Think fast! But ideally don't.

In this podcast, Nick, Peter, and Fraser discuss Daniel Kahneman, the Nobel-prize-winning psychologist whose scholarship challenged economists' presumed rationality of human decision-making. From coining the term 'cognitive bias' and the development of prospect theory with Amos Tversky in 1979 to articulating the frequent errors which affect our decision-making heuristics, they reflect on his lasting contributions to economic science.

A few things we mentioned in this podcast:

• Daniel Kahneman, Nobel laureate who upended economics
• Dual Process Theory
• Living, and thinking about it: Two perspectives on life
• Peak-end Rule

For more information on Aleph Insights visit our website https://alephinsights.com or to get in touch about our podcast email [email protected]

Hello and welcome to the Cognitive Engineering Podcast produced by me, Fraser McGruer, for Aleph. In this series of podcasts, we take a look at interesting topics and discuss what we think they tell us about analysis and decision making. I'm here with Peter Coghill and Nick Hare of Aleph, and this week we're discussing Daniel Kahneman.

Okay, cool. So actually, two things. One, I can definitely see why we're talking about this guy. And indeed, I hadn't realised it. I think I've got his book on my shelf.

Oh, yeah, you should. You should definitely crack open. I should definitely read it. Oh, it's excellent. It's very readable. It's not kind of dense academic work.

And also, actually, yeah, we talk implicitly and actually explicitly, I now realise, about his stuff on regular occasions here, right? This whole system...

Yeah, I mean, as far as I'm aware, he actually invented the term cognitive bias. And the fact that you hear that all the time tells you enough, I think, about the impact he's had.

So it's clear to me why we're talking about him. And then you've gone on to say kind of how we want to, that's what I was going to ask, how are we going to go at this? And you said, is we're going to do a survey kind of thing? Is that a sort of survey approach and a skip through his stuff? Is that about right?

Yeah. And I think that kind of respects his methodology as well. Because, I mean, I don't think we've got to this bit yet. But just, he was a real documenter of phenomena. He wasn't a big theorist. Like he wasn't very into trying to work out why these things happened. He was much more interested in documenting the weird things that happen. And you know, he never really tried to integrate them into a big theory of, you know, which somehow explains all of them. It's more like, well, actually, here's an interesting thing, here's an interesting thing, here's an interesting thing. So I think doing it that way makes sense.

Okay, so how are we going to go about this? Look, let's start off with yourself, Peter. Can you give us some headline stuff, please?

So Nick touched on it. One of his early collaborations with Amos Tversky was on prospect theory, which is the thing, is one of the key things that kind of underpins the behavioural psychology science that has come out of Kineman's work. So I always find it really amazing that until he came along, everyone was quite happy, all the economists were quite happy, oh, we've got these rational agents, they're these simple models. And they just behave in a rational way, weighing up costs and benefits and making sort of solid, rational decisions. But in retrospect, it's now obvious that that's completely false, because you know, in your own mind, that you don't do that, right? Quite often, the decisions you make are snap ones, or rushed or suboptimal. But had you taken more time or had more time or had more cognitive space to think about it, gathered a bit more data, then you might have made a better decision. But you know, you make decisions all the time. So I always find that remarkable that it took so long for some people to sort of question that really, but he did a brilliant job of explaining that. But prospect theory, it describes how people make decisions under risk and uncertainty. So if you're buying a coffee, but you're not sure if you want the latte or the flat white, maybe you've never had coffee from this coffee shop before, how do you make the decision there, given that there might be a cost involved with monetary cost, but there's also a cost in terms of your enjoyment? How do you make your decision in that situation? And this theory really was the thing that challenged the underlying economic assumption that people were rational actors, and they make balanced decisions.

When did he publish this?

And so one of the things you've said there is he sort of, you know, observed that, and then went on to go, well, what might be the implications of this be, for example? Did he also go on to examine further into why this question of risk aversion, or did he kind of leave that, you know, like, hey, there's something going on with the abyss?

