How a Single Source of Information Can Seem Like Several

It must be great if everyone is talking about it, right? Learn how we overweight a single source of information when exposed to it through multiple channels.

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:06Katy Milkman?

  Stanley Quencher tumblers. They're tall, stainless steel, vacuum-insulated drinking containers with a handle and straw. They're also a cultural phenomenon. These $45 water bottles took the internet by storm around 2021. People waited in line for hours outside Starbucks and Target and sprinted towards shelves holding the colorful containers. Cup swap groups cropped up, and even resale markets. There were spreadsheets tracking launch dates for new products.

• 0:51Katy Milkman?

  There were even heists involving the theft of these popular mugs.

• 0:54Speaker 01

  Allegedly stealing nearly $2,500 worth of these stainless steel statut-

• 1:00Katy Milkman?

  On TikTok, it was hashtag #Wattalk. The cups appeared in the hands of women everywhere, which was surprising for a 100-year-old brand known mostly for camping thermoses. So what propelled these tumblers into the hydration-status symbols they became? Three marketers with an e-commerce site and an eye for what women want. The trio loved the Stanley Quencher, and so did the traffic on their site, the Buy Guide. So when they learned their beloved bottle was about to be phased out, they stepped in and reached out to the company. They convinced executives that these bottles were desirable and saleable. They demonstrated this with an Instagram post, where influencer Emily Maynard of The Bachelor

• 1:45Katy Milkman?

  fame shared her gifted Quencher. The company eventually came around to the idea of influencer marketing and let social media and its wave of seemingly organic endorsements do the rest.

• 1:58Katy Milkman?

  Soon, it looked like everyone loved the Stanley Quencher. What we saw were lifestyle creators and wellness influencers showing off their cups. What we didn't necessarily see was the affiliate connection. And by the time people in our circle started carrying their own Stanleys, we'd lost sight entirely of what set these sales in motion. We figured they were truly a superior drinking vessel and not the product of a single social marketing campaign. The Quencher boosted Stanley's annual sales from 73 million in 2019 to 750 million in 2023.

• 2:34Katy Milkman?

  In this episode, we'll look at how one source of information, then echoed by seemingly independent sources, can distort our judgment.

• 2:42Speaker 02

  Love my Stanley. Everyone should have one.

• 2:54Katy Milkman?

  I'm Dr. Katie Milkman, and this is Choiceology, an original podcast from Charles Schwab. It's a show about the psychology and economics behind our decisions. We bring you true and surprising stories about high-stakes choices, and then we examine how these stories connect to the latest research in behavioral science. We do it all to help you make better judgments and avoid costly mistakes.

• 3:25Speaker 03

  The cold is something that has bothered people ever since we've been around. It's a very unpleasant thing. For seven days, you are plagued with these symptoms of coughing and sneezing and whatever. And people have tried everything.

• 3:42Katy Milkman?

  This is Joe.

• 3:44Speaker 03

  Hi, I'm Josh Schwartz.

• 3:45Katy Milkman?

  Joe is a professor of chemistry at McGill University in Montreal, Canada. He's also the director of the university's Office for Science and Society. That office has a mandate to separate sense from nonsense, myth from fact.

• 3:59Speaker 03

  Maimonides in the 12th century recommended chicken soup for the common cold, which, of course, we still try today, even though there's no real documentation. People are very keen to jump on anything that offers a simple solution to a complex problem.

• 4:15Katy Milkman?

  You may remember us covering this tendency to prefer a simple explanation in a previous episode of Choiceology. It's called "The Fallacy of the Single Cause." But what we're talking about today is actually quite different. It's about how we can get mixed up about credibility when there's a single source for a piece of information.

• 4:35Katy Milkman?

  This story starts with a famous scientist.

• 4:38Speaker 03

  Here you have a top-notch scientist telling you that he's got the answer. It's not surprising that you jump on that bandwagon.

• 4:46Katy Milkman?

  The top-notch scientist is two-time Nobel Prize winner, American chemist, and peace activist Linus Pauling. The bandwagon? Vitamin C.

• 4:57Speaker 03

  Pauling was a brilliant scientist, starting from early childhood where he built himself a lab in his house.

• 5:05Katy Milkman?

  His house was in Portland, Oregon, where Pauling was born in 1901. He attended Oregon State University and then the California Institute of Technology, or Caltech. He earned his PhD in chemistry there, and then they hired him to join the faculty. By 1930, he had risen through the ranks to become a full professor.

