Our Noisy Minds

Psychologist Daniel Kahneman says there are invisible factors that distort our judgment. He calls these factors “noise.” The consequences can be found in everything from marriage proposals to medical diagnoses and prison sentences. This week on Hidden Brain, we consider how to identify noise in the world, and in our own lives.

If you like our work, please consider supporting it! See how you can help at support.hiddenbrain.org. And to learn more about human behavior and ideas that can improve your life, subscribe to our newsletter at news.hiddenbrain.org.

Additional Resources

Books:  

Noise: A Flaw in Human Judgment, Daniel Kahneman, Olivier Sibony and Cass Sunstein, Little, Brown Spark, 2021

The Wisdom of Crowds, James Surowiecki,  Anchor; Reprint edition, 2005

Research Studies:

Refugee Roulette: Disparities in Asylum Adjudication” Jaya Ramji-Nogales, Andrew I. Schoenholtz & Philip G. Schrag, Stanford Law Review, April 2020 

“Clouds Make Nerds Look Good: Field Evidence of the Impact of Incidental Factors on Decision Making” Simonsohn, Uri, Clouds Make Nerds Look Good: Field Evidence of the Impact of Incidental Factors on Decision Making, SSRN, 2006 

“Algorithm Aversion”Berkeley Jay Deitvorst, Publicly Accessible Penn Dissertations, 2016

Measuring the Crowd Within: Probabilistic Representations Within Individuals” Edward Vul, Harold PashlerSage Journals July 2008

“Human Decisions and Machine Predictions”, Jon Kleinberg , Himabindu Lakkaraju , Jure LeskovecJens Ludwig , Sendhil MullainathanQ J Econ.,  2018

The transcript below may be for an earlier version of this episode. Our transcripts are provided by various partners and may contain errors or deviate slightly from the audio.

Shankar Vedantam: This is Hidden Brain. I'm Shankar Vedantam. We're going to start today with a little experiment. I'll be the guinea pig. I'm going to open the stopwatch app on my phone. I'll hit start and count off five seconds while looking at the phone. One, two, three, four, five. Okay, let me do that again. One, two, three, four, five. Okay, now I'm going to hit start and count off five seconds without looking at the phone. One, two, three, four, five. It was 5.43 seconds. Let's do it again. One, two, three, four, five. Much better, 5.2 seconds. Last time. One, two, three, four, five. 5.59 seconds. The errors I made seem trivial, but it turns out they are not. Multiply the small mistakes I made in milliseconds over all the countless decisions I make every day, and you can end up with a serious problem. Multiply the errors I make as an individual by an entire society made up of other error-prone humans, and you can get disaster. What makes these mistakes insidious is that they are rarely the result of conscious decision making. Human judgment is imprecise, and imprecise judgment produces unwanted variability, what a Nobel Prize-winning psychologist Daniel Kahneman calls, "Noise."

Daniel Kahneman: Wherever there is judgment there is noise, and there is more of it than you think.

Shankar Vedantam: This week on Hidden Brain, the gigantic effect of inadvertent mistakes in business, medicine and the criminal justice system, and how we can save us from ourselves.

Shankar Vedantam: Daniel Kahneman's insights into how we think have revolutionized many areas of the social sciences. He was my guest on Hidden Brain for our 100th episode. We talked about his early research and his first book, "Thinking Fast and Slow." As we close our 200th episode, we wanted to bring him back to talk about a set of ideas he's been working on for several years. They're described in his new book, "Noise: A Flaw in Human Judgment." Daniel Kahneman, welcome to Hidden Brain.

Daniel Kahneman: Glad to be here.

Shankar Vedantam: I want to begin my exploring what you mean by the term "Noise". You spent some time studying an insurance company, and one of the things that an insurance company needs to do is to tell prospective clients how much their premiums are going to cost. An underwriter says "If you want us to cover you against this loss, here's this quote." From the insurance company's point of view, Danny, what is the risk of offering quotes that are too high, and also quotes that are too low?

Daniel Kahneman: A quote that is too high, you are very likely to lose the business because they are competitors and they'll offer a better price. A quote that is too low, you're leaving money on the table and you may not be covering your losses if you do that a great deal. Errors in both directions are costly. We define noise as unwanted variability in judgmental decisions, that is if the same client would get different quotes from different underwriters in the same company, this is bad for the company. Variability is a basic component of error.