So I think there's a common theme that I would say generally, no, like, I don't think he considered that his job to speculate about what the origins of these things are. Personally, I think speculating about the origins of these biases is really interesting, because I think, I suspect that all of them come from, there must be some kind of optimal trade-off in terms of, you know, if you think about the evolutionary environment, where, you know, perhaps this simply reflects the fact that the evolutionary environment is not just uncertain, but rife with extremely fatal consequences if you make the wrong decision. And it sort of, in some sense, seems completely rational to go, right, I have a thing. That's quite unusual. We're very lucky to have that thing. We do not know when we're going to find another thing like this. So let's really guard that, you know, whatever it is, whether it's, you know, you've got a fruit tree that happens to bear fruit all the time. And you think, well, that is, you know, we could go, we might go to the next valley and there'll be five of them. But, you know, the thing is that the consequences of doing that, you know, on average, I think tells you about the uncertainties in the evolutionary environment. In some sense, we sort of, actually, it's not usually worth taking those risks. But anyway, he didn't really do that. But actually, you know, the similar kind of thinking can, you could apply to lots of the other things he looked at. So I'll just say one of his other famous discoveries.

Just before you do, if I might interrupt you for a second. This just years ago, I had a really dismal job where I've had lots of dismal jobs. Not this one, of course. No. And where I was a telesales guy ringing up people doing cold calling. Yeah. And as much as I, you know, didn't enjoy it and soon got fed up with it, I actually learned lots of wonderful things that have stood me in all sorts of good stead for all sorts of things in my life. Anyway, one of the key things they taught us in sales, basically, you're doing lots of things. But if you want to strip away everything, all you're really doing is trying to elicit or tune into two motivations, which are fear and greed. And that's all you're trying to do. And that really ties into all this. Absolutely. Yeah, those things work.

So I think another, we should probably zoom in on two or three of these before zooming out a bit to some of the much more fundamental implications that his work has. But I think we can't talk about Daniel Kahneman unless we talk about Linda the bank teller. I agree. So have you heard of Linda the bank teller? No. She's very famous in cognitive psychology. Okay, cool. I tell you what, well, if you haven't heard of her, let's do this experiment. I can be Linda. No, you're not Linda. I'm Linda's customer. We're talking about Linda. Oh, okay. Yeah, no, Linda is not actually known to you, but I can tell you about her. She's 31 years old, single, outspoken and very bright. She majored in philosophy. As a student, she was deeply concerned with issues of discrimination and social justice and also participated in anti-nuclear demonstrations. Now, what's more likely? A, that she's a bank teller, or B, that she's a bank teller and active in the feminist movement? Which of those is more likely?

I mean, my honest answer is I don't know.

What is the appraiser's gut telling?

What's my hunch? Is it more likely she's a bank teller or that she's a bank teller and active in the feminist movement? It should be C. Sorry, it should be – there's not even a C. It should be B. It should be the latter. Right. That's what I meant to think.

Well, is that what your gut tells you is the case? Do you feel like the answer is actually B? No, I don't, actually. No, well, good. What I really feel like is that she's just a bank teller. Well, good. Because, actually, the thing is, like, if she's a bank teller and active in the feminist movement, she must also be a bank teller. So her being just a bank teller must be more likely than that she's a bank teller and active in the feminist movement. It must be, right? Yes. Yeah, yeah. So this is the conjunction fallacy, where if I add something to a scenario, it can sometimes make it seem more likely. Which it should never, should never. If I add conditions to something, it must make it less likely. Yeah, yeah. So what this touches on is the fact that our assessment of probabilities is not based on some kind of mathematical calculation where we go, well, how many bank tellers are there? How many of them are feminists? And let's multiply these together and so on. What instead we do is kind of imagine the thing and then go, well, how close is it to this other thing? Right? So bank teller, well, that's a bit weird. Like, you'd expect her to have become a, you know, maybe a writer or a journalist or something. So that doesn't seem very likely. But, ah, feminist bank teller, yeah, that's Linda Roll, right? Yeah, yeah. And so that's really, I mean, that tells you something really interesting about the sorts of ways that, you know, what's going on in our head. Very similar phenomenon of the availability heuristic, which is another thing that he identified, where if we're trying to think how likely something is. So if you, if you are someone, you know, what's more likely that, you know, there'll be a terrorist attack or that you'll die of bullous pemphigoma. People can easily think of terrorist attacks and they seem much scarier. And people tend to think that they're more likely. People think that murder is relatively likely, for example. And all these things are much, much less likely than things like dying in an accident in your bath. But examples don't come as easily to mind. So that's, that's, and so that's the availability heuristic is instead of looking at the data and dividing one thing by another thing, you go, how easily can I think of examples of this? And that is what you do instead of trying to judge the probability of something.