• 5:25Speaker 03

  He was very dedicated. At a very young age, he published very impactful studies. So this was not a quack, you know, by any means. He was a very legitimate scientist who certainly contributed a great deal to quantum mechanics and the chemical bond. And he almost got the structure of DNA before Crick and Watson. He thought that DNA was three strands instead of two, but he was on track there all by himself. So there's no question that Linus Pauling was a stellar scientist.

• 6:03Katy Milkman?

  Joe Schwartz was studying chemistry himself when he first learned about Pauling.

• 6:07Speaker 03

  If you were into chemistry, you came across Linus Pauling and his classic textbook. He wrote what, you know, at that time was the definitive book on the chemical bond, and everything in chemistry is based on chemical bonds.

• 6:25Katy Milkman?

  It was Pauling's work on chemical bonds that would win him a Nobel Prize in Chemistry in 1954. And then later, he won a Nobel Peace Prize for his campaign against nuclear weapons and his advocacy for international peace.

• 6:37Speaker 03

  He is the only person in the history of the Nobel Prize to win two unshared Nobel Prizes. That's something.

• 6:45Katy Milkman?

  Pauling really was a remarkable scientist. He made fundamental contributions in a range of fields: chemistry, biology, and medicine. He's one of the most celebrated scientists of the 20th century and one of the most trusted.

• 6:59Speaker 03

  When you're a Nobel Prize winner, you're basically scientific royalty. So, anything that you say gets listened to. And when he started with his vitamin C business, of course, it got immediately reported. I mean, it would have been ignored if it was anyone else. But this was Linus Pauling, who, you know, was arguably the greatest chemist of the times.

• 7:21Katy Milkman?

  If you've ever heard of the healing properties of vitamin C, Linus Pauling is likely the reason why. And this is where it started.

• 7:30Speaker 03

  Pauling was diagnosed with a type of kidney disease. And his physician recommended a diet, which was a low-protein, low-salt diet, on which he did much better. And this initiated in his mind the concept of what came to be called orthomolecular medicine. That was his brainchild. Ortho means right. So, orthomolecular means using the right molecules. So, because of his success with this treatment, which was nutritional, he thought that the body just needed all the right molecules in order to work efficiently.

• 8:12Katy Milkman?

  In 1966, after an awards event where Pauling gave a lecture expressing a desire to live another 15 to 20 years, he ran into biochemist and vitamin C advocate Erwin Stone.

• 8:25Speaker 03

  After his lecture, Stone came to him and told him about his ideas about how humans lost the evolutionary vestige of producing vitamin C. He was quick to buy into this, even though there was no physical evidence. It just sounded right to him.

• 8:47Katy Milkman?

  On Stone's advice, Pauling and his wife started taking 3,000mg of vitamin C daily. For context, you need about 10 to 20mg a day to ward off scurvy, and the recommended daily allowance is closer to 100. At one point, Pauling and his wife were taking 18,000mg daily. The mega dosing seemed to make them feel good. They claimed to catch fewer colds, and Pauling recommended everyone give this a try.

• 9:15Speaker 03

  That is what gave rise to the original book, Vitamin C and the Common Cold, which of course took the world by storm. Because here was the most famous scientist in the world, arguably at that time, telling people that they should be taking vitamin C. And sales, of course, jumped. Vitamin producers noticed that right away. Sales kept increasing because so many magazine articles and newspaper articles talked about it. You know, it's the kind of thing, you repeat something often enough and it becomes true by repetition.

• 9:50Katy Milkman?

  There's another Choiceology episode from a couple of seasons ago about how simple repetition can make something seem more true. This is called the illusory truth effect. But here, the information about vitamin C preventing, or at least minimizing, colds, was repeated not just by Linus Pauling, but also by myriad other seemingly independent sources.

• 10:11Speaker 03

  So when you see articles all over the place, even though they all start from the same bit of information, but you just think, you know, well, all of these articles, they can't all be wrong.

• 10:24Katy Milkman?

  But they were wrong, because they all came from the same source, a reputable one at that, but as you'll see, an incorrect one.

• 10:33Speaker 03

  Here was a guy who had published hundreds of papers in the scientific literature, obviously all peer reviewed. He had a full understanding on how scientific methodology works. And here he was publishing Vitamin C and the Common Cold, based on anecdotes, his own and his wife.

• 10:54Katy Milkman?

  The scientific community wasn't impressed by this lack of rigor, but they were intrigued by the claims.