Shankar Vedantam: I think of the insurance business as being driven by mathematics. That's my stereotype, that they are hard-nosed statisticians who work at these companies. I would not expect a quote from one underwriter to be widely different from the next. You asked executives of this insurance how much variability they expected between underwriters. What was their estimate of this kind of subjective variability?

Daniel Kahneman: It turns out that there is a very general answer to that question, and people have a very general idea about what that will be and it's around 10%. When we actually measure that in an insurance company, the answer was 55%. That was a number that was an amount of variability, as we call it "an amount of noise" that no one expected. That really is what set me off on this journey that led to this book.

Shankar Vedantam: The difference between 10% and 55% might seem trivial. Who cares? Well, the consequences of this variability were anything but trivial.

Daniel Kahneman: I asked people what actually would be the cost of setting up a premium that is too high or too low. When they carried out that exercise they thought that the overall cost of these mistakes was within the billions of dollars. What was in some sense saving that company was that probably other companies were noisy as well. But if you have a company that is noisy while others are noise-free, the noisy company is going to lose a lot of money very quickly.

Shankar Vedantam: With the insurance company, it's not just that the insurance company is losing money, there is also a cost that's being paid by all the people who are trying to get insurance. It might be that if you happen to get a quote that's too high you might end up being uninsured, or you might be spending more on insurance that you need to be spending. There is sort of a general human cost to these errors, not just in terms of the bottom line for the insurance company.

Daniel Kahneman: Of course, when you have a noisy underwriting system, then the customer is facing a lottery that the customer has not signed up for. That is true everywhere, that is wherever people reach a judgment or a decision by using their minds rather than computing. Wherever there is judgment there is noise, and there is more of it than you think.

Shankar Vedantam: I want to look at a few other places because in some ways what's striking about your book is both a number of different domains where you see noise, and the extent of noise in those different domains, including in places where you really feel this should be a setting where noise does not play a role. You cite a study done by Jaya Ramji-Nogales and her coauthors who found that in asylum cases, this was a courtroom in Miami, one judge would grant asylum to 88% of the applicants, and another granted asylum to only 5% of the applicants. This is more than a lottery. This is like playing Roulette.

Daniel Kahneman: This is a scandal. Clearly, the system isn't operating well. In many situations, it's just that when people look at the same data, they see them differently. They see them more differently than they expect to see them more differently than anyone would expect. That's the basic phenomenon of what we call "System Noise" that is, when you have a system that ought to be producing judgements or decisions that are predictable, they turn out not to be predictable and that's noise.

Shankar Vedantam: You also describe in some ways there are different kinds of noise. If you're an asylum judge and I'm an asylum judge, and we have very different subjective readings, that can produce very different answers. It goes to be that if you are reviewing a case in the morning, and you are reviewing a case in the afternoon, it's possible that just within yourself your own judgments can be noisy. Can you talk about that idea as well?

Daniel Kahneman: It's not only possible, it actually is the case that when people are asked the same question or evaluate the same thing on multiple occasions, they do not reap the same answers. For example, radiologists were shown the same image on two separate occasions, and are not reminded that it's the same image, really with distressing frequency reach different diagnoses on the two occasions, that we know. It's true even for fingerprint examiners, whom we really would not expect to be noisy at all, but actually they vary when you show them the same fingerprints twice. By the way, that's important. They do not vary in the sense that somebody would make a match on one occasion and would positively say it is not a match on the other, but fingerprint examiners are allowed to say "I'm not sure." Between "I am not sure" and "I am sure", that it's a match or it's not a match, there is variability.

Shankar Vedantam: One of the things that you point out is that you don't expect that the lottery of who is reviewing your file is going to make a huge difference, or that extraneous factors would play a huge role. The researcher, Uri Simonsohn found that college admissions officers pay more attention to the academic attributes of candidates on cloudy days, and the non-academic attributes when the weather is sunny. He titled his paper "Clouds Make Nerds Look Good." Talk about this idea that extraneous factors, whether someone's hungry and what the weather is like, that can affect people's judgment too.