Which, I mean, two things, which is, it's interesting, which is, you know, this information and this study being available and known. And yet it's interesting to what extent that people think any differently or behave any differently. Or I guess another direction you can go is, OK, this is great information. And having this information, we can, we can, we can help ourselves think differently. And but how do you actually do that either on an individual level or a societal level? Let's say it's tricky. I think we should come back to that at the end, because I think I think that's really the key question is, what do we do about this? And sorry, just second thing, I know you'll enjoy this analogy. Well, this is what Gary Lineker says as well. I'm sure it is. Which is analysing a football game. I remember once him seeing and he was, he was actually famous as a striker. Well, not as famous. That's what he was. But he was famous as a kind of goal poacher. And he was just the guy who is not a spectacular goal scorer, but he gets the goals. That's what he does. And I saw him analysing this match once and essentially had two people charging in on goal. And the choice for player A, striker A, is either pass it or shoot yourself. OK. And what Gary Lineker said is regardless, every time the striker A should just go for it, he should just shoot. And the reason why is by passing it, even though the other guy might be in a better position. I mean, obviously there's judgments around this. Even though you might be in a better position, you're actually adding another level of something. You're adding another level of something can go wrong. OK. Which is a bit like, is it Linda? When you're adding another bit of information. Right. Another possibility. Just by adding a possibility, another condition, it makes it, it just logically, it must make, mathematically makes it less likely. I'm sure it's exactly the same.

Well, and I think the thing is that every fibre of my being wants to say that it's more likely she's a feminist bank teller than that she's a bank teller. I know that's wrong. But I can't help feeling it. Right. And I think and so, you know, I think we probably want to touch on. Well, actually, we probably want to discuss the two systems. But I think, you know, so just that recognition. OK, well, there's something in me is wanting to give this answer. But learning not to trust that something, at least at face value, is a very important first step. Yeah. So, I mean, I think, you know, we could sit here and list a load of interesting cognitive biases. There's no need. Wikipedia has done that. And so, you know, or just by the book. But I think the work that he did touches on some quite fundamental things. So one of them being this idea of utility, of like what basis it is that we make decisions. What are we trying to maximise? Are we trying to maximise, you know, happiness or well-being? What do those concepts mean? And, you know, and how good are we at predicting it?

So, I don't know if psychology is the study of how our brains process and generate well-being and happiness. And so Kahneman did quite a lot of work in this area. And a lot of his work was about how people perceive and remember previous experiences of comfort or pleasure and then pain and discomfort. And again, similar to the prospect theory, whether or not we're rational decision makers, there's an asymmetry in what we remember. And similarly, we remember bad things more prominently than we do the good things, which has all sorts of effects on what choices that we then go on to make. Just wanted to throw that in there.

It's interesting, just to say, like a lot of super, super uber bright people, he's meant to be an economist, right? I don't think he'd have described himself as an economist. What did he describe himself as? A cognitive psychologist. Cognitive psychologist. Because, you know, I feel like, oh brilliant, let's get an evolutionary biologist in here and talk on this. But there's so, like all great thinkers, like truly great thinkers, just crosses over into so many areas. Exactly, yeah.

Yeah, and I think this, so he identified this thing called the peak end rule, which is that how enjoyable you remember something being or how bad you remember it being depends actually quite crucially not on like how long it was or how bad it was on average, but how bad the worst bit was or how good the best bit was and then kind of how it ended. So he discovered that if you, let's say that you have to plunge your hand into really painfully cold water, for example, this is a famous experiment. And then, you know, after say a minute, you take your hand out and then compare that with where you've got to do that again for a minute. But this time the water then gets warmed up so that you kind of end with your hand in, you know, not painfully cold water. It turns out people remember that experience as being more pleasurable because it ended in a better way, even though they had to go through the same experience at the beginning. So, you know, well, that's, that touches, that indicates something, which is, I think, one of his most profound discoveries, which is that our experiencing self, the thing actually having the experience and momentarily experiencing pleasure or pain actually has very little to do with our decision making.

