Christie Aschwanden: On February 26, 2015, I was on a train traveling from New York City to Boston. 

It was a Thursday, and I was glued to my laptop, trying to get some work done. 

Suddenly my chat groups started blowing up—over a dress. You probably know the one.

[CLIP: David Greene and Linda Wertheimer speak on NPR’s Morning Edition.Greene: “Good morning, I’m David Greene, thinking maybe we should reconsider how we use our time–a debate about the color of a dress on the Internet has been consuming people, including people in this studio.” Wertheimer: “Definitely looks black and blue to me.” Greene: “You’re wrong, Linda; it is totally white and gold—whatever. This image we’re looking at here was posted on Tumblr. People online took sides and yelled at each other. Wired Magazine put a photo editor on the case. They say the dress is blue and they are wrong.”]

[CLIP: Theme music]

Aschwanden: I’m Christie Aschwanden, and this is Uncertain from Scientific American.

Today: part two of our five-part series on uncertainty. In this episode, we’ll talk with two researchers whose work probes the uncertainty surrounding how we perceive the world around us. 

It turns out that what we see may not always be a perfect reflection of reality. 

Consider “the dress.”

Pascal Wallisch: My name is Pascal Wallisch, and I serve as a clinical associate professor of data science and psychology at New York University.

Aschwanden: In the nine years since the dress blew up the internet, Pascal has made a career out of studying what the hell everyone was fighting about.

Okay, can you just start by telling me a little bit about the dress? What does it look like?

Wallisch: I must confess that I am not very good at describing dresses.

Aschwanden: So you don’t, you don’t have a minor in fashion?

Wallisch: No, no. If you had told me 10 years ago that one of your key research programs is going to be about the way people will perceive dresses, I would have said, “There’s absolutely no way that—that will never happen.” I don’t care about dresses. I don’t have a dress. I’ve never talked about dresses. And now I talk about it a lot.

Aschwanden: Oh, uncertainty! It can be so cunning!

Wallisch: You literally never know what’s going to happen next ...

Aschwanden: Or how some stranger’s fashion conundrum could change the trajectory of your career. And this dress—my first impression was that it’s really ordinary.

Wallisch: It’s a pretty standard dress, don’t you think?

Aschwanden: Yeah, it seems like the kind of thing you wear to someone’s wedding when you don’t want to be the center of attention.

Wallisch: And that is exactly what it was bought for. So I talked to the person who took the picture, Cecilia Bleasdale...and she said that was the purpose. She was wanting to buy a dress for the occasion of the wedding of her daughter, I believe. So they took pictures, and then they revisited the pictures to decide which dress to get, they realized they disagree about what the color of the dress was. They were like, “What do you mean ‘it’s white and gold?’” or “[What do you mean] ‘it’s black and blue?’” And that’s how this all got started.

Aschwanden: So what color is it?

Wallisch: Christie, you’re in luck. Let me show you something. Oh, I have the dress right here.

Aschwanden: Oh, my God, whoa! It is—definitely not the color I saw in the photograph. Wow. Wow. I need a moment to absorb this. Oh wow. 

Okay, but before we talk about the actual color, I want to hear about the first time you saw the image of the dress. What colors did you see?

Wallisch: I can tell you that like it was yesterday because it was such a big day in my life.... It’s a Thursday, and I was coming back from N.Y.U. on the train.... I had finished teaching for the week. And I was minding my own business, and [I] was at peace with the world.

I was on the train, and I was checking Twitter. And a student had tagged me. And they were like, “Professor, what do you see here?” And ... I was like, “You know, this is obviously, obviously a white-and-gold dress....or, an image of a white and gold dress” And a student asked, “Does it appear at all ambiguous?” I said, “No, why? Why do you ask?”

Aschwanden: And that’s how you learned that people were fighting over it.

Wallisch: I said, “Well, I guess I can see how somebody could see it slightly off-white, but it is clearly white and gold to me.”