Daniel Kahneman: Indeed. It's been established in the justice system. If you're a defendant you have to hope for good weather because on very hot days, judges assign more severe sentences. That is true, although judges have air conditioning, but it's the outside temperature nevertheless seems to have an effect. It's been established in at least one study that for judges who are keen on football, the result of their team Sunday or Saturday depending on whether it's professional or college, will affect their judgment they make on a Monday. They will be more severe if their team lost.

Shankar Vedantam: That's a terrifying idea, isn't it Danny, that you're sort of hoping that your judge's football team wins the Sunday before your case is heard.

Daniel Kahneman: Yes, absolutely. You're also hoping to find a judge who is in a good mood, to find a judge who is rested and has a good night, who is not too tired. Your chances of being prescribed antibiotics or painkillers differ in the course of the day, so doctors tend to prescribe more antibiotics towards the end of the day when they are tired than earlier in the day when they are fresh and they're more likely to prescribe pain killers later in the day simply because it's an effort to resist the patient who wants pain killers and when you're very tired and depleted that effort becomes more difficult. So, completely extraneous factors have a distressingly large effect.

Shankar Vedantam: Noise in medicine often shows up under a different name, "Medical Mistakes". Danny, can you talk about these two different dimensions of noise in the medical sphere, the ways in which it might cause us to get diagnosed with conditions we might not have, but also for doctors to miss conditions and problems that we actually do have?

Daniel Kahneman: The contribution of noise, that which physician looks at the data, makes a difference. There is a lot of that, that as we know, that physicians disagree on diagnosis and they also disagree on treatment. That is a little shocking that there is that element of lottery. So, errors could happen for many reasons, including luck, which is not an error in judgment but where information was missing. In some cases, the errors cannot be described in any other way than noise that has different doctors looking at the same case reaching different conclusions.

Shankar Vedantam: It might seem obvious from these examples that noise is a big problem, and that combating noise makes a lot of sense. Who could argue against reducing arbitrary decisions and inconsistent rules? It turns out a lot of people have a problem with doing just that, and one of those people might be you. You're listening to Hidden Brain. I'm Shankar Vedantam.

Shankar Vedantam: This is Hidden Brain. I'm Shankar Vedantam. We've seen how noise pervades many aspects of our personal and social lives. It can lead to widely different estimates on our insurance premiums. It affects judgements doctors make about our health. It can determine whether we get a job or a promotion. In their new book, "Noise: A Flaw in Human Judgment," Daniel Kahneman and his coauthors Olivier Sibony and Cass Sunstein, show that noise also shapes what happens in the criminal justice system. It affects decisions that send people to prison, or sentence them to execution.

Shankar Vedantam: Danny, Judge Marvin Frankel worked as a United States District Judge, and he made a name for himself by pointing out inconsistencies in the criminal justice system. He once heard a case about two men convicted for cashing counterfeit checks, both amounts were for less than $60.00. One man got a sentence of 30 days in prison. The other got 15 years. What did Judge Frankel make of such disparities?

Daniel Kahneman: He thought "I'm just." He thought that it's extraordinarily unfair, which it seems to be on the face of it. He really felt that the justice system should be reformed to avoid this role of completely unpredictable unreasonable factors that determine the fate of defendants.

Shankar Vedantam: You know, Danny, I feel like in the last year I've seen dozens of stories that talk about disparities of all kinds, including disparities in the criminal justice system. Invariably when I read these stories about disparities, they talk about the idea that it's about bias, that it's about racial bias or gender bias, or some other kind of bias. When Judge Frankel comes along and says defendants are being given vastly different sentences, the very first thing that pops in my head is maybe these defendants were of different races and what we're really seeing is racial bias at play rather than noise. How can we tell the difference between racial bias and noise?

Daniel Kahneman: It's actually easy to do, because when you want to measure noise you can conduct a kind of study that we call a "Noise Audit". You take professionals, for example judges, and you show them a fictitious case. And you ask them to make judgments as they would normally. You know that it's the same case. They've all been given the same information. They should give you the same judgment. The differences among them cannot be attributed to bias. Indeed, what Judge Frankel caused to happen, he called many noise audits to be performed. He actually conducted some himself. In the most famous one, 208 federal judges evaluated 16 cases and assigned sentences to 16 cases. This gives you an idea of the lottery that the defendant would face in that where the average sentence is seven years in jail, the probable difference between two judgements is over three years. So, that seems to be unacceptable.

