QAA Podcast - Trickle Down Episode 18: Viral Science (Sample)

Episode Date: March 1, 2024

Brian Wansink was the Head of the Food and Brand Lab at Cornell University. While there, he did exciting research that suggested eating healthier and encouraging others to eat healthier was as simple as providing subtle hints and nudges towards the preferred behavior. He’s perhaps most famous for the claim that if you take a larger plate to the buffet, you’ll eat more than if you were to take a small plate. For a long time he was easily the world’s most prominent voice on food psychology. He published bestselling books. He was constantly interviewed by national media. He even had major influence in the government. He served for two years as the Director of the US Department of Agriculture’s Center for Nutrition Policy. It’s fair to say that Brian Wasnick affected how millions of people prepared, presented, and ate food. However, his reputation came crashing down from 2016 through 2018. It was discovered, partly through an accidental confession by Brian Wansink himself, that his papers were frequently based on shaky and perhaps outright fraudulent data analysis. The scandal was so bad that Cornell determined that Wansink had committed scientific misconduct and removed him from all teaching and research positions. This is a story about what happens when someone treats science like a business with the goal of gaining the most influence and media coverage, rather than a project of gaining an empirical understanding of the real world. Subscribe for $5 a month to get an extra episode of QAA every week + access to our archive of premium episodes and ongoing series like PERVERTS, Manclan, and The Spectral Voyager: https://www.patreon.com/qanonanonymous Theme by Nick Sena (nicksenamusic.com). Additional music by Pontus Berghe. Editing by Corey Klotz. www.qanonanonymous.com

In 2012, a Finnish graduate student landed an enviable position for any young academic. She earned the privilege of working at the world-renowned Cornell Food and Brand Lab, which was led by the charismatic Brian Wonsink. Wonsink produced mountains of scientific research about people's behavior around food. With this research, he advised individuals, corporations, schools, and governmental organizations on encouraging better eating. And this phase could be seen on every major news and talk show in the country. Soon after the grad student arrived, she discovered how the lab was able to publish

so many papers each year. Once they handed her a data set from an online weight loss program run by the lab, and a brainstorming session, once they proposed about six papers that could come out of the data. But when the grad student examined the data set, there wasn't enough information to support the the conclusions he wanted. You can't just pull a significant finding from an existing data set because you think the finding would be cool. She eventually used the data to draft the paper she was comfortable with. It was simply a straightforward review of how the program worked. Wansink was fine with the paper because it wound up in a notable journal, the

Journal of Medical Internet Research. Though Brian Wonsing was listed as a co-author, he had a little to do with writing. His few contributions made her uncomfortable. She later told the reporter this about her experience. Like he hadn't really looked at the results critically, and he was trying to make the paper say something that wasn't true. That's when I started feeling like, this is not the kind of research I want to do. She was right to feel uneasy.

Unbeknownst to her and the rest of the world, within six years, Brian Wansink would become the subject of one of the most notorious scientific misconduct scandals of the 21st century, casting doubt on the most high-profile behavioral food science ever published. I'm Travis Vue, and this is Trickle down, a podcast about what happens with bad ideas flow from the top. With me are Julian Field and Jake Rockatansky.

Episode 18. Viral Science I wanted to do an episode on modern scientific misconduct. And man, there is a lot of topics to choose from in that category. Even if I were just to limit myself to research psychology, the field is currently embroiled in a reproducibility crisis. And this is where like many widely cited and high profile studies can't be replicated. And that's bad because in good science, you're supposed to be able to find roughly the same result when you conduct the same experiment over and over again. But I was really drawn

to the story of the rise and fall of Professor Brian Wansink. He was the head of the Food and Brand Lab at Cornell University. He did really exciting research that suggested eating healthier and encouraging others to eat healthier was as simple as providing subtle hints and nudges towards the preferred behavior. His experiments supposedly examined the cues and conditions that make us eat the way that we do. According to the research, he published little things

like the size of an ice cream scoop and the way something is packaged, whom we sit next to, all influence how and how much we eat. He's perhaps most famous for the claim that if you take a larger plate to the buffet, you'll eat more than if you were to take a small plate. And so simply having smaller plates that is inferred can actually help you cut back on calories. Well, not true.

Not your experience. You go back for seconds. Yeah, there was a place in Chicago called Flat Top Grill. I don't know if they still have it anymore. But it was basically you could put whatever meat you wanted on the plate. You could put whatever veggies. And then you would stick a little flag in it to let the chefs know how to prepare it. It was like make your own stir fry.

And it came in these tiny little bowls. And the reason that you would go to the restaurant is because you would, could have as many bowls as you want. So me and my friends, we would get two, three, four bowls. I mean, you know, this was, we were eating quite a lot. Interesting. That's probably reproducible. I also would like to present my own findings, having two plates. Oh, yeah, double fissing plates. Didn't consider that. People forget that one. For a long time, Brian Wonsink was easily the world's most prominent voice on food psychology. He published bestselling books. He was constantly interviewed by national media, even had major influence in the government.