I presume he reversed it as well because you could, because on balance, you're still, let's say, so I don't want to distract us, but on balance, there would be a little bit of like better stuff during that, let's just say. But if you put the better, I wonder if like putting the better stuff at the beginning, so making it warm and then sort of uniformly cold, I don't know. So you end up with that cold. He probably knew what he was doing, I think. I think they thought about that.

Yeah, no, but basically that peak and end, how good was the best bit and how did it end? That turns out to be like by and large determines, you know, to a first approximation how enjoyable we think something was. Anyway, but so this distinction between the experiencing self and then the remembering self, right? So even momentarily after something has happened, your remembering self is the one which is interpreting what happened and how you felt about something. And of course, your remembering self is the one actually that sort of predicts what our utility will be in the future and makes decisions based on how good or bad we think something will be. And so an unpleasant, you know about sort of type one fun and type two fun, type one fun is fun while you're doing it. And then type two fun is horrible while you're doing it. But fun afterwards, you know, like doing some big exertion, like marathon running or rock climbing or something. I've got something on this in a minute. But I think this sort of distinction between experiencing self and remembering self means that actually experiencing self can have some really quite a lot of dis-fun, of non-fun. But it doesn't matter because it's the remembering self who's going to decide to go and do some rock climbing next week because the remembering self said, oh, that was really good, actually. But they're actually really two fundamentally different selves and they disagree. The experiencing self has totally different experiences to the remembering self. But Kahneman said, we are our remembering self. That is completely contrary to our opinion where we're the same thing. I'm the same thing experiencing it as remembering it. Well, that isn't true. And when I'm talking to you, I only ever get to talk to the remembering self. The experiencing self is the poor bastard trapped inside having to suffer for the remembering self's benefit or pleasure.

There's some intoxicating truth in this. It's absolutely true. And, you know, Kahneman will be pleased to know, if he could be pleased, that I agree with him. This is resonating. Well, more than resonates, I confirm his results.

As if the Nobel Prize wasn't enough. Yeah. Yeah. It's just got a whole lot better. Yeah.

I know the reason why, just really briefly. I was in Finland recently. And as part of that, you know, I'm a sauna guy. I like saunas. I love plunge pools. And what a great country to do those two things. And I had a sauna every single day. But there was one particular special occasion where we were at this outside hut and go ice lake. Right. And I said, right, I'll do that. You know, I'll plunge into the ice lake afterwards. And sure enough. And it was one of the most unpleasant experiences of my life as I was going through. It was. It was. I just wanted to curl up and cry. And it was great. It was primal what it was like. I did it four or five times. And why? Because afterwards, this surge, whatever it is, the endorphins. And it's just wonderful, amazing feeling at the end of it. Yeah.

And the remembering self doesn't have to go through the misery of being in the cold water. It's the experiencing self that has to do that. So anyway.

Sorry. Go on.

No. And I think, well, I mean, it's interesting that I think a lot of what people mean by mindfulness is about really trying to focus on your experiencing self and, you know, saying, am I actually enjoying this pizza? Like remembering self gets a nice warm glow after eating a Domino's pizza. Experiencing self actually probably isn't having as great a time as the remembering self thinks it is. So, I mean, it's things like that. It's going, actually, am I enjoying this cigarette? Am I really enjoying it? You know, what does this feel like? And we don't, we rarely pay any attention to that guy, the experiencing self.

And yeah, that is a key part of CBT techniques is to sort of be mindful, enjoy the moment or be aware of why you're not enjoying the moment. Be able to do something. Be empowered to do something about it. So that and, yeah, so that's sort of focus on momentary well-being. But also techniques, you know, some of his work is actually quite actionable stuff that has fed into mental health treatment, such as day reconstruction method, which is how you go about retrospectively, like rebooting your experiencing self so you can relive that experience in a more real way. Rather than just remembering, oh, yeah, yesterday was a white day because of X, Y and Z happens. Day reconstruction method is about taking you through a process which helps you relive that almost, but not like hypnotherapy. It's more like conscious walking through the day again and reliving it. Look, we need to talk about system one and two. We do. So think about rationality.

Look, we need to round this off a bit. It's a shame because we could talk about this forever, right?

The podcast is serving its purpose of being a sort of advert. Well, yeah, absolutely.

That's kind of what we are trying to do on this one, I think. So systems one, systems two. What's all that about, Peter?