Shankar Vedantam: Based on the work of Judge Frankel and others, congress eventually passed a law that basically limited the amount of discretion that judges had. Talk about the effects that this law had on reducing noise. Were there studies conducted to actually figure out if these were reducing noise?

Daniel Kahneman: Yes. Studies were conducted, and actually you can look at many cases and look at the variable of judgements in many cases, and you find that the variable would be significantly diminished, which indicates that the noise was in fact reduced. However, something else happened. The judges hated it. They hated this restriction on their ability to make free decisions, and they felt that justice was not being served.

Shankar Vedantam: Even as the data was showing that the noise was reducing in sentencing, in other words, sentencing was becoming more consistent, many judges were upset that their discretion was being taken away. Judge Jose Cabranes was one of those who spoke up. I want to play you a clip of something he said in 1994. This was a discussion at Harvard University where they talked about these guidelines that were aimed to reduce ethnic disparities and sentencing by limiting the amount of discretion that judges had. Here is Judge Cabranes:

Judge Cabranes: These arcane and mechanistic computations are intended to produce a form of scientific precision, but in practice they generate a dense fog of confusion that undermines the legitimacy of the judge's sentencing decisions.

Shankar Vedantam: Danny, I want to draw your attention to what Judge Cabranes is saying when you limit the variability of sentencing, you're telling judges "For this offense you have to X, for that offense you have to do Y." A lot of judges feel their hands are tied and they feel the art of law is being reduced to a mechanistic science.

Daniel Kahneman: If it takes a mechanistic science to produce justice, then I think we should seriously consider some mechanistic science. What seems to be happening is that from the perspective of the judge, they feel that they're evaluating every detail of the case, and that they are producing a just judgment because they're convinced that what they're doing is a just judgment. Somehow, it's very difficult to convince judges that another judge whom they respect a great deal when presented with the same case would actually pass a different sentence, that argument doesn't seem to have penetrated when Judge Cabranes made that dissolution. That, in fact there is a problem and there is a problem to be resolved. He was in effect, as I hear him, he was denying the existence of the problem.

Shankar Vedantam: Psychologists talk about a phenomenon called "Naive Realism", that in some ways explains why it is I am bewildered that you would not see the world exactly the way that I see the world. Can you explain what naïve realism is, and how it speaks to the question we just discussed about judges not just reaching different conclusions, but being bewildered that anyone would reach a different conclusion than them.

Daniel Kahneman: We feel that we see the world as it is. It's the only way we see it, and what we see is real. What we see is true. It makes it very difficult to believe and to imagine that someone else looking at the same reality is going to see it differently, but in fact we're struck by how different they are in the context of criminal justice, the variable of the sentences is shocking. But when you are looking at it from the perspective of a judge who looks at cases individually, and feels that he or she is making correct judgements for every case individually, then it looks as if any attempt to restrict their freedom is going to cause injustice to be performed. But they are simply not accepting I think the statistics that tell them that another judge looking at the same case would actually pass a different sentence.

Shankar Vedantam: These debates about sentencing reform raged in the 1980s and 1990s, and eventually in the early 2000s the Supreme Court struck down the guidelines that bound the way judges were operating, and sentencing reform essentially went away, giving discretion back to judges. Is what happened what I fear happened? Did noise come back into the system?

Daniel Kahneman: Oh, yes. There is evidence that noise came roaring back, and there is also evidence that judges were a lot happier without the guidelines than they had been earlier.

Shankar Vedantam: One of the ironic things that you and others have found is that even though there is this distinction between noise and bias, when the noise came back after the Supreme Court ruling, black defendants were actually among those who were the most severely harmed by this. Is it possible in some ways there can be intersections between noise and bias? In other words, they can amplify one another?

Daniel Kahneman: Certainly. When you are constraining people and reducing noise, you're reducing the opportunities for bias to take place. Attempts to reduce noise and attempts to control noise are going to in general, not imperially, but are very likely to control and reduce bias as well.

