338 - May Contain Lies - Alex Edmans (rebroadcast)

Introduction to Drum Kit Purchase

0:00Hey, Mom, now that I have an after-school job, I think I want to buy a drum kit. Okay, Rockstar. Let's look at your account. When kids start making money, Chase has easy-to-use tools and expert bankers to help them learn how to manage it. And next month, I'll buy the bass drum and the cymbals. Oh, and maybe some noise-canceling headphones for you. Aw, you've thought of everything. Help kids save, budget, and build financial independence all with Chase. That's good for kids and good for parents. Visit your local Chase branch and get started. Accounts subject to approval. JPMorgan Chase Bank, N.A. Member FDIC.

0:30You can go to kitted, K-I-T-T-E-D, dot shop, and use the code SMART50, S-M-A-R-T-5-0, at checkout, and you will get half off a set of thinking superpowers in a box. If you want to know more about what I'm talking about, check it out, middle of the show.

Podcast Introduction

1:00Welcome to the You Are Not So Smart Podcast, episode 338. So when Alfred Sloan was CEO of GM, he closed a meeting by asking,

1:41are there any objections to my decision? There were no objections. So he said, well, then I propose that we postpone the decision until the next meeting, so that you have opportunities to come up with concerns.

Guest Introduction

1:54My name is David McRaney. This is the You Are Not So Smart Podcast. That was economist Alex Edmonds. And Alex Edmonds is a professor of finance at London Business School. He serves on the World Economic Forum. He is a fellow of the British Academy, the author of several books, and he once delivered a TED Talk that has been viewed more than two million times,

2:25titled, What to Trust in a Post-Trust World. He has a new book out titled, May Contain Lies. And in it, he outlines something called the Ladder of Mismference, which is why he is joining us on this episode. I wanted to ask him, Alex Edmonds, about this ladder. And I will do that. I will do that very thing in just a moment.

Inference Foundations

2:53Because first, I feel like this is a great opportunity to lay some psychological foundations, not only for his ladder of misinference, but for something I've wanted to talk about on this show forever, what this whole inference thing is all about.

3:13Inference in psychology and neuroscience and all the sciences that study the brain and mind refers to the guessing game played by your brain in the presence of every single thing you think, feel, perceive, and do. I think a few years ago, this would be more difficult to explain. But with AI being so intensely hyped these days and part of our lives everywhere we turn,

3:44the concept of a large language model predicting the most likely next word in a sentence at scale is a pretty good entry point to making sense of inference in the brain. Brains do this. They predict the most likely outcome, the most likely conclusion, the most likely result. All brains, from snakes to mountain lions to people. But unlike large language models, people can reason and love and experience emotions and contemplate meaning itself.

4:18But we do so with a lot of help from the portions of our brains that produce inference.

Inference Definition

4:29Okay, so what is this thing? Inference is a cognitive process by which, in the presence of sensory inputs and or mental models, draws upon prior experiences, existing knowledge, and contextual clues to generate assumptions and expectations and predictions about what has not yet happened or what has not yet been presented as evidence. We take all that and then we add our biases and goals and identities

5:04and concerns about our reputations and relationships and well-being and our desires and fears and traumas and all the rest to make a guess. Psychologically, that guess feels like knowledge, but it's not. It's just a guess.

5:25Neurologically, those guesses can come across as straight-up reality. For instance, there's a portion of your retina where the optic nerve exits on its way to the brain and that results in a blind spot in your vision. But as a seeing person, when you look around and don't see a blind spot in your vision, even though there definitely is one there, that's because your brain fills it in by inferring what ought to be in that missing portion of your visual field.

5:59It's doing a little bit of that in your periphery and it's doing a little bit of that all the time. And in that blind spot portion of your vision, you don't see what's there. You see what your brain thinks ought to be there. In a way, it's a guess based on context, based on experience, based on interpolation of visual data. But it's not real. And of course, nothing is real when you get down to it.