Aschwanden: So you think they’re debating about white versus off-white. That’s hilarious and also totally understandable because I also saw a white and gold dress in the photograph.

Wallisch: Do you still, to this day, see it as white and gold?

Aschwanden: Absolutely! The very first time I ever perceived it as blue was when you showed it to me just now. It was shocking! What about you? What happened after that note from your student?

Wallisch: So I go home to my wife and tell her—I don’t know yet that it’s a thing yet— “Oh, you know, people online are so silly. They’re arguing about the color of this dress. And it’s obviously white and gold. I don’t know what they’re talking about.” So my wife looks me straight in my face ...

Aschwanden: Oh, no. She thinks ...

Wallisch: It’s black and blue.

Aschwanden: What?

Wallisch: At that moment is when I realized that we need a bigger boat, so to speak, because she’s a very serious person. She wouldn’t, like, lie to me about this.

Aschwanden: Now it’s really got your attention.

Wallisch: So some people early on said, “Oh, this just reflects different screen settings.” Yeah, but I knew, immediately, two things. Number one: she’s not trolling me. Number two: this has nothing to do with the screen because we are seeing it on the same screen.... I’ve done a Ph.D. in visual perception. I knew that we don’t already know this. This is new. This is completely new.

Aschwanden: You feel yourself on the cusp of scientific discovery. Wow!

Wallisch: I did not sleep at night. No, it’s true. That night, I wrote a screed as to why we don’t know what this is.

Aschwanden: What made it so new?

Wallisch: What I found remarkable from the beginning was you basically locked in to your perception.You can’t…you can’t switch it. Now…you can’t undo it, you can’t unsee it.

Aschwanden: So it’s not like those old-fashioned optical illusions. You know, the ones where the drawing can either be an old woman or a young woman. Or the duck or rabbit? 

And if you know about both, you can pretty easily go back and forth between seeing the two versions—it’s a duck’s bill or it’s a rabbit’s ears; it’s a duck, a rabbit, a duck. 

But you’re saying that with the dress, a single person can’t see both versions.

Wallisch: Most people cannot. And so that must reflect something about how they reduce ambiguity.

Aschwanden: Wait—what do you mean by “ambiguity”? What exactly is uncertain here?

Wallisch: So I want to be very clear and I make that point to everyone who is willing to listen–this is not about the dress. 

If we’re all seeing the dress under normal lighting together here, there is zero ambiguity. The ambiguity comes from the image and that specific image of the dress.

Aschwanden: So it’s something about the photograph.

Wallisch: There’s information missing in the image … in the image of the dress, the lighting is ambiguous. Yes, it could be indoor lighting; it could be outdoor lighting. I talked to the person who took the picture, and it turns out that it is indoors, but it is so overexposed with a Samsung cell phone camera, it’s hard to tell.

Aschwanden: That’s really interesting because I don’t remember the image seeming ambiguous at all.

Wallisch: We know from a long history of research on people and their cognition that if there is uncertainty, they don’t just say, “Oh, no, I have this uncertainty, and now I can’t act.” Your brain, your mind, however you want to call that, fills in this uncertainty, but in a smart way. It's like autocorrect is, like, smart guessing.

Aschwanden: So how does your brain make its smart guess?

Wallisch: Let’s try this right now. Imagine you’re at a staff meeting at Scientific American, and you are facing the whiteboard, not the door, and somebody was late. And in your experience, because you’ve done it before, Bob is always late. You don’t know who it was because you weren’t looking at the door, but who do you think it was, not having seen the door? Bob! Why—because [it’s] always Bob? Yes, exactly.

Aschwanden: But the difference is that in the staff meeting, I’m consciously making a guess. With the dress, I never knew there was any uncertainty to begin with. My brain made its guess on the fly, before I could possibly know what had happened.

Wallisch: Your brain is doing this for you unconsciously.

