computers. So we told them a computer will randomly make you an offer. And that offer may be fair or unfair, and then you have to make a decision. Initially, to our surprise, quite a few people actually rejected offers from computers also.
00:17:14
Speaker
in that they were told, this box will generate a random number and that's what you get. And you can say yes or no. And a lot of people said no. What we realized afterwards was that this was a very kind of initial reaction. So if we told people, just wait five seconds and then tell us, then everybody said yes.
00:17:33
Speaker
Oh, right. And by doing kind of these experiments where we can vary the identity of the person or thing making the offer, we kind of start to see that there's a very initial and as you used, I think a really good word for visceral reaction to unfairness that if I'm being treated badly, I'm going to do something about it.
00:17:52
Speaker
Now, if I'm given seconds to think, okay, this computer clearly isn't targeting me for unfair treatment, I can do the, you know, quote, correct thing. But there's a certain sense of, people are very sensitive to this feeling of being unfairly treated and can often crowd out even the... Like being taken advantage of kind of.
Interdisciplinary Collaboration in Decision Neuroscience
00:18:15
Speaker
Yes, yes. And we've done many other experiments to find out what is it about these unfair offers that really
00:18:22
Speaker
makes people feel so, so mad. And it turns out to be, need that sense of being taken advantage of. So if you're told the other person only can not make you this offer, usually say, okay, that's fine. I don't think the other person is trying to hurt me. It's not intentional. Yeah. So the intentions turn out to be really important in this case. Now that has been
00:18:46
Speaker
When did that paper come out? That came out in 2003. 2003. Now that was kind of the starting, well, I shouldn't say starting point, but when decision neuroscience became bigger and bigger and like really as.
00:19:02
Speaker
a big part on the map of what's possible and areas that is going to be covered. Now, ever since then, I mean, the field has been grown so much. And of course, as you said before, it's super interdisciplinary. You have behavioral economists looking into this. You have also the, let's just call them, in air quotes, standard economists looking into this. Everyone, of course, with a different, slightly different framework of trying to understand
00:19:31
Speaker
Um, are these different groups ever talking to each other? Yes. I mean, one, one of the, one of the big differences in the past, you know, almost 20 years now is that there's actually a quite thriving academic society call a society for neuroeconomics, which is a, a word I don't particularly like. Um, I much prefer decision neuroscience, but the society is called in your society for neuroeconomics. And I'm currently the president of it. So I can't, I can't, I can't bath melt the term too. I should know that.
00:20:01
Speaker
And, you know, we have a meeting upcoming in Dublin, actually, in October, our annual meeting. And we expect about 350 people there. So that's a society that's really emerged. I think this is our 17th meeting. So it's a very new academic society.
00:20:19
Speaker
And we indeed draw economists, psychologists, neuroscientists, occasionally sociologists, anthropologists, quite a few people from business schools on marketing and management. So that's actually been really encouraging to see that people are willing to talk and more than just talk, actually actively collaborate and try and build these models that take advantage of the
00:20:42
Speaker
the various strengths of the component disciplines. It's still small in relation to a large field like neuroscience, but I think it's positive. It's growing every year. There's a sense that people are seeing the benefits of
00:20:58
Speaker
interdisciplinary collaboration more than just paying lip service to it, but I'm actually willing to do it. Yeah, as long as it actually happens, right? Yeah. And that's the trick. I mean, everybody says it's great and every grant says it should be interdisciplinary. And I think sometimes then we minimize the difficulties. It's quite difficult to do. I mean, you can spend an entire conference just getting the sense of what people are talking about. So I was recently at a workshop in the US on
00:21:27
Speaker
I'm kind of the theme of social inequity, social inequality, which I thought, well, I study this, I should know all this. And there were people from many different disciplines and just with completely different perspectives on these things, very interesting perspectives, but quite different.
00:21:45
Speaker
It took us literally a day, a day and a half to even arrive at the same page where we could have a useful conversation because people are talking from very different levels, very different approaches. So interdisciplinary-ness is hard, but worthwhile. And so when it actually occurs, I think you can see the benefits. Right. Now, wait, is this the same conference? That's in your economics conference. The one you mentioned at the beginning, right?
00:22:12
Speaker
Ah, okay. So then I have been to that one when it was in Miami. Okay. It's been quite some time ago. Yes. Yes. That was, I don't know, maybe six years ago, seven years ago. I'm not sure. Yeah. We've been in Miami a few times. Oh, okay. Well, it's nice there. Yes. And it's in this big hotel. We have this great view. Yeah. It's no great hardship to go to a conference in Miami. No, for sure not. But when you look at the field, what are some of the latest findings or A-finding, which do you think is
00:22:41
Speaker
very, very exciting. Some, one of the most recent ones, or well, doesn't have to be from the past year, can also be from, I don't know, past five or 10 years. But where you think, okay, this really is a big step forward. Yeah, I think it's more probably a constellation of findings and, you know, sort of my own subfield of looking at, say, social decisions. There's been a lot of progress in understanding valuation, I think. So understanding how we,
00:23:09
Speaker
how we encode the value of a particular choice prior that we have to make it, right? So I'm given a choice. I can go to Miami or I can go to Dublin. I have to kind of get some essence of what will the value of each of these options be before I decide what to do. And there's been, I think, really a lot of progress in really understanding the mechanisms of how that valuation process works.
00:23:31
Speaker
What goes into me coming up with a feeling of value for Miami versus Dublin, for example? What are the kind of psychological inputs? And then also what are the kind of neurochemicals that are involved in processing that? So how does dopamine help us in code that? So there's been a lot of work on that. And that's really been quite interdisciplinary. So work from rodents where we can really track these
00:23:57
Speaker
these chemicals and these brain areas all the way up to kind of work from economics from very theoretical way of thinking about it. And they're starting to map onto each other, I think, in really interesting ways. So that's one area that there's been a lot of progress within my own area of social decision making.
00:24:13
Speaker
There's been a lot of progress in modeling these interactions in a really meaningful way. So not just going beyond saying if people are unfairly treated, they'll be upset and they'll make a certain type of decision and that certain brain errors are involved, but actually building up models saying what are the key factors in these decisions? What is the brain networks involved?
00:24:37
Speaker
How might this differ as a function of, say, social context or emotional context? What are the individual differences that might play out? It's really building up much richer models of the decision process. I think that's where the
00:24:52
Speaker
the progress has been made, and I think that's the direction we'll potentially see most progress in. Right. But the modeling doesn't only hold for the social decision-making, right? No, no, no. And indeed, the modeling is probably late to the party social decision-making for the modeling. So the modeling really has started off with individual choice. So very simple choices, right? If I have a choice between an apple and an orange,
00:25:16
Speaker
seemingly almost overly simplistic, but we can start there and start building on models of how do we do that valuation up to more complex sets of choices. Now you mentioned specifically valuation that a lot of progress has been happening there.