6:32When it comes to vision, it's all a simulation. But in this portion of the simulation, you're not seeing something based off of actual inputs from the electromagnetic spectrum hitting the back of your eyeball. It's all coming from inside like a dream.

Inference Examples

6:51In moments of intense ambiguity, the brain will do this sort of thing outside of the blind spot. We've discussed this a few times in the show via the example of the dress. It's the one that some people see as black and blue and others see as white and gold. What's happening there is in that famous image, in the photo, the photo makes the dress seem overexposed.

7:21When your brain assumes something is a bit overexposed, it will reduce that overexposure before you experience it in consciousness. If it's too blue, your resulting experience will be a little bit less blue. If it's too yellow, your resulting experience will be a little bit less yellow. With the dress, the nature of the overexposure is almost perfectly ambiguous. It's unclear whether it is exposed more so in natural light or more so in artificial light.

7:56Natural light tends to be a bit more blue. Artificial light tends to be a bit more yellow. So the more experiences you've had over your lifetime, in which you've seen objects overexposed in natural light, the more likely your brain will assume the dress is overexposed in natural light and thus will remove the blue tint, resulting in white and gold. The more experiences you've had over your lifetime, in which you've seen objects overexposed in artificial incandescent light,

8:27the more likely your brain will remove the yellow tint, resulting in black and blue. In both cases, in a moment of ambiguity, your brain disambiguates the ambiguous before it reaches your conscious experience by making a guess based on your prior experiences, or, as they say in psychology and neuroscience, your priors. An enormous amount of our day-to-day lives are built on disambiguation via our priors.

8:57If you hear a knock at the door, you will infer the origin of those sounds based on context, on expectations, on prior experiences. You will then infer what will happen if you take all manner of possible actions based on your inference. And then you'll produce a complex output of inferences, of inferences, of inferences, of inferences, all the way down. That, in the end, will not feel complex when you choose how to react.

9:29As you grow from baby to child to wherever you are now, all your billions of experiences, all the causes that regularly have led to effects, they all become a foundation of priors you use, mostly unconsciously, to generate the pattern recognition that you use, mostly unconsciously, to produce probabilistic predictions, expectations, and explanations, mostly unconsciously. Most of our cognition is inferential in this way. The world is too complex, too fast, too ambiguous for the brain to process it in real time.

10:07Instead, it constantly generates hypotheses, tests those against inputs, and then updates them so that we can make more and more useful assumptions.

10:19This all leads to a term in psychology that isn't really a term in psychology, but is sometimes used by psychologists and that I love, called the makes sense stopping rule. The makes sense stopping rule is a cognitive shortcut, a heuristic, by which your brain stops seeking more evidence, more sensory input, more data, more contemplation, once it reaches a conclusion, an inference, that it feels is plausible.

10:52And as a social primate, plausible often means that which you feel other people will consider to be reasonable, in the sense that if you have to defend yourself, it will be supported by good reasons. Not that any of that is usually conscious, but it is how the processing seems to flow. The problem is, thanks to the makes sense stopping rule, you will stop seeking more information when something makes sense to you, regardless of whether it's actually true.

11:23And that can lead to a false sense of confidence in your decisions, and or a false sense of understanding when it comes to a topic or a situation. The more formal term for all these mental phenomena coming together in sort of a makes sense and stop looking for more stuff kind of way is satisficing. When it comes to satisficing, we tend to take on a thinking strategy that aims for a satisfactory outcome instead of a perfectly optimal one.

11:56Whether shopping for socks or making plans for dinner or making plans for your future career or home or relationship, we tend to commit to a decision when we meet our individual threshold of acceptability. The same is true of conclusions. The same is true of interpretations of news stories. The same is true of deciding whether or not we're going to share a bit of content in our information streams.