• 11:00Speaker 03

  Because of who Pauling was, they said, okay, well, you know, let's just take a look at it, just in case there is something to this. And there were a number of studies carried out, some on marine recruits, very good population to do such studies because, you know, if you have a marine recruit and you tell him to jump, the only question he asks is how high? So if they were told to take two grams of vitamin C a day, they did that, you had no doubt. And it didn't show that it prevented the common cold. There were a number of other studies done on that as well. Some showed a minimal effect if you take regularly a few hours less of the length of the cold. I mean, not very encouraging.

• 11:44Speaker 03

  So the evidence just wasn't there.

• 11:48Katy Milkman?

  In a perfect world, those inconclusive studies would have counteracted Pauling's claims. But the information was now snowballing, and businesses saw an opportunity.

• 11:58Speaker 03

  By that time, the supplement industry had wound up. And, you know, when they start promoting things, it works. But there's just no evidence for Pauling's original contention, and certainly not of taking 18,000 milligrams. His concepts were scientifically valid. It's just that I think he went overboard.

• 12:25Katy Milkman?

  Pauling didn't stop with the common cold. He went on to propose that the supplement had benefits for other diseases, too. He wrote a book about cancer and vitamin C, and he linked vitamin C to heart health.

• 12:37Speaker 03

  I actually heard him at a conference in Ottawa. There must have been 1,000 people in the ballroom. He was the keynote speaker at a chemistry conference. He showed a graph. It showed an increased rate of heart disease and decreased intake of vitamin C in the population. One of the real boogies in science, of course, is the correlation causation business, you know? And I remember sitting in there thinking, I mean, here is this hero. He was a hero of mine, you know, just throwing up that graph and saying conclusively that this demonstrated that taking more vitamin C would reduce heart disease.

• 13:21Speaker 03

  Nobody said anything. By that time, he was such an exalted figure that you didn't dare bring that up.

• 13:28Katy Milkman?

  But every scientist there knew that correlation does not equal causation. And these wild claims have not stood up to rigorous testing. So why is this idea about vitamin C still out in the world and on our drugstore shelves? One answer is that people have heard it works so many times from so many seemingly different sources that they believe it must be true.

• 13:52Katy Milkman?

  For now, taking high doses of vitamin C is like taking a hope pill.

• 13:56Speaker 03

  People live by hope. Anytime that there's some study which shows some potential, they will blow that out of proportion. It just keeps rolling along. I mean, there's so much money to be made.

• 14:13Katy Milkman?

  Dr. Joe Schwartz is the director of the Office for Science and Society at McGill University, where he is also a professor of chemistry. You can find more details about his work in the show notes and at schwab.com/podcast.

• 14:34Katy Milkman?

  Have you ever had something recommended to you on social media? Say a book. Someone says they've heard it's amazing, and then a few more people say this too. Maybe in a group chat or another post? And maybe you thought, wow, everyone loves this book. But what if all those people heard it was amazing from the same source? A review in a popular magazine? Or maybe from Oprah's Book Club? To you, it sounded like a bunch of separate convergent opinions. But really, just one common source has shared information. This can happen with books, restaurants, running shoes, stock tips, workplace rumors, and medical advice. And when we don't realize that all of the information we're getting is connected,

• 15:19Katy Milkman?

  we can end up way off base when making decisions for ourselves. Our failure to account for the fact that information from seemingly different sources might in fact be related or come from the same single source is called correlation neglect.

• 15:35Katy Milkman?

  Economists Florian Zimmerman and Benjamin Enke conducted experiments where participants tried to determine an unknown number based on guesses other people shared with them. In one condition, participants received multiple independent guesses, five and seven, for example. And then, most participants wisely averaged across these guesses to come up with six. In another condition, though, participants were told one person's guess, five, for example, and the average of that guess with someone else's. Let's say six, implying the other person's guess was seven. Here, participants failed to think clearly about what to do. Instead of guessing six under these circumstances,

• 16:19Katy Milkman?

  which would have been the right path, many guessed the equivalent of 5.5, averaging the five and six, and showing a bias where they treated correlated information as if it were independent. Florian Zimmerman joins me today to talk about the implications of this research. Florian is a professor of economics at the University of Bonn in Germany. Hi Florian, thank you so much for joining me today.

• 16:45Speaker 04

  Hi Katie, my pleasure.

• 16:46Katy Milkman?

  I'm really excited to talk with you about correlation neglect. And I was hoping we could just start with a definition.

• 16:52Speaker 04

  Yes, of course. So by correlation neglect, we mean that if people get information from multiple sources and these sources are somehow connected, thereby generating a correlation. If they neglect this correlation, if they ignore this correlation in their belief formation process, this is then what we mean by correlation neglect.

• 17:12Katy Milkman?