00:25:34
Speaker
Now, when it comes to valuation, I think you put it very, very nicely as like, hey, do you want to go to Miami or... Sorry, what did you say? Dublin. Dublin. Well, I mean... Always go to Dublin. Always go to Dublin. If you're in Europe, go to Dublin. Anywhere I go to. Anywhere you go to Dublin. Unless you're in Dublin, then you should go to Miami. Right. That's, of course, that's always possible. But when it comes to valuation, right, I mean, you gave this example of, okay, Dublin versus Miami.
00:26:01
Speaker
Now, there's this talk, and people can, of course, look this up on YouTube, where Dan Ariely talks about decoys. And I thought that was, or I think most people in the industry know this maybe more like as nudging. And I thought this was very interesting, what that does to the valuation process. Can you talk a bit about that? I find that quite interesting.
00:26:26
Speaker
That's an interesting reason because I think we don't know what it's sold out. The idea behind a decoy is you put a slightly lesser appealing option in and it can actually change people preferences somewhat. So if I'm choosing between Miami and Dublin, I may be fairly indifferent. But if I give you a slightly worse European city,
00:26:46
Speaker
I'm not going to bad math a particular European saying, yes, somewhere that's not quite as nice as Dublin. Even though it's never really in the running for being chosen, it may push you towards Dublin because by comparison, it looks better. And of course, marketers are well aware of this trick, and I've been using this for many, many years.
00:27:03
Speaker
I think these are the areas actually that we still have to make progress in is understanding mechanistically how these decoys actually work. Why would that necessarily lead to a change in valuation for the ultimately preferred option? I think that's where we're starting to see
00:27:24
Speaker
interdisciplinary nature, hopefully having effect. So we know these things from psychology and from marketing and neuroscientists now are starting to take them seriously. So I think previously maybe they had taught them as they're kind of these little tricks that psychologists can play, right? And you get to shift some preferences a little bit, but maybe they aren't, they're, you know, parlor tricks basically, as opposed to
00:27:46
Speaker
really meaningful examples. I think they're more meaningful than just tricks. And it's starting, I think, to filter into the research where people are interested in understanding how this works. Not so much to understand how this works, but again, maybe to come back to our earlier example of a cognitive illusion, how this helps us understand the system better. So what are the factors, for example, that go into valuation? This suggests that some kind of comparison process is important too.
00:28:13
Speaker
So it's interesting you bring that up because I think that's the next phase now of where we want to understand these
00:28:23
Speaker
I guess aspects of decision making that marketers and lawyers and salespeople are well aware of from a practical perspective. I think it's quite interesting to try and understand that from a brain slash behavioral perspective with the view again of building up better models of the system. Yeah. Do you think it would be useful to look at it from the perspective or viewpoint of perception research? Because that's what I used to do.
Nudges and Decision-Making Biases
00:28:50
Speaker
And when you mentioned illusions,
00:28:53
Speaker
you just tweak a little bit of the sensory input you're providing, and the illusion is altered. And even though it's the same illusion, but it's not perceived the same way, and you just tweaked one tiny parameter a little bit, basically throwing off your reference, and now you have a new reference, which then, because of that, changes the illusion.
00:29:16
Speaker
I think that's a good approach. This whole program of heuristics and biases, the Kahneman diversity policy was really explicitly derived from perceptual research the exact same way. We look at the inputs and we look at, see where the
00:29:35
Speaker
Where the system is, you know tricked and what is that? What does that tell us? And so I think certainly perception provides a good fundamental basis for this and sometimes our decisions are you know The perception is the the reason sometimes it's memory. Sometimes it's attention So but I think that approach is a is a good one. Yeah, sure. Yeah. Yeah. No, I would
00:29:59
Speaker
Yeah, if I had to think about it, I think I would tackle it from a systems theory perspective. But we can talk about that later. We don't have to talk about it. It's just, it's my curiosity coming out. Sorry. Okay, but you mentioned that nudging and these, well, let's just say again in air quotes, lots of these tricks are being used so much in marketing or sales, advertising, and actually
00:30:26
Speaker
all over the place if you are aware of it, of these principles. Do you think there are some of these principles which are used a lot, I mean talked about a lot as in their usefulness in industry, but seem to you as if they are being misunderstood?
00:30:48
Speaker
Misunderstood in the sense of what? We can break it down as in one, concept and two, effectivity. Okay. I think from the first perspective, I wonder does anybody care, like in industry, whether they conceptually understand the decoy effect or not? Because I always only present if it works, it works.
00:31:08
Speaker
or you perceive that it works. Why it works is less interesting. That's kind of the province of science to understand. Effectiveness, I think that's a really good question. I have colleagues in marketing that I discussed this with at length.
00:31:26
Speaker
Personally, I don't know how effective these things are because I've never seen really good data on the effectiveness of a decoy. I know it works in a lab setting. It works for some number of people in a lab setting. I think that's another important point. Often, these things are portrayed as being universally effective. They're usually not. They work on
00:31:48
Speaker
more people choose Dublin after another European city is introduced than if not, but it's not a wholesale shift from 0% to 100%. It works, I think, at the margins. I think all of these nudges work at the margins to some degree. Perhaps they're additive that you can pile a few nudges on top of each other and that'll be more effective.
00:32:09
Speaker
I guess I'm a little skeptical about the overall effectiveness of these nudges, to use a generic term, without seeing them really rigorously tested in the real world. I believe they work in a lab environment. Do they work in a real world that places that marketers would be interested? I just don't know. I guess that's
00:32:34
Speaker
There probably isn't data that's non, at least that's, I'm sure there's data that's proprietary, but there's no data easily available for people like me to look and say, okay. I was about to say they would, I don't know any company that does it that would be willing to share. Of course. I mean, if you have insight into this, it's a competitive advantage.
00:32:53
Speaker
Why would you like this? Well, okay, then let me rephrase then what I meant with concept. I think what I meant was not necessarily mechanistically how it works, but to know enough about the theory to apply it into practice. Okay. Yeah, I honestly don't know. I mean, that's something I guess in my sense would be probably
00:33:20
Speaker
I honestly don't know. I mean, it's always a bit of a mystery to me how much people in industry know where they get their information from, how much of it is based on, say, primary articles or primary sources versus how much is gleaned from, you know, the Sunstein and Taylor book and blogs. And so then, you know, things often, of course, get
00:33:45
Speaker
misrepresented somewhat. Even in papers, sometimes they're misrepresented. So once they go away from that, I think it's very easy for people to think, oh, there's this nudge that just works. So I think the whole nudges
00:33:59
Speaker
I feel a bit ambivalent about, because on one hand, I think they've been really useful in demonstrating to, I mean, for me, my interest will be more in policy than in business, but to both of those fields, nudges have been really useful in demonstrating the usefulness of studying human behavior, studying psychology, basically, and what that can add to our understanding of decision-making and how it can be applied to help people make better decisions, ideally.
00:34:29
Speaker
But I think there have been a little bit overhyped at the same time. So my fear is that at some point people say, well, these don't quite have the benefits that they've been promised. And you kind of get this backlash. Yeah, yeah, exactly. I mean, new remarketing at the same backlash. And I actually think well deserved because it just went
00:34:54
Speaker
It's similar to what you described. It was so hyped up. Actually, it doesn't only hold for neuromarketing. Actually, anything that is so hyped up and so much expectation is put onto it is just sooner or later going to, well, implode. Yeah. Even if it worked, but the expectations are just so high.