Aschwanden: So it’s smoothing over the uncertainties by making assumptions about what I’m seeing that are based on my previous experiences.

Wallisch: Yes, exactly.

Aschwanden: And you’re saying that if, say, it’s near sunset, and the light is getting dim. And then, I see a fire engine come down the road. Probably I’m going to assume that it’s red.  

Wallisch: Correct. That’s called the true-color effect.... Your brain is constantly color correcting.... As the day progresses, the mix of wavelengths that reaches you from the atmosphere, they shift from, like, bluish during the day [to], in dusk, more reddish. So as these shift, your brain has to take that into account.

Aschwanden: Okay, so my brain is like a fancy camera. It adjusts the color balance to make the colors look right in the image I’m seeing. It makes sure the fire truck appears red to me in the morning light and also some shade of red in the evening light. 

Wallisch: People are really good at that. But that presumes that you know what lighting is.... In the dress image, we don’t know what the lighting was—could it have been artificial lighting indoors or sunlight.

Aschwanden: Pascal had a hypothesis: what determined whether you saw the dress as white and gold or blue and black was based on what assumptions your brain unconsciously made about what you were seeing. 

If it assumed that the photograph was taken in bright outdoor light and the dress was in shadow, your brain color corrected to account for that, and the dress looked white. If, on the other hand, if you assumed the photo was indoors under artificial light, there wasn’t the same shadow to correct for. And therefore you saw it as blue.

Wallisch: The reason for that is shadows are bluish black if you look at them under [a] photometer. So if you mentally subtract [the] effect of a shadow, it will then look yellowish and whitish.

Aschwanden: The tricky thing about this is that we were all making these assumptions subconsciously, which meant that he couldn’t just ask us. Instead he had to come up with a proxy. So, some other way of getting at what assumptions your brain had made when you first saw the image.

Wallisch: I knew that people have a chronotype, meaning that some people like to get up in the morning and just rise to the sun, and other people, basically they’re like night owls. They are staying up late. And this is a genetically set kind of thing.

Aschwanden: Okay, this gets a little complicated, so I’m going to pause a moment here to explain.

Pascal guessed that just like we’d assumed that Bob was late to the meeting, our brains would default to the assumption that the light we were seeing in the photograph was the same kind of light we were most accustomed to seeing.

He surmised that over their lifetime, early birds get more short-wavelength blueish natural sunlight. On the other hand, people who stay up late get more long-wavelength artificial light.

In other words, he used chronotypes as proxies for what assumptions our brain was subconsciously making about the ambiguity of the lighting. 

Larks should be more likely to see the dress as white. They’re going to fall back on the assumption that it’s the brighter outdoor light. They see a lot of that kind of light. So they’ll correct for the shadow effect so that it appears white.

[CLIP: Lark calls]

Aschwanden: Meanwhile owls will be more likely to see it as blue. Their brain will assume it’s the kind of incandescent light they see inside at night. And under that kind of light, the dress is blue.

[CLIP: Owl calls]

Aschwanden: He tested this theory with a sample of about 1,000 people, many of whom he recruited from an article he published in Slate.

Okay, back to Pascal.

Wallisch: What we found is the more you identify as a night owl, the more black and blue you think that image is—it’s a dose-dependent effect. These are very, very strong effects.

Aschwanden: But still—chronotype is just a proxy for the thing Pascal was really trying to measure, which was what assumptions your brain made when viewing the image. He needed to go deeper.

The next step was to replicate the result in a new context to show that it wasn’t a one-off fluke and what you saw really did come down to assumptions your brain made about ambiguity.

Wallisch: So here’s what I thought, “What’s the principle here?” So first of all, we need to introduce uncertainty. We need to take an image, an image that could be any color, and put it under artificial lighting and take all context cues away.

Aschwanden: So you’re creating another condition where it’s not immediately obvious what the colors should be.