12:28And this is most likely because for most of our evolutionary history, we had neither the time, resources, nor information necessary to take an exhaustive, extra-super-rational, unbelievably scientifically sound path toward plausibility. And so, today, we are biased toward speed, not accuracy. So yeah, many systems, biological in nature,

12:58combine efforts inside your brain to produce what we might call the architecture of assumption. Its output is not perfect, and it doesn't aim to be perfect. It aims to be good enough. And most of the time, it is. Most of our inferences are good enough. But here's the thing. When they aren't good enough, we often don't know they're not good enough. And not only do we not know they're not good enough, we believe, we infer that they are good enough.

13:32As Mark Twain once said, it ain't what you don't know that gets you into trouble. It's what you know for sure that just ain't so. Which is a quote I used to use many times and cite Mark Twain as the origin, but when I fact-checked it, it turned out that, yeah, he never said that. It just made sense to me that he did, and then I shared it, believing he had.

14:15After the break, we will discuss how this all contributes to the ladder of misinference and what we can do to avoid the mistakes it produces with economist Alex Edmonds, author of May Contain Lies, a book about that ladder and all manner of ways we get fooled by misinformation.

14:48Hey, Mom, now that I have an after-school job, I think I want to buy a drum kit. Okay, Rockstar, let's look at your account. When kids start making money, Chase has easy-to-use tools and expert bankers to help them learn how to manage it. And next month, I'll buy the bass drum and the cymbals. Oh, and maybe some noise-canceling headphones for you. Aw, you've thought of everything. Help kids save budget and build financial independence all with Chase. That's good for kids and good for parents.

15:18Visit your local Chase branch and get started. Account subject to approval. JPMorgan Chase Bank and a member FDIC. You know, one of the nice things about Hertz right now is that they have a whole fleet of new cars. So whatever trip you're planning, it's easy to find a new car that fits your adventure. Heading out for a sunny drive, then a new convertible with very little roof is for you. Considering a fishing trip with your mini-me, then you need a new minivan that's anything but mini. And if you're planning on some stargazing, a new SUV will give you plenty of space.

15:49Get paired with our newest fleet yet at Hertz.com.

15:54The School of Thought. I love this place. I've been a fan of the School of Thought for years. It's a non-profit organization. They provide free creative commons, critical thinking resources to more than 30 million people worldwide. And their mission is to help popularize critical thinking, reason, media literacy, scientific literacy, and a desire to understand things deeply via intellectual humility. So you can see why I would totally be into something like this. The founders of the School of Thought have just launched something new

16:26called Kitted Thinking Tools. K-I-T-T-E-D Thinking Tools. And the way this works is you go to the website, you pick out the kit that you want, and there's tons of them, and the School of Thought will send you a kit of very nice, beautifully designed, well-curated, high-quality, each one about double the size of a playing card, matte cello 400 GSM stock prompt cards, and a nice magnetically latching box that you can use to facilitate workshops, level up brainstorming

16:56and creative thinking sessions, optimize user and customer experience and design, elevate strategic planning and decision making, mitigate risks and liabilities, and much, much more. And each kit can, if you want to use it this way, interact with this crazy cool app. Each card has a corresponding digital version with examples and templates and videos and step-by-step instructions and more. You even get PowerPoint and Keynote templates. There's so many ways you could use this. Here's some ideas.

17:27If you're a venture capital investor, you could get the Investor's Critical Thinking Kit and use it to stress test and evaluate different startups for Series A funding. If you're a user experience designer, you can get the User Design Kit to put together a workshop with internal stockholders for a software product. Or if you're an HR professional, you could mix and match these kits to create a complete professional development learning program tailored specifically for your team over the course of the next two years. So, if you're the kind of person who is fascinated with critical thinking

17:58and motivated reasoning and intellectual humility and biases, fallacies, and heuristics, you know, the sort of person who listens to podcasts like you are not so smart, you're probably the kind of person who would love these decks. If you're curious, you can get a special 50% off offer. That's right. Half off offer right here. You can get half off of one of these kits by heading to kitted.shop. K-I-T-T-E-D dot shop and using the code SMART50 at checkout.