Challenges in Bridging Science and Industry
00:35:13
Speaker
Yeah. I mean, I have some sympathy for people trying to use science and industry because
00:35:20
Speaker
Uh, I mean, I think that the goals are quite different. So, I mean, in science, we're looking for, you know, to sound very pretentious. We're looking for truth, right? We'd like to know some, some, some baseline knowledge that we feel, okay, this is how the system works, whether it's, this is what the system outputs and behavior are. This is how the system works in mechanisms, but that's kind of our goal. Of course, in the industry, the goal is somewhat different, right? It's, and each industry or each application will have a separate goal.
00:35:50
Speaker
But I remember a conversation I had a long time ago with a neuromarketer, a famous neuromarketer, and their idea was that, well, no, we don't, and my, I was using the scientific, I was on my scientific soapbox basically saying, you know, you can't possibly tell anything to any degree of statistical certainty from, you know, putting four people in a scanner and showing them a commercial. You just can't do that.
00:36:16
Speaker
And they, you know, after some point they said, okay, that's true. And I agree with you. However, what I'm trying to do is give executives, advertisers information and they don't necessarily require it to be beyond, you know, reasonable statistical certainty, right? They just, if it's a preponderance of the evidence or if it's a, it's a hint in favor of one commercial versus another commercial, that's what they need. So that did maybe
00:36:45
Speaker
I mean, made me realize that the criteria are quite different. The criteria are different, but I would disagree with him. That is what those people would want, but you can only do that if you have enough statistical power, even if it's just a hint. I would rather say that this is a question of effect size and not statistical significance per se.
00:37:14
Speaker
I think from a statistical perspective, yeah, you're right. I think from their perspective, it was like, okay, you know, we won't get bogged down in technicalities, but they were saying, you know, if we have a value of P.2, right? Which in a scientific paper, that would be laughed at, right? People would not take that seriously. They were saying, we view this as there's some evidence for
00:37:37
Speaker
commercial A over commercial B. And they were going to present it to the executives there. We can't be sure, but at least there's some hint that the brain prefers commercial A over commercial B. And their argument was that that's useful information to know. And if the executives want to take that on board, or if they were just going to flip a coin anyway, why not use this? Right. I mean, OK, I agree with
00:38:02
Speaker
providing the executives with this type of information, but I do think that that type of threshold is the wrong one to pick. I agree with you on that. There are other ways of approaching it. There are other ways of how you can approach it. I guess what struck me at the time and what struck me was that the concept kind of opened my eyes, but it's just a different
00:38:27
Speaker
Yeah. You have different goals, right? You're not trying to discover the truth. You're trying to help make a decision about a particular advertising campaign or a particular way of promoting your product or whatever. Right. And so I have to add a little bit more sympathy for in your remarks. No. Yeah. No, it's the same. What you're describing now is also something that I had to realize over time, slowly and slowly, because
00:38:52
Speaker
I mean, when we started, I was so, no, this has to be done by the book properly. And the more I tried to do it by the book, the more the industry actually asked me, why are you doing it this way? We don't understand what you just put here on the screen. What are all these gray shadings around these curves? And my answer was, oh, that's the standard error. And they asked, what's that? So I tried to explain it. And after I used the word distribution, it was clear.
00:39:21
Speaker
this was more confusing than helpful. And they just asked, hey, can you remove it? So I removed it and they literally said, isn't this better already? I mean, when I started to understand what you said, basically, that they are just trying to
00:39:36
Speaker
They want to use data to use it in order to make a decision. And the decision can be fairly simple. It's not about any mechanisms. It's not about any of those highly complex models. It's exactly A, B, or maybe C is in there, but it's in essence just a comparative thing. And that should then also
00:39:57
Speaker
can have a comparison across previous other commercials, let's say, just as a benchmark, which, yeah, that took me a while to wrap my head around. And how to actually do that while not feeling I completely abandon the statistical
00:40:19
Speaker
not properties. That's the statistical rigor. Yeah, exactly. And I mean, in all honesty, sometimes when they say we don't care about this, I'm like, well, okay, then that's, that's up to you. Yeah. I mean, we've been approached a few times here at the donors about, you know, doing research, which is more applied, um, or in the kind of newer marketing area. And I found it ultimately impossible to do
00:40:45
Speaker
do both research that I felt was kind of scientifically rigorous to a standard I'd feel comfortable writing a paper about and gave the customer in a sense what they wanted. Because once you start listing, this is the requirement of your, you want to know A or B. Okay. Well, it's going to take me three studies, each with 50 people. It's going to take two years. And then I'm going to give you a conditional answer.
00:41:11
Speaker
It might be a, or there's a good chance it's big. And then it's like, well. And it's going to cost you a fortune. Yeah, it's going to cost you a lot of money. So in the end, I think it's just really hard to do both. For me, I think it's impossible to do both. I don't think.
00:41:27
Speaker
I mean, they're just two different things. I mean, that's been the tricky thing. I've been to a few neuromarketing conferences, and that's the one thing I've found difficult is when people have a foot in both camps. I think that's quite hard to do because I think you end up invariably have to put your weight on one foot, and then the other side is going to be, yeah, unbalanced, I think.
00:41:50
Speaker
But I don't know, is that common in the industry? Do people really think of him to look at his industry and science and they don't have that much to say to each other? That was my experience, that they really don't talk. But I have to say, it's neither of them are at fault. No, I don't think there's a fault here. The scientists really try to. It just, unfortunately,
00:42:17
Speaker
whatever insights or let's say new findings come out of science, it just often doesn't sound newsworthy to the media outlets. And those are the ones that often the... Right, so everything is going to get filtered through the media basically. Not only, but it's really...
00:42:40
Speaker
So, for example, we collaborate with a few market research companies or UX testing companies and so on. And they have been very curious about scientific concepts of decision making.
00:42:54
Speaker
So they have asked me if I can give them courses in it. And I literally told them, look, this is not my field of expertise, like a good scientist would say. This is not my field of expertise. And they said, well, I'm sure after you read that book, you understand more than we do. That's also fair. Yeah, I think that's reasonable. That's reasonable. So then I basically, I worked through the neuroeconomics book. OK, yeah.
00:43:22
Speaker
That was quite interesting. Learn a few new, not new tools, but different ways of looking at it. So I learned something for sure. And then when I taught them, they were just blown away with
00:43:38
Speaker
all this information of, okay, now we talked a little bit about nudges, about biases, set size, default options, and so on. So I kept talking and talking and they were really like, all this information is out there? Yes. And it's not even very new. Yeah. No, that's an experience I've had multiple times now that I think
00:44:03
Speaker
I think many scientists think that what we know is, well, everybody must know this, but the basics, not the detailed mechanism. That's also what I kept thinking. Time and time again, I've given general audience talks to intelligent audiences and also sometimes to audiences in specific domains. I've talked to insurance companies about risk.