He served for two years as the director of the U.S. Department of Agriculture's Center for Nutrition Policy. He was responsible for the oversight of the 2010 dietary guidelines for Americans, the website MyPyramid.gov, and various other food-related programs administered by the USDA. He and the people he worked with were responsible for hundreds of published papers, many of which were cited under the Obama-era Smarter Lunch Rooms Movement. in American schools. So this was a program to reorganize lunchrooms in schools like around research published by Brian Wansink. About 30,000 schools participated. That's about

one-third of all the public schools in the entire country. It's fair to say that Brian Wansink affected how millions of people prepared, presented, and ate food. However, his credibility came crashing down from 2016 to 2018. It was discovered, partly through an accidental confession by Brian Nwancing himself, that his papers were frequently based on shaky and perhaps outright fraudulent data analysis. A group of data detectives started closely examining his papers, and they found that they often relied on a technique called pee hacking. Common way to do this is by combing through a data set to find correlations between variables that are just statistically significant enough to create a paper that gets published in a peer-reviewed journal. And this

frequently leads to false positives. The scandal was so bad that Cornell determined that Wan Sink had committed scientific misconduct and removed him from all teaching and research positions. So he broke the very basic rule, and I'm not a big science understander that correlation is not causation? Yeah, that's basically it. Is that, yeah, so correlation is causation, but if you look through a data set over and over and over again, you can find correlation that look, that look meaningful enough to get published. That's basically what he cared about. That's so cool. Oh, my God. So he, yeah, he broke like, just the entry-level thing. Fantastic.

One way that you might phrase it is that Cornell and most of the rest of the scientific community believe that the purpose of science is to serve mankind. Brian Wancink, however, seemed to regard science as some kind of dodge or hustle. His theories were the worst kind of popular tripe. His methods were sloppy, and his conclusions were highly questionable. Oof. Travis is coming for you, buddy. He's good. Let him cook.

This is a little treat for Jake. This was a Ghostbusters quote. So this is a story, but what happens when someone treats science like a business with the goal of gaining the most influence of media coverage rather than a project of gaining an empirical understanding of the real world. So Brian Wonsink was born and raised in Sioux City, Iowa. He had humble origins. His father, John, worked in a bakery. His mother, Naomi, worked as the secretary for a county attorney. Wonsink spent summers at the northwest Iowa farm of his aunt and uncle who raised corn,

a few hogs and chickens on their 140 acres. Wonsink has said that his fascination with food was born on that farm. In an interview that was later done by the National Institutes of Health, he said that as a young boy, he was curious why people bought food and some other people didn't. I grew up in Iowa, which is total farm countries, and all anybody does they work with food. And selling vegetables growing up as a little boy, I was always amazed at why one house would buy everything I had in the wagon. And the other house would look at me like I was Karen Kryptonite. And it really sparked this curiosity.

You know, why can you take something that the smartest person you know in the world cannot explain why they ate a salad instead of soup this morning? Or why they ate one breakfast cereal instead of another. In my lab at Cornell in the food brand lab, what's acknowledging is that there's a lot of mindless habits that get formed that can easily be changed, not by education, but by essentially changing the environment. Cryptonite, like, that they would, that they're supermen? It doesn't understand kryptonite, but also, like, what makes it a food and brand lab? Like, why brand? There's something, let me tell you something about that word. Because it was part of the business school.

Yeah, exactly. Awesome. So its original goal, they had a strong connection with the food business industry. Oh, interesting. Wow. And so the goal was to basically figure out why people eat so that, you know, you can create more successful food products. Yeah. Fellas, we need to create an easier Mac. Easier Mac and cheese.

It's fairly easy now. It could be easier. Now, listen, we know the Big Mac is killing people, but what if it was a bigish Mac? What if it was three patties instead of two? You'd be eating more meat, more healthy stuff. What if we put it in a smaller box? Hey there, you've been listening to a sample clip of Trickled Down. This is a side project that I've been working on.

It's a 10-episode series about misinformation and bad ideas that flow from high authority sources. I think it's fascinating. And, I mean, it's a way for, I guess, me to explore the way people who should know what they're talking about don't always, actually. I'm not going to lie, some of it's kind of a bummer. But if you're anything like me, that's actually more of a reason to dive into the subject matter. Like with the premium episodes of Q&N anonymous, all the episodes of Trickledown are available to people who support us through Patreon. Still the same five bucks a month. Double the extra content, same price that we've been doing since 2018.

We are inflation-proof. Inflage proof, inflage proof, inflage proof.