18:30That's SMART50 at checkout. 5% of the profits will go back to the school of thought so you're supporting a good cause that distributes free critical thinking tools all over the world on top of receiving a set of thinking superpowers in a box. Check all of this out at kitted.shop or just click the link in the show notes.

18:53And now we return to our program.

Interview with Alex Edmonds

18:56I'm David McCraney. This is the You Are Not So Smart podcast. Our guest in this episode is Alex Edmonds, a professor of finance at London Business School. In his book May Contain Lies, he examines how narratives, statistics, and studies can mislead us if we're not really aware of how inference works and where we are on a ladder of potential misinference.

19:26especially when we align our preconceived notions with the thing that we are scrutinizing or most likely not scrutinizing if it confirms our preconceived notions, our biases, our priors, and what we would like to be true, what we wish was true, or what just seems to make sense. All right. Here is my interview with Alex Edmonds.

19:56Let's just start from like first principles here. What is confirmation bias, Alex? As if it's the first time anyone has ever heard this term before, what are we talking about? It's the temptation to accept a result uncritically if it confirms what we want to be true and to reject a result out of hand if we don't like what it says. So that's what I'd call biased interpretation. So that's one part of confirmation bias, which is that we respond to information

20:26we receive based on whether we like it. But there's also a second part and therefore it's important that you ask for the definition. The second part is biased search. So the information that we look for to begin with is information that we'll like the sound of. So if we're on the right wing, we might only look at Fox. If we're on the left wing, we might look more at MSNBC. So it's not only about the interpretation of information, but the search for information. Okay, there's a couple things in there I want to pull out briefly. One is the term like and want

20:56and agree with. Like if we come across information that we and it confirms something we want to be true, that word want is humongous to me and I want to get a little bit deeper in there. What do you mean by want and what is it that's fueling and motivating and driving this want it to be true statement? Yeah, so this is important. So want is there could be you want it to be true because there are clear tangible benefits from this.

21:27So if I'm a proponent of sustainable investing because I've written books about it, I'm an advisor to sustainable investing firms, I would like results claiming that sustainable investing pays off. But also this idea of want can be much, much weaker and much more subtle. we might want something to be true just because it confirms our worldview. And so there might not be anything in it for me, but if it confirms our worldview, we might accept this. And so this is really important because we might think confirmation bias only applies

21:58to huge things where there's a lot of skin in the game or if it's ideological like your views on immigration or abortion. But even something small, such as we think that something natural is better than something artificial, that is sometimes enough to trigger us to be exhibiting this bias. You use these two terms like you talk about naive acceptance of things, which is a really good way to put it. I love that. But also this blinkered or as I've often heard selective skepticism. So I'm looking over things and I see things, okay, well, I'd really rather

22:29that not be. And this other side of the coin, I love how this comes into play. It could be something where you're actively searching, you're Googling up, you're Wikipedia-ing, but it also could just arrive, it could just land in your lap, you open the newspaper metaphorically because that rarely happens anymore, but information arrives somehow and where there are times and you're like, that sounds true to me, and there are other times you're like, hmm, and all of a sudden you become

22:59skeptical and it's selective or blinkered as you put it. Let me hear a little bit more about that. Yeah, so one example is the Deepwater Horizon disaster. So there, there was an inconvenient truth, which was the rig was not safe and so you could not remove the rig without there being a potential explosion. They did a test known as the negative pressure test. They did the test three times. Every time it failed, they just didn't want that to be true. So they thought of another escalation to explain why the

23:30test was not reliable in that circumstance. they invented a different test that they ran instead that different tests passed and so this led to the disaster. Another example could be Silicon Valley Bank. So their own models predicted that because they'd gone so into treasury bonds, then they would have huge losses if interest rates rose. They didn't want that to be true. They, they, they enacted blinkered skepticism and they said, well, let's try to come up with