00:44:26
Speaker
And I would go in thinking, I'm not going to be able to tell an insurance company anything about risk. I mean, that's their principal business is understanding risk in theory. And maybe that's true at an institutional level. At an individual level, I mean, I've given talks which is very basic psychology and neuroscience at risk. And people have been really like, we had no
Decision-Making Research Impact on Policy
00:44:47
Speaker
idea about that. And I was recently at a
00:44:49
Speaker
an environmental conference and talked about some very basic psychological principles to a very well educated and accomplished audience. And it was all completely new to them. So I think scientists have done quite a bad job outside some of the popular science books. I've done quite a bad job about getting out the message of these are pretty foundational principles. You know, we don't need to talk about every edge case and every new finding, but we know
00:45:17
Speaker
Basically, we know a lot about how the decision process works. And I think we've done a poor job of making that available to people who could use the information. So for me, as I said, I'm interested in policy and how policy decisions can be better informed. And that's, I think, an area that we need to do much, much better in.
00:45:37
Speaker
Can you tell us a bit about that? Well, I just think that we know a lot about how people make choices and often policy is about helping people make better choices for them slash their country, their environment and so on. And the knowledge that we have usually isn't present in these policies and often
00:45:57
Speaker
policies are dominated by economists, economists who've studied decision making, but from quite a different perspective, it's the optimal decision making, right? The people who would take any amount of money offered to them in an ultimatum game because $1 is always better than $0. And that's the kind of the model, the rational model is used as kind of, this is how people will decide if we give a policy to them. Often it's in the area of incentives.
00:46:22
Speaker
That's the economists primary tool is we'll incentivize people to do something. Sometimes that works so we can charge money for a plastic bag at the supermarket and people will use less plastic bags. That can work quite well. Sorry for interrupting you, but just if I understand correctly,
00:46:41
Speaker
When it comes to incentives, I mean, there's not only monetary incentive, right? No. But that's what they mostly use. They typically use financial and moral incentives. But there's also like ethical and moral and social incentives. I mean, that's one area of research that we're very interested in the moment, the social incentives. Can you incentivize and promote good behavior using social motivations? And you can.
00:47:05
Speaker
if everyone who recycled got a gold star, that would have an effect. We've done some experiments demonstrating that. That's an example, I think, of where your economics can make a decision. Neuroscience can have a role to play in understanding behavior better and trying to
00:47:27
Speaker
don't use the word nudge, but at least alter behavior in a way that's kind of for the greater good. And so that's a direction we're taking in my lab. Recently, we're interested in questions of inequality and poverty and environmental sustainable behavior. And so how can we use what we've learned, not just from us, but from the field to kind of better inform. And that turns out to be difficult because you have to
00:47:51
Speaker
maybe similar to the professionals you mentioned in industry, they're often not aware of these findings. So you have to somehow get access to them. You have to make the findings compelling and understandable with the risk that you simplify down to an unacceptable degree. So it's a difficult or it's a challenging road to take, but I think it's one that's really important.
00:48:16
Speaker
I totally agree with you. And to my surprise, I constantly thought that we need to, so from the scientist's perspective, when we communicate something, we need to, you know, again, air quotes, dumb it down or simplify it. And then running the risk of, okay, it's not being, it's not one, it cannot make any sense anymore if you do that too far, or it's actually not accurate anymore. Either way is not great.
00:48:46
Speaker
But I started to, the more I was talking to people in industry, they actually didn't want this at all. I thought that's what they were demanding, but actually it wasn't. They wanted to know, maybe not the details, the technicalities from the scientist's perspective, but just enough so they can grasp it. They didn't mind if it's complex. They actually liked that.
00:49:14
Speaker
So I, yeah, just from my experience, I would say it's actually, again, some media outlet in between that says, oh, no, you have to make the short and snappy people have less time. This needs to be a one minute read. How much do you learn in one minute? Nothing. Nothing. I think that's a good insight that I think people aren't afraid of, uh, of complexity. I mean, especially when you're talking about human behavior, I think if you come up with some glib one liner about how, you know, we can get people to do X by doing Y. Yeah.
00:49:41
Speaker
Yeah, it's also implausible. We're not quite that simple. So I think it's a good suggestion to not necessarily remove all complexity. I think it has to be compelling. I think a scientist's theater disadvantage, maybe we have a scientist that we're very
00:49:58
Speaker
reluctant to state anything with any degree of certainty because we know that it's impossible to prove things. So it's very hard to get a scientist to stand up and say, I am sure that if you do this, this other thing will happen. It's always couched in some degree of likelihood or probability, which is what we're trying to do. And that's often not what a policymaker does not want to hear. That's true. They want to hear, OK, I want people to recycle. Tell me what I need to do to get them to do that.
00:50:26
Speaker
And I think as scientists, maybe we have to be willing to put our money where our mouth is a little bit and say, OK, here's a policy recommendation that you should carry out. And we believe it will have the best chance of achieving this outcome, as opposed to being saying just, well, here's the data. We don't know. We don't know. See you later. I mean, economists are quite good at being sure of themselves and sure of their models. And maybe we need to take a leaf out of their book, too.
00:50:54
Speaker
Yeah, no, I think it's really, I'm not necessarily saying, I mean, I also, during my PhD time, I've been told way too often, you sound way too confident about what you're doing, about these findings and so on. But I mean, I was thinking a bit more in, I think what the policymakers would like, or not only policymakers, people in industry,
00:51:20
Speaker
to have it a bit more concrete. So as in, let's say you have a nudging effect, just as we talked about this. Let's say you have a nudging effect of you have product A, you have product B, and what is it? 50-50 normally in the purchasing. You add product C, which is, of course, nobody was going to buy that because it's the inferior product. But it shifts towards, let's say,
00:51:46
Speaker
30% now choose for product A and 70% choose for product B. That's quite an effect size. And if we were to just take the experiments, were to take the effect size and say, well, if we do this just from the experiments we did, given what your numbers are, this is what
00:52:08
Speaker
what the direction of the outcome would be. I think they would already be much more convinced. And even though you still hesitate a little bit, you don't say this is what's going to happen for sure, but you know a shift in that direction is likely going to happen.
00:52:25
Speaker
Sure. I think you can have some confidence in that. I mean, I think the concreteness is an interesting point because I think another reason for maybe the reluctance of scientists to make confident predictions about industry or policy is that
00:52:41
Speaker
the nature of our experiments. And the nature of our experiments are extremely well controlled. And we try and strip away any bit of context that might be, you know, an air mine's confounding or confusing or affect the result. So we want to only focus on the exact variable that we can manipulate. So in the case of an ultimatum game, if we're interested in whether
00:53:06
Speaker
the person who offers you the money is a man or a woman, that's all we change. Or if we're interested in knowing, does it make a difference if it's $10 or $100, that's all we change. Or if it's your products, you make sure all the products are equally visible for the same. We even change for different people, see them in different visual positions. So we make everything as controlled as possible so we can make some inference by results.