Shankar Vedantam: If noise produces many of the adverse outcomes we see, if noise produces much of the unfairness we see, why is it that critiques of disparities invariably talk about bias? Turns out, that's because of the way our minds work. As we discussed in a recent series of episodes, the brain is a storytelling machine. A story of bias caters to our hunger for simple explanations.

Daniel Kahneman: Clearly, bias in general is a better story. As you see something happening, it had the character of an event, it had the character of something that is caused by psychological force of some kind, variability noise is uncause. Noise doesn't lend itself to a causal story. Really, the mind is hungry for causes. That leads us very naturally to think in terms of biases, that errors must be explainable.

Shankar Vedantam: If I get a misdiagnosis because a doctor doesn't like the color of my skin, that might not make me feel good, but at least I can make sense of what happened. Once I settle on an explanation of racism, or sexism, or homophobia, I tell myself I have every right to get angry. When I discuss what happened with others, they'll get angry too. By contrast, a misdiagnosis produced by noise is by definition no one's fault. The error may have harmed me, but I can't lay the blame on someone's evil intentions. Noise is the very opposite of a good story. It's meaningless, and that can make me feel even worse. Here's another problem, when I see a judge pass a really harsh sentence or a very light sentence, I can come up with a story of bias to explain this individual case. You cannot do that with noise. You cannot spot noise by looking at any individual case. You have to measure it in the aggregate. It shows up only when you look at the statistics, and many of us are uncomfortable turning to data as our guide to the truth. We prefer stories and anecdotes, and stories and anecdotes are better at illustrating the problem of bias.

Daniel Kahneman: Stories and anecdotes are what the mind is prepared for. Statistical thinking is alien to us, and statistical thinking is the only way to detect noise because it's variability. It's sort of absurd to say about any single case that it is noisy. You say that if you have no idea of how it came about, but noise is a phenomenon that you observe statistically, and that you can analyze only statistically. That is not appealing.

Shankar Vedantam: There's an even deeper problem in the fact that noise is detectable only through statistics, whereas bias you can tell a story about bias. For many people making decisions, the data is simply not even available. So, at a statistical level you can see an insurance company is demonstrating noise, but many of the decisions we are making are decisions we make as individuals. If I want to propose marriage, and I feel like proposing marriage on a moonlit night in the springtime, I have no idea if my decision to propose marriage on that evening is being shaped by noise or not. I don't have a statistical set of how I would behave under different circumstances.

Daniel Kahneman: The truth of the matter is that no one can tell you that this decision was noisy. What you can tell is that when you look at the collection of decisions of people deciding to get married, that collection is noisy. There is no reason to believe that these steps which improve judgements in the statistical case do not apply when somebody decides to get married. If noise is present in the decisions where you can observe it, it's also present when you cannot observe it.

Shankar Vedantam: Some years ago I interviewed the researcher Berkeley Dietvorst. He talked about how people respond when a mistake has been made by a human versus an algorithm. I want to play you a short excerpt of something he told me.

Berkeley Dietvorst:

People failed to use the algorithm after they'd seen the algorithm perform and make mistakes, even though they typically saw the algorithm outperform a human. In our studies, the algorithms outperform people by 25-90%.

Shankar Vedantam: He's basically saying the algorithms are significantly better than the humans, but when a mistake is made, and algorithms of course can make mistakes and humans can make mistakes, he's saying that you prefer the human to make the mistake. I think intuitively, that feels correct to me. If I'm going to get a misdiagnosis when I go to a doctor, I would feel better if it's the doctor who has made the mistake rather than an unfeeling, unthinking algorithm.

Daniel Kahneman: I think that's absolutely true. When we're looking at a road accident, we somehow feel less bad about it if it was a driver error than if it was a self-driving car that caused the accident. Algorithms, they make errors. The error they make, by the way, are different from the errors that people would make. They look stupid to people. Algorithms make errors that people think are ridiculous. We don't get to hear what algorithms think of the errors that people make. We do know that algorithms just make far fewer of them in many cases, and you have to trade off the higher overall accuracy against the discomfort of abandoning human judgment and trusting an algorithm.

Shankar Vedantam: Yeah. This might actually be a subtext of much of your lifetime's work, Danny, but it seems to me that fighting noise requires a certain amount of humility. It seems to me that humans are not humble.