24:01a different model to give a better answer that our bank was not at risk. So this is dangerous because you might think, well, these biases, only dumb people should fall for it. I'm a smart listener to your podcast. I will never fall for this misinformation. But actually, some evidence suggests that smarter people are more likely to fall for misinformation because the smarter you are, the more able you are to engage in what's known as motivated reasoning. You come up with reasons to excuse why a particular

24:31piece of evidence should be dismissed. The, yeah, the, the evidence is pretty clear. And I guess there's many meta levels to our conversation, considering that we are discussing evidence and how it may or may not confirm our hypothesis. But the, the evidence so far is pretty strong that the smarter you are and the more educated you become, the, it's just, you're just become better at justifying and rationalizing, which leads to the bizarre downstream effects of what you're discussing. So I love that you point this out too, very early on in the book.

25:02I think this is a pretty good introduction to confirmation bias. It reminds me of Dan Gilbert, uh, had this, when he was trying to make the shortest, uh, possible metaphor, uh, you step on a scale of, if it, if you like what it says, you just go about your day. If you, uh, if you step on a scale and you don't like what it says, you step on and off about five more times. That's a, it's a great way to put it. And also I love that you mentioned in the book, the, uh, the research, um, the scientist who did the fMRI and, uh, had

25:34the strong, oh no, I'm getting attacked by a bear feeling when their amygdala lit up in the presence of information they didn't like. Um, I got to interview them right after they did that research. And at the time they had no idea why they, they, they had, they were, they were just like, look, I have this data and I don't know. I just, all I can tell you is I have this data. Okay.

25:57So we have confirmation bias. These are, these are the big two and you open the book with the big two. The other is black and white thinking. You really get into it and demonstrate how dangerous and difficult this can be for people who think they know what we're talking about or have never heard this before. What are we talking about when it comes to black and white thinking? So this is the idea that something is either always good or always bad. There's no shades of gray.

26:28For example, this was how Atkins approached the Atkins diet. He said carbs are always bad. And this is black and white in many ways. Number one, it's black and white in that it's always bad. There's never any moderation. So it's not that carbs are okay as long as they're 20% of your daily calories. No carbs have as few as possible, but it's also black and white in that it ban, it bunches all types of carbs under the same umbrella. It doesn't matter whether it's refined sugar or simple carbs or complex carbs. Anything which is called carbs is bad.

26:59And so why is this so appealing? Well, it's just easy. And we like simple shortcuts and heuristics. It makes our complex world simple. And therefore, anything which gives us simple advice like eat as few carbs as possible, that's something that can catch on because it accords with our biases. I dig how you describe it. You describe it as, you know, why would we go this way? Because as you write, we like things that can be learned and applied quickly. This is the reason we have so many of these heuristics and a speedy good enough thing is often better than a complex

27:30nuanced. I have to really think about this to achieve something approaching total accuracy. But you add something to that that I had never seen before. And I think it's so easy to just say the world is made of shades of gray and there's a gradient to things that it's a bit more complicated than that and almost no matter what it is you're discussing, it's always more complicated because you could always go into more and more rich detail. But you add that before you hand wave

28:00with the phrase shades of gray, consider the concepts of moderate, granular, and marbled. I would love to hear you talk more about those three things. Yeah, so these are three reasons for why something might be shades of gray. So I want to be precise as to why the world is not black and white. So the first is moderation, which is it's rarely the case that something is always bad or always good. It might be good only up to a point or bad only after a point. So one example is drinking water. Particularly if you're running a marathon,

28:31water drinking seems to be a really good thing. You need to hydrate, but you can suffer from water intoxication. And sadly, this happens sometimes to runners where they're told, hydrate as much as possible. They follow this advice dutifully. They hydrate a lot. And this leads to sometimes death because there's so much water that it dilutes sodium and other essential minerals to tiny concentrations. So that's the idea of moderation. So something is not always in one direction. It might start off positive and then turn negative like the