00:53:29
Speaker
And I think once then we have to talk about decision making in the real world, whether that be in the supermarket or in government, we're well aware that there's all of a sudden now a hundred other factors that come into play and that we have no idea how they interact with this one thing that we've measured. And I think that's a cause for
00:53:52
Speaker
uncertainty in scientists. I think it's also a reason we should be doing experiments where we try and start to build context back in. We have a good understanding of the basic mechanisms of the decoy effect. Well, now let's start studying this in as more and more realistic fashion as we can and see
00:54:08
Speaker
is the decoy effect overwhelmed by some very basic attentional process, or is it overwhelmed by the type of product it is? So I think that has to be the next stage of research where we've gone to strip everything down so we can do an experiment, then we need to start kind of building it up again. Yeah, I totally agree. I think that'd be useful to do.
00:54:25
Speaker
I mean, it makes total sense to, you try to break it down, understand the little parts, and then try to see how they interact. Yeah. That makes total sense to me. We've done very well with the first part. We're very good at experiments, and we're very good at isolating the one thing we want to look at. But then we don't tend to, so we take a real life problem, we distill it down to an experimental scenario. But what we don't typically do is go back to the real life problem again. So we kind of stay down in our little scientific hall.
00:54:52
Speaker
a bit and I think we could do a better job of coming back closer to real decisions that people are faced with. Can you name a few examples where industry or whoever implemented it used decision-making theory in an applied setting to actually cause some type of behavioral change for the
00:55:15
Speaker
for the better of whatever they wanted to achieve? I guess I'm not sure about industry. You'd know better than me. I know there has been some initiatives in government to make use of some of these facts we've learned about decision making. So you mentioned earlier defaults. So we know that, for example, defaults tend to be quite powerful. People tend to be
00:55:39
Speaker
We'll typically choose a default option over switching to something else. One well-known example is the people willing to sign up for organ donation. For many years, this was kind of
00:55:56
Speaker
idiosyncratic by country. So some countries, it was the default that you were a donor, unless you explicitly said, I do not want to donate my organs. So that was the case, I think, in places like Belgium, places like the Netherlands up until recently had the opposite default. That is, you have to actively sign up and say, I want to donate my organs after I die. And so what happened is,
00:56:16
Speaker
From an economic theory point of view, this is a trivial, the decision to either tick the box or not shouldn't really matter. People's preferences should be to either donate their organs or not. The mechanism of whether you're automatically in or automatically out should be a very small amount. Turns out it has a huge impact. It was something like a five-fold difference between countries in willingness to donate organs depending on whether they were opt-in or opt-out. Wow.
00:56:44
Speaker
where, obviously, the more effort you have to go. That's crazy. Yeah. And so that has changed. So the Netherlands are thinking of moving towards an opt-out system where you're automatically in unless you do something. And that's going to change public health in, I think, a pretty dramatic way.
00:57:02
Speaker
There's been other examples of defaults in the US where people have been automatically enrolled in retirement schemes where a certain percentage of their money goes to retirement. The higher the default, people just leave it like that. You can have people actually almost unwittingly save quite a bit more money by changing the default.
00:57:23
Speaker
The other one that I know of that I think is going to be quite important is the idea of social comparison and social norms. And this is a bit closer to my own research where we know that people can be quite influenced by the behavior of other people and also what they view as the kind of norm of behavior in a particular country. So from your homeland, if you stand at a
00:57:46
Speaker
pedestrian crossing and it's, you stand there until the little green man comes on. Otherwise you'll stand there forever. Yes. Um, yes. Where as you go somewhere like Ireland, where I'm from, people would look at you as though you're insane. If you just stood there and there's no cars coming, just cross the road. Um,
00:58:04
Speaker
And from going back and forth between these countries, I tend to conform with the norm. So when I'm in Germany, I stand there with everybody else. Even when there's nobody there, I stand there because that's kind of the norm in this country. And so I think people are starting to recognize you can take advantage of these norms and these social factors to try and
00:58:22
Speaker
motivate behavior. So I think one nice example is they did a big test in this in the US where people got their energy bill every month. And in addition to being told they used X kilowatts of energy or so on. And they were also told how much you use relative to your neighbors. Because people have no real sense of is this a good amount or bad amount if they're just given a number of kilowatts.
00:58:48
Speaker
And the benchmarking basically comes in here, right? Yeah. And so the benchmarking was you're using more or less than your neighborhood average. Yeah. And that was all. So there was no sense of this is good or bad, just this is the case. And what they found was that people who, of course, use more energy than their average tended to reduce their energy accordingly. So people wanted to be kind of closer to the norm, closer to the average, and move down. Now, one
00:59:15
Speaker
One issue, of course, with nudges is that you're not always nudging people in the right direction. So this worked. But the problem was there was also people who use less energy than average every month. And what they found with those people is they start using more. So there was a perception that you're kind of supposed to use the average. And if I'm well under average, well, I'm going to start blow drying my hair every morning and leaving the fridge open or whatever. Don't have to feel bad about it.
00:59:39
Speaker
Yeah. So I use more. And so this is called, now we call this a boomerang effect where there was a good intention, which turned out to be bad. So in the second set of studies, they used the social norm, but they also use some kind of social labeling. So the second round they gave people also, this is your neighborhood average and you're below or above.
00:59:58
Speaker
But when people were below the neighborhood average, they got a little smiley face, right? So it was like a good job. You're doing good. And when they were below the average, when they were above the average, they got the frowny face. So it was like, you know, you should do something about this. I get one of those.
01:00:13
Speaker
There you go. Not the negative smiley face. You get the smiley. Yeah, but there is a smiley on my energy bill. I mean, I can tell. There are smileys on it. You just need a smiley. You must be doing good if there's a smiley. And in that test, it had the desired effect. So people who use more than average tend to reduce to the average. People who use less than average stayed below the average because they were getting their smiley faces and reward.
01:00:38
Speaker
I think that's a nice example of where you're not going to cause radical changes in energy use, but you get enough at scale that it can actually have a real effect. Also, it's basically free to the provider. It's free to the energy company. You don't need to give any financial rewards for reducing energy. It happens naturally as a consequence of people's
01:01:03
Speaker
willingness and desire to follow these norms. So I think that's going to be, these are, these kind of interventions, I think have potential at least up until people feel they're being tricked, which I think is another, uh, an issue with nudges that people can feel manipulated quite easily. Right. And sometimes it has the opposite effect or you feel people are feel, okay, the government are trying to tell me what I should do. And then the trust is gone.
01:01:28
Speaker
Yeah, so I think there may be a limited range of areas where these are useful. Things like public health, maybe, or environmentally sustainable behavior, where most people agree this is pretty much a good thing to do.
Social Media and Behavioral Impact
01:01:43
Speaker
But I think if these
01:01:46
Speaker
these things overreach with extreme taxes on sodas, which some governments have tried to do that hasn't worked so well. I think there still needs to be research on what are the limits of people's willingness to trust, as you say, and also feel not
01:02:05
Speaker
Yeah, talking about a boomerang effect. So this is this is the story how I heard it. I mean, whoever's listening, please correct me if you hear something wrong as a fair chance. This is actually not true. But what I heard was that before on Wall Street,
01:02:21
Speaker
There was no, it was kept secret, you know, the big bonuses and salaries. And of course, back then, it was also outrageous how much bonuses and so on have been paid out to executives or managers in certain positions.