Wallisch: My collaborator, Michael Karlovich, came up with the idea of using Crocs and socks.

Aschwanden: Ah, Crocs—those funny plastic shoes that come in a whole range of colors. There’s no default color for them.

Wallisch: I think there’s something like 28 different kinds of Crocs, yes?

Aschwanden: Okay, so I’ve got the Crocs and socks image up on my computer screen right now. The socks are almost like a neon green. And the Crocs are a very bland color—maybe gray? I guess I’d call them beige.

Wallisch: Great. Now here’s the thing, here’s the crazy thing...so we were able to do the following. So, first of all, I have the…it’s too bad, if I had known that…I would bring the Crocs. I have the Crocs. The Crocs…the Crocs are pink–undeniably they are just as pink as the blue dress is blue.

Aschwanden: Really?

Wallisch: If you put the Crocs under—on a black background, and we did this in my attic—we blacked out my attic, and then you illuminate the Crocs ... and you put them under green light, that will add up to gray. And at that point, it’s unclear if you’re looking at gray Crocs on a black background under normal light or pink Crocs under green light.

Aschwanden: So it’s just like the dress—what you see depends on what you assume about the lighting.

Wallisch: But there’s one dead giveaway. It’s not just the Crocs. It’s the Crocs and the socks. Here’s the kicker: when I was young, socks like that were always white—always, always, always. So if I shine green light on these pink Crocs, even though they look gray, I will know the socks will look green, subjectively. But I know that the socks are white, so I can mentally subtract the green from the socks to then color calibrate.

Aschwanden: Wait, the socks are white? Oh, wow! This is so bizarre!

As you were talking about the white socks, I had this immediate recognition that, “oh, these are those white tube socks that everyone used to wear when I was a kid.” As that recognition clicked, the Crocs turned pink. 

It was almost instantaneous. It’s wild. I’ve never experienced anything like this.

Wallisch: We asked people, “What do you see the Crocs as? And what do you see the socks as?” And to make a long story short, and I have to send you the paper on this…the people who saw the socks as white and who are sure about this ... 80 percent or something like that see the Crocs as pink, even though they actually look gray at face value.

Aschwanden: So what’s happening here is that people who assumed the socks could be green, which is the color they objectively look, just based on wavelength, see the Crocs as gray or beige. 

But people who saw Bob—that is, they knew the socks were white because they’re always white—they color corrected and saw the Crocs as their true color, which is pink.

Once again, your brain is processing ambiguity by relying on your prior assumptions, and you don’t even realize it.

Pascal, do you think there are larger lessons we can take away from this?

Wallisch: Well, I think the biggest lesson is: don’t be so sure of yourself. Just because you’re sure does not mean you’re right. This uncertainty is hidden from you, yes?

Aschwanden: Oh, I was absolutely certain the dress was white. And when I heard other people saying they saw blue, I felt incredulous. 

But you’re saying that maybe next time, instead of immediately assuming that everyone else is denying reality, I should be more open to the idea that I am the one who is wrong?

Wallisch: Your brain doesn’t say, “I have absolutely no idea....” It doesn’t say that. Your brain tells you, “I think it’s white and gold.” That’s it. It doesn’t flag to you, “Oh, and by the way, I made a guess here.”

Aschwanden: Your brain is just making a judgment call, and you never even realize how uncertain it was in the first place.

Wallisch: Yeah. It’s like a game show. It’s our best guess; it’s our final answer. But it doesn’t tell you that–that it made a guess.

Aschwanden: So if we’re not careful, we can end up with a bunch of ignorant overconfidence.

Wallisch: I think we need a culture of awareness and intellectual modesty that counteracts this evolutionary tendency to jump to conclusions and then commit to them.

Aschwanden: We think that what we see represents stone-cold reality. And that can be a problem because it turns out that we have a natural blind spot for the ways we might misinterpret reality.

Psychologists have a term for this: “naive realism” is the assumption that the way we see the world is the way it really is. We are blind to our blind spots.