Daniel Kahneman: They're not humble for a fairly straightforward reason. We do not go through life imagining different ways of seeing what we see. We see one thing at a time, and it feels right to us. That is really the source of the problem of ignoring noise. This is why it is so difficult to imagine it.

Shankar Vedantam: I want to talk just for a brief moment about places where noise can potentially be useful. Let's say for example you have a company that's trying to innovate and come up with new ideas, or you're in a creative enterprise where you want to pitch different ideas for movies. In some ways you might want to actually maximize the variability of the ideas you get. Noise is not always bad. Sometimes it can actually lead to good things.

Daniel Kahneman: Yeah, we don't call it noise in those cases. We reserve the term "Noise" for undesirable variability. There are indeed many situations in life in which variability is a blessing. Certainly in creative enterprises, also evolution. Anything that allows you to select the better one of multiple responses, wherever there is a selection mechanism, variability is a good thing. But variability in the absence of the selection mechanism is a sheer loss of accuracy. Those are the cases that we talk about. If you had a way when you have multiple underwriters of finding out who is doing a better job than whom, and using that in order to improve their training, that would be a case where you could make positive use of variability. In the absence of such a mechanism, that variability just is a sheer loss.

Shankar Vedantam: When we come back, how to fight noise. You're listening to Hidden Brain. I'm Shankar Vedantam.

Shankar Vedantam: This is Hidden Brain. I'm Shankar Vedantam. Noise is endemic. It's also very difficult to fight in part because judges and doctors, and police officers don't like to think of themselves as capricious. We don't think of our judgements as being arbitrary, certainly not when it comes to really important decisions. Even when we are told about how noise is affecting our judgements and decisions, we hate to be shackled by rules. Danny, in 1907 Charles Darwin's cousin, Francis Galton, asked 787 villagers at a county fair to estimate the weight of a prize ox. None of the villagers guessed the right answer, but then Galton did something with their answers that got him very close to the correct answer. What did he do, Danny?

Daniel Kahneman: Well, he simply took the average. The average, I think, was within two pounds of the correct weight. That led to a lot of research that was summarized in a recent book by James Surowiecki on the wisdom of crowds, and the fact that when you take multiple judgements, independent judgements, and average them you eliminate noise. This, by the way, is guaranteed to eliminate noise. If you take multiple judgements, there is no guarantee that it will reduce bias because if the judges agree on the bias then the bias will remain when you take the average. Indeed, it would be even more salient. What is absolutely guaranteed is that when you average independent judgements, you're eliminating noise. When you take four independent judges you're reducing noise by one half. When you take 100 you're reducing it by 90%. There is some mathematics of noise that lends itself to analysis, that doesn't apply to bias.

Shankar Vedantam: It's really remarkable. The correct weight of that ox was 1,198 pounds. As you said, that was one or two pounds off the correct weight. I want to point out that the reason averaging the responses produces a better answer, is that noise is random. You're taking advantage of the fact that various estimates will be randomly high or low, and that's why when you average them out you're going to get closer and closer to the correct answer.

Daniel Kahneman: What happens when you have different people making the same judgment of the same object, and then you're going to average them, then the errors they make cancel each other out. When people make judgements about different cases, errors don't cancel them out. If you set too high a premium in one case and too low a premium in the other case, that doesn't make you right. That just makes things worse. This idea that errors cancel out, you have to apply it quite precisely. They cancel out when you average judgments of the same thing.

Shankar Vedantam: Also judgements have to come from people who in some ways are independent of one another. If I am seeing the judgment you make, and then I make my judgment afterwards, my judgment really is just a reflection of your judgment, not an independent one.

Daniel Kahneman: That's right. What happens basically is when you have witnesses who talk to each other, the value of their testimony is fallibly reduced because in effecting the extreme, if you have one witness who is very assertive and all the other witnesses fit their story to his, then you have one witness regardless of how many testified.

Shankar Vedantam: One of the most remarkable aspects of the wisdom of the crowd that you describe in the book has to do with how you can elicit the wisdom of the crowd just from yourself. You cite research by Edward Vul and Harold Pashler that ask people to make judgments about the same thing separated by a certain amount of time. What did they find when you average out these different estimates?