01:02:38
Speaker
So the government has made it made it obligatory that those numbers will be made public with the idea of, oh, no, there's going to be so much public outcry that those executives will be.
01:02:55
Speaker
sort of like shamed into submission. But that's not what happened at all. What happened was that the people working with those executives, they saw, I'm doing the same work. How does he get so much more? I want more. And that's then what happened. Actually, it started to inflate everything even more instead of, well,
01:03:18
Speaker
pulling it down. I think that's a really good example if it's true. I think it's a really good example of how you can enact policy to have one effect and fail to consider how people will psychologically respond. And it's still transparent, by the way.
01:03:33
Speaker
As far as I know, it's still transparent. But of course, you know, you're going to mean what we started talking about. You're going to generate these feelings of unfairness and inequality. If I'm doing the same job, maybe even working harder, then this guy is making twice the money I am. I mean, people really don't like that. And even if all of a sudden they're perhaps previously
01:03:55
Speaker
pay packet, they're perfectly happy, but now seems completely inadequate. How could I possibly live on this? It feels like it must feel like you've been taken for a ride. It feels like you're being taken advantage of and someone else is profiting off here. It generates all these, I think, these cognitions that are not conducive to a happy life.
01:04:16
Speaker
But I think there's an example where psychologists slash cognitive neuroscientists slash decision neuroscientists kind of, I think, usefully inform on this is likely to happen if you enact these things. And again, often these
01:04:31
Speaker
Policies are based on these economic principles of kind of rationality, which is basically saying, I should be happy with my packet. What you're making is irrelevant. But of course, it's far from irrelevant. It's always a comparison, I guess. What was it? The neighbors
01:04:50
Speaker
Lawn, a grass is always green. Grass is always green. On the other side, always. Yeah. And I think that's true. And it's very hard not to look. Not to look in your neighbor's yard. It's so much greener than mine. So I think that's the nature of us as humans that we're social animals and we engage in these social comparisons constantly. It's unrealistic to expect us not to do it. So I think we have to understand how it works and can we
01:05:18
Speaker
Can we, you know, use it to help people as opposed to cause people misery? Yeah. Talking about fairness, I've seen this very, I think by now it's a super famous experiment where two monkeys are in the cage and they have to do a certain task and one monkey gets a grape and the other monkey gets a cucumber.
01:05:43
Speaker
So what happens, I mean, whoever has not seen that video, don't worry, it's gonna be in the show notes. I'm not gonna say the link out now. You can just go to show notes and then watch it. But what basically happens is, so it starts off with those two monkeys doing that task, and they can also see each other, right? And that's key. Yeah, that's key. They can see each other. And the first monkey does the task.
01:06:07
Speaker
gets a cucumber, eats a cucumber, all fine. Second monkey does the task, gets a cucumber, eats the cucumber, all fine. Second round, the first monkey starts to get a grape, and the second monkey starts to get a cucumber.
01:06:22
Speaker
Now the second monkey is very upset about getting a cucumber because it saw, the other one just got for the same thing, a grape. So given that experiment, and they did plenty of variations with that, it looks like that the concept of fairness does not seem to be a particularly only human thing. Where do you think
01:06:50
Speaker
where do you think this type of finding will lead for us reflecting upon ourselves? Yeah. So that's a, I mean, it's a, it's a wonderfully entertaining video. It is. I still watch it. It's basically a monkey version of the task I started out talking about. It's a monkey version of the ultimatum game where people are kind of unfair, unequally treated when the, the,
01:07:17
Speaker
the norm or the perception is that it will be treated the same way. So what I think it shows the fundamental nature of this feeling or state that we're kind of hardwired to
01:07:34
Speaker
object to unfairness. And we've a fairly sophisticated sense of what unfairness is. So it's not necessarily that we don't get the same thing, but it's that we're being treated worse than somebody else when we should be treated the same way. I mean, we've done some experiments in children with similar tasks where they're dividing candy between them, and he sees similar findings. So it's something that is present early on. It seems to be present in primates.
01:08:01
Speaker
And so I think that shows us that it's a pretty important principle of how we evaluate others' behavior towards us and how we make decisions often. Right. For me, when I saw that video, I just couldn't believe it. Because this is something normally not talked about so much.
01:08:23
Speaker
Yeah, I think recently there's quite a bit of interesting work now on these kind of primate social behavior. There has been work obviously on primate social interaction and things like hierarchies within primate groups and so on. Looking at social decision making has been a more recent avenue of research where people have looked at how primates, they cooperate. Also rodents now there's some work looking at say unfairness and inequality in rodents.
01:08:52
Speaker
So, I think people are understanding it may be a lot more fundamental than we think. From our early work in this, one
01:09:02
Speaker
One, I think, interesting contribution of the brain part is that the areas that we found in the brain that were sensitive to unfairness. One was an area called the anterior insula. This was particularly interesting to us at the time because up until then, people had seen involvement of the anterior insula in quite different processes. These were processes involved in very sensory discomfort.
01:09:28
Speaker
So for example, if I was to poke you with a stick, or set fire to you, or make you hungry, or make you thirsty, you would see activation. Yeah, often pain, or physical discomfort, or stress, or negative affect, basically. Now, in our ultimatum studies, we weren't physically hurting anybody, but people felt kind of, I don't know, morally injured maybe. And it was interesting to us that we saw the same area involved.
01:09:54
Speaker
as had been shown in these much more physical sensory negative affective states. And so it also gets us thinking that maybe these are quite primary emotions or motivations. They're not necessarily kind of sophisticated cognitive representations, but we may be quite, at a very basic level, we respond to things like inequality and unfairness. Yeah.
01:10:21
Speaker
Sticking a little bit to the inequality and unfairness, what do you think is happening or is going to happen when it comes to the future of decision making and how it is being, well, let's say the concepts we just talked about, do you think they are
01:10:43
Speaker
they're going to be regulated at some point. And let me put this in a bit more of a context. If you think of the current social media platforms of using specific perceptual decision-making, memory, attention, tricks based on neuroscientific findings, psychology findings, about principles of slot machines,
01:11:10
Speaker
Do you think that, and talking about being manipulated, right? Do you think that at some point, the governing bodies will step in and say, okay, this is going too far?
01:11:24
Speaker
I don't think so. Just for my opinion as a citizen, I can't see that happening. I think it maybe should be considered. The principles, it's really interesting to see how many of the social media platforms operate on a system
01:11:43
Speaker
almost perfectly designed to promote constant checking with intermittent reinforcement and also not just social media, but also things like video games. It's amazing. My son plays a video game sometimes and I sit in there and I watch it and I go, this is like really clever the way they designed this. It's very hard to resist.