So these are characterizations of two modes of thinking, which Kahneman and Dvorsky kind of described in various publications. So system one, you can think of that as your kind of primitive brain. It's fast. It's very broad bandwidth. It takes in lots of data from all of your senses, including your own body, like how hungry you are, tired, et cetera. It's automatic and intuitive. So you don't have any control over it. It's the thing that makes you catch something when it's falling off a bench or shout at somebody when you're angry and tired. It's the thing that you so often wish you hadn't done the thing that it made you do. That said, though, and it's error prone is particularly in the modern world. It's often very, very good. It's the thing that stops you. Very efficient. Very efficient. It's the thing that stops you walking out into the street and getting run over by a bus. It's very important. So that contrasts system one to system two, which is your conscious thinking. That's the thing that does mathematics. That's the thing that writes your essays as you're thinking about various topics and colliding them together. And that's slower, more deliberate, more logical, less error prone in the sense that it's more rational. But being slower, it can't do a lot of things that system two do. System one does.

I presume it's actively trying to draw on more sources of information. Yeah.

But because it has to pass through a narrow little hole, which is your conscious mind, it's limited in the throughput that it can achieve. So they come in and sort of characterize these two systems and apply that to why certain circumstances cause people to make poor decisions or aren't able to be the rational agents that everyone sort of assumes that they are. And so they have a huge influence on behavioral economics, but also things like advertising and every day to day stuff. Like, you know, where do you put the chocolate? You put it at the end of the supermarket aisle by the checkout. So when your brain is fried from trying to process all this information, you're tired and hungry, you're going to reach for a chocolate bar. Or that explains, this is what system two explains all of this.

Yeah, it's completely contrary, I suppose, to that idea of a kind of unified rational self. It's just a very different picture of what's going on with our decision making, but an important one to have. It's not totally new in the sense that, you know, people have talked about the idea of these sort of two different fundamental approaches. You know, people used to talk about left brain and right brain, and it's not totally dissimilar from that. But really, just to codify it and to put it out there and say, here's a mental model you need to be aware of. Ask yourself, is this system one going off on one? Or is, you know, is this something that I should check? Do I need to check system one's homework here? That's really a fundamentally different sort of model. And it suggests, you know, that it's not very transparent what's going on in our heads, which is surprising to us because it feels like it would be. It feels like we understand where we're coming from, but we don't. But I think just so I think the Kahneman's, some people are inclined to overstate like what the implications of Kahneman's work is. And how do you mean? Well, Kahneman wouldn't have wouldn't have said this, but I think a lot of people are inclined to go. Oh, look, people are totally irrational. We now know that people are completely rational, that they just, you know, we're just a big bag of biases and stuff. That just, Kahneman's work doesn't suggest that. And, you know, if anything, I mean, you can look at the fact that generally we don't just constantly get killed. We're not like money pumps is the phrase sometimes used for like if you were totally irrational, I could sell you some beans for two pounds and then buy them back off you for a pound. I just keep doing it by changing, you know, the color of the packaging or framing it in a different way or something. Clearly, that's not true. What these tell you is that there are certain circumstances like when you have limited information or when time is really scarce and you're under stress or when the problems are quite abstract where, you know, our reasoning that works most of the time because we've evolved that way. So, you know, it obviously works most of the time, but not all the time. And that's the point of a heuristic is it's going to be wrong sometimes. And so I think it's really useful to have these models where you can go, OK, am I in one of those situations? Should I expect, should I trust my own reasoning? You know, I'd say that's probably his legacy is to give us these ideas and these models, which allow very practical, as Peter says, they're quite actionable. You can go right. Am I likely to be getting this wrong? And to be able to sort of stop yourself and say, hang on, wait a minute, system one, put those cigarettes down.

Yeah. And there's the rub. Yeah, actioning this stuff, how you do it in the moment. And yeah, it's yeah. OK, let's stop there. OK, so I can see why we've talked about Daniel Kahneman. I mean, we've got our sort of philosopher, you know, our patron. We've got Derek Parfit.

We've got our writer, of course, Borges.

Yes, yes. And now we've got our cognitive psychologist, Daniel Kahneman. We commend them all to you. All right, we're going to stop there. So that was lovely. Thank you, as always, for listening to the Cognitive Engineering Podcast. I'm Fraser McGruer. I've been here with Nick Hare and Peter Coghill of Aleph. Until next time. Goodbye.