29:01effect of water on marathon performance. But the second idea is granular. So this means that there's not just one big bucket. Within that bucket, some things might be good and other things might be bad. So within carbs, yes, it may well be the case that simple carbs are bad, but complex carbs might be good for you. Similarly, you learn at school that cholesterol is generally bad for you, but there's good cholesterol, which is high density lipoprotein. And so rather than the simple idea that's avoid as much cholesterol as possible, there are some

29:33things that are good and some that are bad. Finally, the idea of marbled is that certain things might be neither unambiguously good or bad. So why I call it marbled is if you think about some marbled meat, it's got streaks of fat intertwined with it, so you don't know whether you're getting fat or whether you're getting muscle. And so my field of sustainable investing is the fossil fuel industry necessarily bad. Yes, climate change is a really important threat,

30:03but sadly, in many countries, we don't yet have enough renewable energy to get by without fossil fuels right now. In Africa, 600 million people don't have any access to electricity, so the idea that we should deprive them of fossil fuels and the potential for economic development is something which would be a drastic outcome. So even cases in which we think that something might have no redeeming qualities, it might still have them, and so there's much more nuance needed rather than never invest in fossil fuels, never

30:34hire somebody who's ever been in prison for your company, we can really rehabilitate people who've been in prison, so it's just to encourage more nuanced thinking.

30:45I had a guest on the show, Laurie Santos, who coined the G.I. Joe fallacy, which was knowing that knowing is not half the battle is half the battle. It was, you do have to recognize the half part of that, so the knowledge is not enough. What are we going to do with it? What are good ways to apply the fact that we have learned there are more than 200 cognitive biases? So, and most of them are just sort of variations or deeper explorations of

31:16confirmation and motivated reasoning and black and white thinking. So, you have this beautiful thing, and I love it very much. I'm going to share this with the world starting with this podcast, but I will cite you endlessly and show this to people every time I get a chance to talk about these topics going forward. I love your ladder of misinference. I could describe it and then have you tell me little things about it, but that would take a valuable opportunities for you to actually just get going. I will just say it's, you have these rungs, it starts with statement, goes up to fact, data, evidence,

31:47and proof. What is this thing and how can we use this? I have other questions to follow up, but you can just get going.

Ladder of Misinference

31:54Thanks.

Ladder of Misinference

31:54So, after highlighting the biases, I wanted to look at, well, how to address them. So, how to correctly interpret information. So, what I wanted to do was to categorise all of the types of misinformation out there into just four buckets for easy application. And so, these four buckets are four steps on the ladder of misinference, which highlights the four types of mistakes that we might make. So, the first rung of the ladder is the difference between statements and facts. And I highlight that a statement is not fact, because

32:27it may not be accurate. So, what do I mean by this? So, it may well be that you hear a famous quote from somebody, but that quote might have never been said by somebody. It might not be giving the full context, or it might be a misquote. So, we often think, well, if there's a footnote at the end of a sentence, and there's a paper cited in the footnote, this must mean that the sentence is gospel. But it's not. We need to check the facts. Now, the second step of the ladder is a four-

32:57fact is not data. It may not be representative. So, let's say something is absolutely true. We've checked it. It still may be misleading because it could be a hand-picked example. So, then you might think, well, it's not the best solution to that, having hundreds of data points, so that you don't have one isolated example. But then that's the third step of the ladder, which is data is not evidence. It may not be conclusive. So, what's the difference between data and evidence? So, what is evidence to

33:28begin with? Evidence is something which points to one interpretation of the data and not the others. Even if you have a robust correlation with tons of data points, it might not be causation. There could be other things going on. The final step of the ladder is evidence is not proof. It may not be universal. So, evidence, even if it's watertight, only applies to one situation. I'm wondering, how do we

33:58apply the knowledge of a statement's not a fact, a fact is not data, data is not evidence, evidence is not proof. If I'm out there Googling, what are some simple tips for applying this ladder to what I'm searching for? So, first, it's to see what step on the ladder you're on, and then ask questions to make sure that you're not making a misstep to the next rung.