01:12:06
Speaker
especially for developing mines. It's very hard to resist. Whether there should or shouldn't be regulation, I do think there should be awareness of that these
01:12:18
Speaker
the way these media operate are going to encourage a certain type of behavior that can be quite difficult to not adhere to. I think there should be greater knowledge about that. When we talk about, say, the power of Facebook or the power of social media, I think there should be better understanding of how they work and what kind of triggers they
01:12:42
Speaker
they hidden us and what buttons they press because they're not designed accidentally. They're designed by smart people who have very smart people. Whether they read the scientific papers or they just have some kind of implicit understanding of how people operate, they really, as I said, hit all the buttons. And so I think it behooves society that we understand that. And even if we're not going to regulate it, we should at least be able to say, OK, this is why it's hard for people to
01:13:11
Speaker
to stay away from these media. So I think, again, science could do a better job about getting involved in that. We have a tendency not to get involved in things because we want to stay neutral, and we want to show we're just about the truth. But I think there are certain platforms we could speak up a bit louder in. And again, not necessarily with a value judgment, but just to say, if you're operating
01:13:39
Speaker
an app or a game or a social media platform which gives this intermittent reinforcement and encourages people that you really need to kind of keep up or you're going to miss out. That's going to promote certain behaviors in the population that maybe we want to think about are they positive or not.
01:13:54
Speaker
No, I like what you said with, I mean, these are very smart people. And I don't remember where I heard this quote, but somebody, maybe it was a friend who said that. It's like, look, the smartest people on the planet, and all they do is try to get you to click on something. And I'm like, oh, that's...
01:14:10
Speaker
Yeah, I mean, they talk, I've seen some quotes, at least from, you know, Silicon Valley people talking about things like dopamine hits, right? They do have some awareness, I think, of... No, I think they are very well aware. I just think the users are... The users are probably oblivious to this. And of course, we could probably think of a better use for all this intellect. That's true. Now, I think it's quite interesting because there are...
01:14:38
Speaker
There are studies in for some time now longitudinal studies that look into the effects of social media use on your mental health.
01:14:48
Speaker
And I was very concerned when I read those papers. It's not just one. I mean, they have been replicated and so on. I thought this is alarming. And I think then the scientists have nothing to worry about because the data is crystal clear. I'm not saying regulation. I'm not saying I know what the solution is or something. But of course, this isn't new. I mean, we've known this. I mean, slot machines make a perfectly designed mechanism to
01:15:18
Speaker
keep you coming back and give you just enough reward to keep the temptation going. We've had slot machines for a long time. These are just being applied in a different format at a different scale now, but sometimes there's nothing new under the sun, as they say.
01:15:37
Speaker
I agree. It's just a different packaging, I guess, and it perception-wise looks new. Also with a much bigger outreach. Everyone doesn't have a slot machine in their pocket. Maybe that was a way of implicitly regulating it. You create a building that you have to physically show up in and put your money in, and that's maybe one way of implicitly regulating it. That, of course, goes at the window when
01:16:04
Speaker
everybody has it in their pocket and on their desk and that home and everywhere. So that changes things. Yeah, exactly. No, I find that, I find that very concerning. And I think it would be great if we all just talked a bit more about this. Yes. Also, of course, all the people listening to your podcast and their electronic device. That's also a good use for these. Yeah. Yeah. I think that's a good use. And I hope they're not just sitting around. I hope they're cooking something while listening. Yes. That's what I like about the podcast format. You don't have to.
01:16:31
Speaker
Yeah, it doesn't tie you to your device. Exactly. It doesn't tie you to it. I have a follow-up question to what you said before. So you talked about default options. But the examples you gave, they both had a temporal discounting.
01:16:50
Speaker
mechanism tied to it. So you gave the example of organ donation and you gave the example of retirement. Now both of those are of course super long term and people are terrible at thinking about the future. They often don't even want to do it. Do default options also have a strong impact if there's no temporal discounting factor
01:17:11
Speaker
Yeah, that's interesting. I hadn't thought about that, that most of the examples do have this sense of I'm choosing now for some outcome in the distant future. I mean, I think they do because, I mean, you go to these internet sites or web forms and there's often an option chosen for you and you just don't click away. So I think those kind of defaults still work. I think it's a matter of current debate, why they work is it's
01:17:38
Speaker
Are people too lazy to click away? I don't want to do it. I'll just do what I'm told. Or do people have this implicit assumption that because something is being defaulted that somebody has decided that this is the best thing to do, right? So it's a way of basically saying, I trust whoever has defaulted this for me, that this is the best. I don't care enough to think again.
01:18:02
Speaker
But I do think they probably do work. I mean, I guess in marketing, there must be quite a lot of knowledge on defaults, right? That if you go to a site or you're trying to purchase something, one thing will be, here's what you're going to have, or you can choose something else if you want. So I think there is a power of defaults, even
01:18:23
Speaker
even in the present domain. Yeah. Yeah. All right. Okay. There was this personal curiosity of mine. I just, something I noticed. Yeah. That's a good point on the previous research. I just have two more questions. Sure. Um, so the first one is it's a bit of a segue back. Cause you mentioned the book thinking fast and slow by, uh, by Danny Kahneman, Daniel Kahneman, I shall say, I have never met him, uh,
Understanding Decision-Making Frameworks
01:18:47
Speaker
unfortunately. And I mean, this book I hear in industry, this, I hear this being cited all the time.
01:18:53
Speaker
All the time. All I hear is system one, system two, system one, system two. I hear this all the time. And then I hear, oh, a system one, a subconscious system two is conscious. And then all these labels are just thrown onto these two things. Now, what do you think were good things that came out of this book being published and maybe some quite bad consequences? Right. So, well, personally, I think
01:19:17
Speaker
more good than bad, for sure. And the positives are simply exposing this area of research to a broader community. We talked about many people aren't aware of these kind of what people like me in the field will consider really basic and obvious stuff. And much of the research cited in the book is quite old. I mean, it's stuff they've done in the 70s and 80s and still stands up. But it's kind of classic findings from
01:19:45
Speaker
from judgment and decision-making research. And so I think anything that can be done to bring science in a non-hypey way, which I think is what he does quite well in the book. He doesn't sensationalize and preach, I think is a good thing. It shows how experiments operate, it shows how scientists work, and it shows the kind of conclusions you can draw after a long program of research. So I think that's really positive.
01:20:12
Speaker
Um, the system one system two thing I feel has been maybe over, uh, abused. Yeah. Appealing people have taken it a little bit more literally than I think it should be. I mean, in my mind, it's a, because he does say explicitly in the book, I just, just to point it out, you say explicitly say in the book,
01:20:35
Speaker
these are just concepts to make it easier to understand it from a framework perspective. That's all he says. But it's very easy to go down the road of thinking, okay, because it makes very nice intuitive sense. We've had this metaphor for many, many, many years. I think it was Aristotle talked about the
01:20:57
Speaker
the voice of reason on the back of the animal of instinct. Many people have come up with two system models over the years. It's a very compelling concept that you've got rationality and emotion battling it out, and it captures a little bit. It's Freudian almost.
01:21:18
Speaker
But I think there's a danger of taking it too literally, is that we literally have a part of the brain that's system one, and a completely different part of the brain is system two, or somehow fighting it out all the time. Which is just not true. No, it's not true. And we're becoming much, I mean, somebody earlier research, including some of mine, did talk about this, again, in a kind of a conceptual sense, that we may have areas of the brain that are more predisposed to kind of deliberative, cognitive processing, such as areas of the frontal lobe.