Pascal’s experiments are on visual perception, but they highlight a concept that’s true in many aspects of science: the uncertainty we don’t perceive can warp our view.

Act 2: The Many Hypotheses Study

In this episode, we’ve been talking about uncertainties we don’t easily see and how they can shape our vision. 

But it’s not just our visual representations of the world that can be influenced by unrecognized uncertainties. The results of scientific studies are also influenced by uncertainties that are not always apparent to us. 

These invisible uncertainties are something that researchers are starting to explicitly study in hopes of understanding just how much they can change experimental results.

Justin Landy: My name is Justin Landy, and I’m an assistant professor of psychology at Nova Southeastern University.

Aschwanden: Justin was one of the organizers of a large study that aimed to look at how the methodologies that scientists use can influence their results.

Landy: What we were interested in was answering a question that basically boils down to how much of what we observe as the results of scientific studies comes from the decisions that the researchers make in designing the studies.

[CLIP: Music]

Aschwanden: So the things like what to measure and who your test subjects will be and which statistical tests to deploy.

Landy: Anytime a scientist sits down to test some hypothesis, what we have to do is we have to take kind of the high-level conceptual hypothesis, and we have to turn it into something concrete and measurable. We call this operationalizing.

Aschwanden: Right. Sort of like how, earlier in our show, Pascal had to figure out a way to quantify what assumptions people were subconsciously making about the lighting in that photograph of the dress.

Landy: For any conceptual-level variable, there are really an infinite number of ways you can operationalize it to sort of bring it down to Earth and make it concrete. And so we were curious: How much do those kinds of decisions matter to whether you find support for your hypothesis or not?

Aschwanden: So what Justin and his colleagues did was take five different research questions in psychology and then crowdsource an investigation into them.

Instead of a single research team doing the study, he recruited about a dozen different teams with researchers from all over the world to design and conduct their own ways of testing the hypotheses.

Landy: We just said, “Go make the study that you would run to answer this question.”

Aschwanden: Each team designed its own study to come up with an answer to the question, which meant that each question had about a dozen studies that were run to test it.

Landy: We recruited two gigantic samples online through two different recruitment methods. We ended up with over 15,000 participants total.

Aschwanden: So you used participants you gathered through Amazon’s Mechanical Turk service and a survey company called Pureprofile. And these 15,000 participants didn’t just take part in one study—they were each randomly assigned to take part in multiple experiments.

Landy: So you essentially take part in five separate studies in a row—one for each question. But if you got Team One’s materials for research question one, you might get Team Seven for research question two.

Aschwanden: Ah, so it’s a way to use the same research subjects for all of the studies. Very clever!

Landy: What this lets us do is, basically, we can examine “How much do the results of these different studies differ from one another?” And unlike in sort of the usual practice of science, where one study is published, and then somebody else goes and runs a conceptual replication, if they get a different result, the original author can come back and say, “Oh, well, you know, this isn’t really testing my hypothesis the right way” or “Oh, well, maybe your results are different because you tested a different group of people,” right...? Something like that. They’re all off the table here.

Aschwanden: Okay, so each team designed its own way of testing each hypothesis, and it ran its studies with the same sample everyone else used. 

Now, without getting too deep in the weeds, let’s talk about the five hypotheses or research questions. The first one asked Do people report an awareness of hiding negative automatic associations about members of negatively stereotyped social groups.

Landy: That one, we got a lot of different designs focused on a lot of different groups, everything from racial groups to people who are overweight to, to…I think we got one related to, like, immigrants—just lots of different groups there.

Aschwanden: Another hypothesis was that negotiators who make extreme first offers will be trusted less relative to negotiators who make more moderate first offers.

Landy: The next one was “What are the effects of continuing to work despite having no material/financial need to work on moral judgments of the individual?”