  So what are some situations where we might get information from multiple sources and they might be correlated or uncorrelated? Just so our listeners can get kind of an intuition for what you have in mind.

• 17:23Speaker 04

  Yes. So think about forming a belief, an opinion about something, and you're perhaps consulting your social network. So you ask a bunch of friends. This would already be a situation where you are getting information from multiple sources. And now, this is also an example where I think oftentimes information will be correlated in an interesting way. For instance, picture a situation where... There is a friend in your social network and this friend is very influential in that network, meaning that he talks to many other people in your network. And by talking to these other people in your network, probably the opinion or the information you will get from these friends will partly reflect what they heard from this very influential friend.

• 18:05Speaker 04

  We often call this a common source of information. And this information or the knowledge that this very important friend in your social network has is getting partially repeated by other people in your social network. And that creates a specific correlation structure.

• 18:21Katy Milkman?

  So say I hear from three different friends about an amazing financial opportunity and I think, wow, this must be really good. But a correlation structure would be something like they all heard it from the same central friend and then they repeated that information to me. And to me, it felt like it was three independent people telling me, wow, this is a great investment opportunity. But in reality, it was actually just one who had that kind of diffusion. Is that kind of what you have in mind?

• 18:48Speaker 04

  That's exactly right. And it wouldn't even need to be that they exactly repeat what the friend said. It would be enough if they're sort of influenced by what this common source of information, this common friend says. That's exactly right, yeah.

• 19:00Katy Milkman?

  Got it. And am I right in thinking that it doesn't have to just be through social networks, that this could happen through other channels as well? Like if I listen to the radio and watch the news at night, is that another situation where I might need to worry about this?

• 19:13Speaker 04

  Absolutely, yes. So think of a situation where there's a very influential experts and different journalists are sort of partially repeating what this very influential expert is saying. Then by sort of consuming different news outlets, you're going to be exposed to correlated information.

• 19:29Katy Milkman?

  And if I neglect that, it would mean I treat those as if they're both news sources, both the radio program and the news program giving me some warning or suggesting some course of action. If I am neglecting that they both got the same experts to give them that guidance, then I might put too much emphasis on that information, think it's higher quality than it is?

• 19:51Speaker 04

  Yes.

• 19:52Katy Milkman?

  Okay.

• 19:53Speaker 04

  If you ignore this correlation, you're going to put too much weight on this common source of information. So if this common source of information is overly optimistic on some issue, you're going to be overly optimistic. And vice versa, if this person is going to be, or this news source is going to be overly pessimistic, you're going to be overly pessimistic.

• 20:11Katy Milkman?

  This is really interesting. It seems really relevant to so many different situations. I'm wondering if you can tell us a little bit about what you think causes correlation neglect.

• 20:21Speaker 04

  Yes, we were, of course, extremely curious to delve into this. And perhaps it's useful to think about a sort of very, very simple framework of what could go wrong if you're dealing with correlated information. The first step, if as a respondent in our little experiment, you approach this task, it might be that you simply fail to notice the correlation structure. You know, there's so many things happening, so many different numbers you're seeing, you just don't even notice that there is this correlation structure. That could be sort of the first step where you're failing. The second step could be that, well, you notice it, but what you don't notice is that actually by ignoring this correlation structure, you are making a mistake. You are systematically biasing your own beliefs. That could also be happening, right?

• 21:06Speaker 04

  So noticing might not be enough. You also need to realize that there's a problem. And then the third step would be, well, suppose you've realized that there's a problem, you also need to know how to solve it, so how to correct for the problem. So in principle, each of these three steps could go wrong. And we've conducted a bunch of experiments that essentially tell us that it's not the third part. So it's not that people don't know how to correct for this bias. So it must be something more along the lines of either failing to notice the correlation structure or failing to notice that there is a problem with this correlation structure.

• 21:44Katy Milkman?

  I also seem to recall that you found not everyone exhibits this bias. Am I remembering that correctly? That you can sort of disentangle the types of people

• 21:52Speaker 04

  Yes.

• 21:52Katy Milkman?

  who are most susceptible to it?

• 21:54Speaker 04

  That's right. So if we sort of were to classify our participants, essentially, we would get two groups. One group that really fully neglects or ignores the correlation structure, and another group that actually does a fairly good job at incorporating the correlation structure and correcting for it.

• 22:12Katy Milkman?

  That's really interesting. Do your studies give you any hints about what we could do to eliminate or reduce this neglect in the people who don't behave properly? What might change that so that they could be more optimal in their decisions?