01:21:47
Speaker
versus areas of the brain like the interior insulin mentioned, which seem to be more devoted to more affective processing. But they're certainly not a one-to-one relationship. Another, when you asked about recent developments a while ago, I think very important recent development from a more technical point of view has been this notion of brain networks, that a single brain area isn't going to tell us much, that what we really need to think about is a network of areas working together to perform a function.
01:22:16
Speaker
And I don't think there's very good evidence or really much evidence suggesting that there's a system one network and a system, a separate system two network. Many areas are involved in multiple things. Yeah. So what it's a very useful metaphor. And I think, you know, it's a very nice concept that that's, I think the, if people indeed are thinking in terms of how do we activate system two and these people are, you know, is system one the dominant, I think that's maybe not,
01:22:45
Speaker
Very useful way of thinking about it. Yeah and to be fair to some other authors which also Wrote great books who also wrote great books, which is a robot sale Denise influence. Mm-hmm I think it was that was that was the first book I actually read on this topic. Yeah, I was blown away. That's a great really good He's a terrific speaker. Yeah, and you can I think the follow-up was
01:23:08
Speaker
pre-suasion, which is also great. I suggest to read it in that order because it does use some concepts of the first one. One nice thing, Cialdini did, of course. Oh, Cialdini, sorry. Yeah, I talked to her about
01:23:23
Speaker
about how often we can get stuck down in this little hole of scientific rigor. He has, I think, gone to some quite big efforts to do more of what we might call a field study in psychology, where he actually looks at decision making in the wild, and as much as is practically possible, and in conjunction with trying to have scientific rigor, which is a very difficult combination, I think he's been a really good proponent of those. And so the books feature some really nice examples of that.
01:23:51
Speaker
And one more that I know of is Dan Ariely's Predictably Irrational, which I also enjoyed. So if people want to know a bit more, like other types of resources that are listening, is there something that you can recommend, except for the books? Yeah, I don't really read those books. You don't really read those books. I mean, I hopefully should know what's in them, because if I'm reading the books, then that's a bit of a problem.
01:24:19
Speaker
No, I mean, you know, I've, I've clear I've read those books, actually, I've read Nudge, I mean, so they do a nice job at, again, presenting this research to a, to a, to a, to a, a general audience. And so I think there's a valuable role to play for those. I think
Communicating Scientific Research Effectively
01:24:37
Speaker
you, you, you point out maybe a
01:24:40
Speaker
where, again, an area that scientists could do, I think, a lot better is making their work kind of digestible to a non-scientific audience. And often we don't, or we usually don't do that. And it's left to people like Kahneman and Dan Ariely and Cialdini to translate it in their books. Or there's the medium of TED Talks, which I think are often much too
01:25:08
Speaker
simplified and much the style of it talked. I've given one, so I was told clearly what I was supposed to do. The E, I think, stands for entertainment in TED. I have no idea what it stands for. I think it's technology. I don't know. Certainly, you're supposed to give an entertaining talk.
01:25:28
Speaker
I think we could do a better job about providing little potted modules on our research that people could find and look at. Designed to be informative and somewhat compelling, but not gloss over all the complexities.
01:25:47
Speaker
And that's, I think, a responsibility of scientists to be better at this kind of outreach we could do. Well, coming to my final question, which is actually not one of mine, it's one from a guest from episode number three, Ivar de Lange, who talked about the UX of cocktails and bars. And when I asked him if you had a scientist
01:26:08
Speaker
at your disposal who could look into whatever you want. What is it that you would want him to look into or look into? And he basically said, what are great menu designs
01:26:22
Speaker
for bars and restaurants. And another one, which was basically, what are great furniture designs, and how does posture impact behavior? So on the latter, I don't want to burden you with the latter one, because I actually have no idea who I would have to ask for this one. But on the previous one, the menu designs. Yeah, well, there are people who study this. I mean, there are psychologists who study menu designs, and they study things like, well,
01:26:49
Speaker
There's some things, I think there's two aspects of things that are studied there. One is just a physical layout of the menu. For example, people I think don't tend to order things close to the edge of the menu.
01:27:04
Speaker
You want to, I think, put the items you like ordered kind of more in the middle of the physical menu itself. You know, how do you advertise specials, things like that. So I think there's a little research field on that, people who look at exactly how.
01:27:21
Speaker
how the perceptual aspect of the menu is going to impact ordering. I'm by no means an expert in that, but I know just the physical layout. People typically don't like ordering the first or the last thing. They don't like ordering the cheapest or the most expensive thing. The cheapest glass of wine rarely gets ordered. Neither is the more expensive. I think things you really want to sell, you might want to put more in the middle.
01:27:47
Speaker
Yeah. And when it comes to, I also find curious about this because good friends of mine, they opened their own restaurants and bars. And when I saw their menus, my first question was, why do you have so much stuff on the menu?
01:28:03
Speaker
It took me 10 minutes to read the entire manual, and I didn't even make a decision yet. I still need time to process all this. Can't you just make some type of decision tree here? There's a very well-known concept in decision-making research called the paradox of choice.
01:28:23
Speaker
From a purely economic rationality point of view, the more choice you have, the better. If I give you a choice of 100 entrees, the likelihood of you finding the entree that exactly fits your dietary requirements, your mood today, how full you want to be is increased. That's just mathematically the case. The more choices you have, the better.
01:28:45
Speaker
it turns out people do not want more choices and people would rather have less choices than more choices in a given situation. There's plenty of experiments demonstrating this that given the choice people prefer, they don't want no choice, but they want less choice because it's just simply overwhelming to have pages and pages of menu items. And so that's a well-known case where economic rationality clashes with what people actually want reality. So people don't want infinite choice, despite what
01:29:15
Speaker
Marketers like to give them and supermarkets like to provide. Some reasonable number of items is usually considered preferable because it reduces the stress of choosing. Yeah, that's
Conclusion and Contact Information
01:29:28
Speaker
definitely good to know. I will pass the information on. Yes. I'm afraid our time has come to an end. Thank you so much for your time in this conversation. If people want to reach out to you,
01:29:44
Speaker
What would be a good way of how they could do that? Well, you can, uh, contact me, um, via email. So, um, I think you can probably put it in the notes. Oh, I will, I will. Yeah. Yeah. A.san fee at donders. Okay. Dot or oo dot nl. Um, or you can look at my website, Alan, san fee.com. Um, I think there's contact information there.
01:30:05
Speaker
All right, that's great. No, I will definitely put that into the show notes so nobody has to write anything down. I wish I had Twitter handles on Facebook. Ah, no. No worries. No worries. I stay away from that. Well, basically what we talked about just said, that doesn't sound too bad. Well, thanks again. And to everyone that's listening, you have a great day. Thank you, Adriel.
01:30:29
Speaker
Hey everyone, just one more thing before you go. I hope you enjoyed the show and to stay up to date with future episodes and extra content, you can sign up to the blog and you'll get an email every Friday that provides some fun before you head off for the weekend. Don't worry, it'll be a short email where I share cool things that I have found or what I've been up to. If you want to receive that, just go to ajmal.com. A-D-J-M-A-L dot com. And you can sign up right there. I hope you enjoy.