Aschwanden: So in other words, are people who work when they don’t need the money judged in a positive or negative or neutral light?

And there was also a hypothesis about the reasons people might object to performance-enhancing drugs in sports.

And a final one was a complicated question about moral orientation and personal happiness.

Landy: Every team was given the same question and developed their own way of testing it.

Aschwanden: And what were the results?

Landy: What we find is just massive, pretty-much-unexplained variability.

Aschwanden: Some studies produced results that supported the hypotheses, and other studies had results that went the other way.

Landy: Yeah, so four out of five research questions, we found not just results in opposite directions but statistically significant results in both directions.

Aschwanden: So the results were kind of all over the place, but you did find one hypothesis that was pretty strongly supported by the majority of the studies, right? The one about how making an extreme initial offer in a negotiation reduces trust in the negotiating partner.

Landy: Even that one, we found one result that went the other way. And now it’s just one out of a dozen. But still, if you had just set out to test that hypothesis, and you use that particular study design, which is really no more or less valid than any other, you would have concluded that, oh, no, actually, they trust you more if you make an extreme offer.

Aschwanden: But if you know that there are 10 or 11 other studies that show the opposite, then that one result looks like an outlier.

Landy: So, yeah, there are conflicting results for almost all of these research questions we tested. So if you just ran one version of the study, one particular design, you might come to the wrong conclusion.

Aschwanden: Do we just throw up our hands and say, “Well, we’ll never know,” or was there some sense that the studies might be closing in on an answer?

Landy: Some hypotheses are more supported than others, right? There is truth value there, which is—that’s reassuring. The one about continuing to work, generally we found pretty consistent evidence that made moral evaluations of the person more positive. The other ones, the evidence was much less clear.

Aschwanden: So what this project really says is that those individual studies contain more uncertainty than may be initially apparent.

Landy: We just find so much variance in the results that we can’t really explain as of now.

Aschwanden: And this isn’t the first big study like this to find this kind of ambiguity, right?

Landy: So my colleagues on this paper have another paper where they, instead of having people design multiple studies, they gave them a data set from one study and a question and say, “Go do the analysis.”

Aschwanden: Ah, I remember that study. It happened a few years prior to this one. They gave 29 research teams the same dataset. The goal was to use data to answer this question: do soccer referees give more red cards to dark-skinned players than light-skinned ones.

Twenty teams concluded that soccer referees give more red cards to dark-skinned players, and nine teams found no significant relationship between skin color and red cards. They were all using their best judgment to do the analysis, using the same data, but their answers were different.

Landy: There’s all of these sources of variance in findings, and you can’t account for all of them, I don’t think. The best you can try to do is standardize along as many as you can and quantify how much does one matter.

Aschwanden: What are the larger lessons here?

Landy: The main takeaway is just that these decisions we make as researchers are consequential, right? They matter more than I think we often assume they do. And not only do they matter, but they matter in ways that maybe we wouldn’t anticipate going in.

Aschwanden: There’s the unseen uncertainty. How do you wrap your head around it?

Landy: We need to be maybe a little less confident in our high-level, conceptual-level conclusions from any given study of any given topic—and a little more open to the idea that we may not have an answer for a while until we’ve tried this in a bunch of different ways with a bunch of different groups of people, a bunch of different analytical approaches, and so on. That is really the only way we’re going to get a real firm handle on what’s going on.

Aschwanden: What does that mean for how we think about science?

Landy: I think that [what] we should be learning from this is that uncertainty or variability is just part of the scientific process, right? 

And so we shouldn’t be putting maybe a ton of stock in any particular single study. And what we should be doing is stepping back trying to bring together results from lots of different approaches, lots of different designs, to try to see “Okay, across all of these choices that we might make, what is the actual overall support for this finding?”

Aschwanden: It’s a good reminder that all science is provisional—there’s no hard end point.

Landy: That’s exactly what I tell my undergraduate students.... I always try to teach them that science is not a collection of facts despite what you may have been taught in high school. It’s a process.