• 22:28Speaker 04

  Yeah, so I think there's a couple of hints from additional experiments we've done, and they all point to the same thing, which is that if you really focus their attention on the correlation structure, and there's different ways to do it, but you know, if you really put an alarm that says, "Look, be careful, watch out for this particular structure," then they actually do the right thing. So I think one implication would be that in sort of real-world situations where we might worry that people are ignoring correlations, our results would suggest that in these situations, really sort of hitting them on the head with, you know, be mindful of this particular structure of your information environment, that that already might help in some situations at least.

• 23:10Katy Milkman?

  So, you know, reminding people, "Hey, if you heard this information from Sarah, make sure you think about whether or not it ultimately came from Brian." Or, you know, "If you hear this from the Boston Globe, you should check and see if it was the same sources as the NPR show you listened to this morning." Something along those lines, kind of a reminder.

• 23:31Speaker 04

  That's exactly right. If you think about the underlying psychology, it seems that in such situations where people consume information from multiple sources, they are already sort of overwhelmed by the information they're getting, which makes them very susceptible for simply ignoring the specific aspect of the information structure. And then, as you say, by reminding them, by really focusing them on the importance of this specific aspect, that might already help at least some people.

• 23:58Katy Milkman?

  That's great. I'm curious if you do anything differently in your life now that you know about this bias and have studied it thoroughly.

• 24:07Speaker 04

  That's a great question. I'm sure I do. I guess kind of a boring answer is that I more often than before really simply ask myself, "Where does a certain news outlet or perhaps even a friend I'm talking to, where does he or she get his information from?" So, you know, questioning that a little bit more.

• 24:27Katy Milkman?

  That's a great answer. I love that. I so appreciate you taking the time to speak with me today. This was really interesting. I learned a lot, and I know our listeners will be grateful as well. So, thank you so much.

• 24:40Speaker 04

  No, no. Thank you, Katie. It was a pleasure.

• 24:47Katy Milkman?

  Florian Zimmerman is a professor of economics at the University of Bonn. And along with Harvard economist Ben Enke, he uncovered the problem of correlation neglect. You can find links to his work in the show notes and at schwab.com/podcast. There's no shortage of hype, or occasionally hysteria, in the financial media when popular narratives get recycled. For objective, data-driven analysis of both the stock and bond markets, check out the Schwab original podcast, On Investing. Hosted by chief investment strategist Lizanne Saunders and chief fixed income strategist Kathy Jones, new episodes of On Investing drop every Friday.

• 25:29Katy Milkman?

  You can find them at schwab.com/oninvesting, or in your favorite podcasting app. Since learning about correlation neglect, I found myself constantly asking, where is this information really coming from? When two colleagues raise the same concern, I pause to consider, are these independent observations, or did they both witness the same incident? When I see the same product hyped across multiple social media accounts, I ask, are these truly separate endorsements, or are they all drawing from the same original post? The danger of correlation neglect is that it can make weak evidence feel overwhelming.

• 26:13Katy Milkman?

  We assume consensus where there's really just repetition. But simply asking, what's the source and is it the same? When we hear related news, gossip, or advice can help us avoid placing too much faith in, say, vitamin C. It's a small mental habit that can make a big difference in how we judge what's credible and what's not. Of course, I would be remiss if I didn't point out the risk that your knowledge of this bias could turn into a bias itself. It would be a mistake to become overly skeptical of high-quality, repeatedly and independently validated information, say, on vaccine safety, or the risk of buying penny stocks,

• 26:57Katy Milkman?

  or the dangers of climate change. The trick is to stay curious. Don't blindly trust repetition. But don't dismiss consensus either. Instead, for important decisions, try to trace the source of what you're hearing. Understanding where information comes from is one of the surest ways to make better decisions.

• 27:31Katy Milkman?

  You've been listening to Choiceology, an original podcast from Charles Schwab. If you've enjoyed the show, we'd be really grateful if you'd leave us a review on Apple Podcasts, a comment or rating on Spotify or YouTube, or feedback wherever you listen. You can also follow us for free in your favorite podcasting app. And if you want more of the kinds of insights we bring you on Choiceology about how to improve your decisions, you can order my book, How to Change, or sign up for my monthly newsletter, Milkman Delivers, on Substack. Next time, I'll speak with Cornell University psychologist Tom Gilovich. He'll explain our tendency to vastly overestimate the likelihood that other people will notice or pay attention to minor flaws in our appearance, decisions, or behavior.

• 28:16Katy Milkman?

  I'm Dr. Katie Milkman. Talk to you soon.

• 28:26Speaker 05

  For important disclosures, see the show notes or visit schwab.com/podcast.