Daniel Kahneman: For example, if you ask people “What is the population of London,” and you ask it once and then you wait a couple of weeks, and you ask it again, the striking thing is that most people will not give you the same number on the two occasions. The second striking thing is that the average of the two responses is more likely to be accurate than either of the responses. The first response is better than the second, but the average is better than both.

Shankar Vedantam: In one of the studies they conducted, they actually asked people to make estimates that were different than their initial estimates. Then they averaged out the estimates and they found that noise was reduced even further. Why would this be the case, Danny?

Daniel Kahneman: Here, what you are trying to do, and you can do it within an individual, is you are leaning against yourself. You made one judgment and then you asked people to think how could that judgment be wrong and then make another. That turns out to be indeed better than merely asking the same question twice.

Shankar Vedantam: In some ways, this provides a solution to the conundrum I posed to Danny. If noise is detectable only by studying statistical averages, how do I reduce noise in decisions I am making as an individual? The answer: try to make the same decision over and over under different conditions. One way to tell of noises behind my decision to propose marriage is to ask myself whether I would make the same decision under different circumstances, not just on a moonlit night in the springtime, but in the heat of summer, or in the dead of winter. If I reap the same answer in these different settings, it's possible I could still be making a mistake, but at least I can be somewhat reassured that my decision is not the result of random extraneous factors. Scientists are exploring lots of ways to reduce noise. The researcher Sendhil Mullainathan and his colleagues devised an algorithm to advise judges on whether to grant bail to suspects. These are people who have been arrested, but who have not yet been put on trial. Keeping them in jail can cause all kinds of hardship. People can lose jobs or lose custody of their children while they're incarcerated awaiting trial. It's costly for taxpayers to keep people in jail. But letting someone dangerous out of jail can cause harm. Maybe they go on to commit other crimes. The researchers had the algorithm offer advice to judges about whether to grant bail. They found that if judges incorporated the recommendations this could reduce the number of people in jail by 42% without increasing the risk of crime.

Daniel Kahneman: Research goes further than that, in that allowing the algorithm to inform the judge is actually not the best way of doing it. The research suggests quite strongly that when you have a judge and an algorithm that are looking at the same data, with some exceptions it's better to have the algorithm have the last word. This is very non-intuitive.

Shankar Vedantam: Yeah. Besides being actually superior in some ways in terms of judgment, one of the things that algorithms do better than people is that they're not noisy. They're actually much more consistent. Can you talk about this, that in some ways one of the advantages that algorithms have is even when their judgments might not be as good as humans because they have less noise than humans, you're able to get better outcomes.

Daniel Kahneman: Noise is a source of inaccuracy and algorithms by their nature are noise-free, that is when you present the same problem to two computers running the same software they're going to give you the same answer, which is not true of different bail judges. That advantage is in many cases sufficient to make algorithms superior to people, but I don't want to create the impression that our solution to the problem of noise is algorithms because even if it were the solution, there's just too much opposition to algorithms. Ultimately, we're talking about improving judgments. In some domains, algorithms can be used. I think where they can be used, they should be used, but this is a long process, a slow process because human judgment is going to make the important decisions for quite a while.

Shankar Vedantam: Isn't it interesting though, Danny, that when you look at the news and you see the news coverage of algorithms, I feel like just in the last year I've seen dozens of articles talking about algorithmic bias, about how algorithms in some ways can make judgements worse. It is the case that you can have poorly designed algorithms. You can argue that the old sentencing rules that we had, three strikes and you're out, in some ways that is an algorithm. But you could argue the algorithm in some ways was too crude to capture what actually needed to be done. Isn't it striking that there's so little attention that's paid by contrast to the potential good that algorithms can do? Because again, we're so focused with the story of intent of saying a bad outcome happened, an algorithm caused it, clearly algorithms need to be thrown out the window.

Daniel Kahneman: We do not want to accept the errors that blind rules will make. I was talking to someone who designed self-driving cars, and they realized that self-driving cars, it's not enough for them to be 100 times safer than regular drivers. They effectively have to be almost perfect before they will be admitted. It's that kind of bias that is completely human and natural. We like the natural over the unnatural. We prefer human drivers and human doctors to make mistakes rather than self-driving cars and medical algorithms. That's just a fact of psychology.