Aschwanden: I’ve come to think of it as a process of uncertainty reduction.

Landy: It’s not that there’s no truth value. Two of these research questions showed very clear results aggregating across every design. Some of them didn’t, right? There was no clear support for three of the hypotheses, but two of them there were. Those hypotheses seem to be, if you want to use the term, right, or true, right?

But we get to that truth, that knowledge, over the course of many studies and often many years, right? 

And so when science goes back and corrects something and updates a conclusion, that’s not a failure of the process. That’s the process working exactly the way that it’s supposed to, right? We gradually continue to sharpen and refine what we know and approach greater understanding.

Aschwanden: What does this mean for nonscientists? How do we know if results are reliable?

Landy: If I had one kind of practical piece of advice for the public, and I hate to say this to a science journalist, but it’s: maybe don’t trust the headlines so much, right? Because the way the public gets exposed to scientific findings is through journalism. And a lot of journalists are very eager to jump on the hot, new, cool result.

Aschwanden: I mean, you’re not wrong. But scientists are guilty of this, too.

Landy: You know, we scientists, we’re trying to get at the right answer, but we’re also human, and we tend to want our results to be quote, unquote, “right.” We want them to be real, right? And so maybe we have a little more faith in our own results than we should for any, again, any single given study.

Aschwanden: As a journalist, I have to say that I’ve become a lot more tempered over my career. It’s hard for me to get excited over single studies any more. I’ve seen too many sexy results later overturned.

Landy: We don’t want to overstate, I think, is what it comes down to, right? And so you can explain what one study found. But you should be up front that no study is never perfect, right?

There will probably be conflicting results from other ways of approaching the question. And we may not have a real answer for a while. And that’s just how science works. It’s a slow, iterative process.

Aschwanden: People often complain to me that headlines are so herky-jerky. One day coffee is good for you; the next it will kill you. What do you say to that?

Landy: There’s often going to be what appear to be conflicting results, and the knowledge advances by figuring out how you explain the contradictions, right? That’s often what science is. It’s very rarely that one result is true and one is false.

It’s usually—it depends on something, right, and trying to find what that something is and explain the superficial contradiction, I think that’s usually how science advances, or at least those are the biggest advances when we reconcile things that seem to be in conflict.

Aschwanden: What I’m hearing is that those contradictions may actually be fertile ground for discovery.

Landy: One of my undergraduate professors said this at our graduation, that the phrase that heralds scientific advancement is not “Eureka!” It’s “Huh, that’s funny.”

Aschwanden: Right! It’s those things that seem a little weird or unexpected because we haven’t figured it out yet.

Landy: So, I think that’s exactly the way of thinking about it—is that uncertainties, inconsistencies, contradictions, they raise new questions, and we scientists are in the business of answering questions to the best of our ability. And so I think that that’s exactly the right way of thinking about this—opportunity rather than failure.

[CLIP: Music] 

Aschwanden: I’m seeing a theme here—the idea of uncertainty as an invitation rather than a threat.

Next time on Uncertain:

We’ll look at how uncertainties in the scientific process play out and what can be done to address them.

Our show is produced by me, Christie Aschwanden, and Jeff DelViscio. Our series art is by Annaissa Ruiz-Tejada. Our music is from Epidemic Sound. 

Funding for this series was provided by UC Berkeley’s Greater Good Science Center–it’s part of the Expanding Awareness of the Science of Intellectual Humility Initiative, which is supported by the John Templeton Foundation. 

This is Uncertain, a podcast from Scientific American. Thanks for listening. 

Listen to the first episode of Uncertain.

Listen to the third episode of Uncertain.

Journalist CHRISTIE ASCHWANDEN, a frequent Scientific American contributor, is author of Good to Go: What the Athlete in All of Us Can Learn from the Strange Science of Recovery (W. W. Norton, 2019).