Shankar Vedantam: You talk in the book about something you call "Decision Hygiene", and others have talked about this idea as well. What is decision hygiene and why the analogy to public health?

Daniel Kahneman: When you're thinking of dealing with biases like a specific disease, so you think of a vaccine or you can think of medication, which is specific to that disease. When you're washing your hands, you're doing something entirely different. You have no idea what germs you might be killing, and if you're good at it, you will never know because the germs are dead. A similar distinction can be drawn between different ways of fighting errors. There is a difference between procedures that are specifically aimed at particular biases and procedures that are intended generally to improve the quality of the judgment in decisions. The way that this feeds back on the individual is that if there are procedures that are good for organizations and for repeated decisions, they should be good for individuals and for singular decisions.

Shankar Vedantam: Mm-hmm (affirmative). If I am a CEO of a corporation, or if I'm a policy maker and I'm hearing this conversation about noise, can you give me two or three really specific suggestions on ways that I can reduce noise in my decision making or in my company's decision making, or in my organization or community?

Daniel Kahneman: I think the first step would be to ask whether you have a task in the organization that is carried by interchangeable functionaries, like underwriters or emergency room physicians. They're carrying out the same tasks, making the same kinds of judgments, and you would like those judgments to be noise-free, to be uniform. First of all, identify whether you have that case in your organization. If you do, we strongly recommend you measure noise. That is, you actually take those individuals, present them with similar cases, and observe the variable of the judgements. Possibly, that may lead you to want to do something about it. The first step is just to measure noise because our intuitions about the magnitude of noise are systematically wrong.

Shankar Vedantam: Danny thinks we should learn from the saga of the rise and fall of sentencing reform. Once you detect noise in an organization, it may be wiser to avoid trying to fix the problem by asking everyone to follow rigid rules. As we've seen, people hate to have their judgment questioned, they hate to have their discretion limited, and they detest anything that smacks of mechanistic rules.

Daniel Kahneman: The main thing to do if you're attempting to improve the judgment of people in an organization is to convince those people that they want their judgements to be better. If you impose it as a set of rules that all of them will follow, they will resist it, they will feel they're being robotized, and they're likely to sabotage whatever you propose. This is well known in insurance companies that provides the underwriters in many cases with information or even with a technical price with the suggestion about what premium should be assigned. Underwriters are very likely to completely ignore those and to follow their judgment. Basically, I would think it's obvious advice, if you have a group of people who are noisy, have that group try to find the solution to the noise. Have them develop procedures that will make them uniform. Do not impose procedures on them, but work with them to make them more uniform because actually they will recognize that they would like to be in agreement with each other. But letting them feel that what they're doing is what they want to do rather than what they're being forced to do, that is clearly a very important step if people really want to have organizations that improve their judgments.

Shankar Vedantam: Daniel Kahneman, Olivier Sibony and Cass Sunstein are the authors of “Noise: A Flaw in Human Judgment.” Danny, thank you for joining me today on Hidden Brain.

Daniel Kahneman: It was really my pleasure.

Shankar Vedantam: Hidden Brain is produced by Hidden Brain Media. Our production team includes Brigid McCarthy, Laura Kwerel, Kristin Wong, Ryan Katz, Autumn Barns and Andrew Chadwick. Tara Boyle is our executive producer. I'm Hidden Brain's executive editor. Our unsung hero today is Rosalind Tordesillas. She's a producer in New York City who helped us record this interview with Danny Kahneman. Rosalyn got to Danny's place early to set up for the interview, and she was incredibly kind, conscientious, and patient. At various points in my conversation with Danny, sirens blared outside. At one point, a refrigerator in Danny's apartment woke up and started making noise. Through all of it, Rosalyn figured out how to get a crystal clear recording. Thank you, Rosalyn. You are a true unsung hero. If you like this episode and like our show, please consider supporting us. Go to support.hiddenbrain.org to learn how you can help. Every little bit makes a difference and it means a lot to us to see you step forward to help. I'm Shankar Vedantam. See you next week.

Podcast:

Subscribe to the Hidden Brain Podcast on your favorite podcast player so you never miss an episode.

google podcast subscribe
spotify podcast subscribe

Newsletter:

Go behind the scenes, see what Shankar is reading and find more useful resources and links.