Plain English with Derek Thompson - What Experts Really Think About Smartphones and Mental Health

Episode Date: June 4, 2025

I'm very concerned about the relationship between smartphone use and America's mental health crisis. But many researchers don't see things my way. They insist that there is little to no empirical data showing that smartphone and social media use drives up anxiety or depression. So what’s the truth about smartphones, social media, and mental health? That’s the question that the NYU researcher Jay Van Bavel set out to answer with his collaborator Valerio Capraro. They took dozens of claims about smartphones, sent them to hundreds of experts in the field, and asked them if these claims were probably true, probably false, or unknown—and why. The result was a massive survey, one of the largest of its kind in the history of psychology. Today, Van Bavel joins the show to tell us what he found, what surprised him, and why his consensus survey made so many researchers so angry. If you have questions, observations, or ideas for future episodes, email us at [e-mail]. Host: Derek Thompson Guest: Jay Van Bavel Producer: Devon Baroldi Learn more about your ad choices. Visit podcastchoices.com/adchoices

What's up, everyone? I'm Nora Prynciotti. And I'm Nathan Hubbard. And we're coming in like a wrecking ball to announce a brand new series. That's right. It's every single album, Miley Cyrus. Deep dive with us into the career of one of our most creative and confounding pop stars. We're starting, of course, with the best of Hannah Montana. And ending with her brand new album, Something Beautiful, in June. And don't forget about Miley Cyrus and her dead pets. We certainly will not be doing that. So listen now on Spotify or we're all. you get your podcasts. Before today's show, just a quick scheduling note, this month's podcast publishing schedule is going to be a bit jostled around because my family and I are moving back from Chappahill, North Carolina to Washington, D.C.

And to all of our friends in North Carolina, a big, warm thank you for a fantastic year in the Research Triangle. To all my friends in D.C., I cannot wait to come back home. And for all you listeners, as far as scheduling goes, some weeks like this one will only have one episode, but we will be back to our normal twice a week schedule very soon. Today's show is about the expert consensus on smartphones, social media, and mental health. Here are a few things that I personally think are true about smartphones and social media. Number one, I think high use of smartphones is common among young people. Number two, I think this leads to elevated rates of anxiety and isolation among young people.

And number three, I think many adolescents and, frankly, many adults would be better off if they use their phones and social media less. I really do believe these claims, many of them popularized by Jonathan Heights's massive bestseller, The Anxious Generation. I also want to be right. And the truth is, in the last few years, many psychologists have made an even stronger claim. They say everything that I just told you is wrong, totally unsupported by evidence, and that actual experts in the field have concluded that social media is in fact not harming adolescent mental health. In a 2024 paper entitled, How Should We Investigate Variation in the Relations between Social Media and Well-Being,

researchers concluded that, quote, most scholars have reached consensus, concerns about general social media use don't seem warranted. Another 2024 paper by Candice Odgers put things even more startly. Quote, hundreds of researchers, myself included, have searched for the kind of large effects suggested by Jonathan Haidt. Our efforts have produced a mix of no, small, and mixed associations. Our time is being spent telling stories that are unsupported.

by research. End quote. So what's the truth here? What's reality when it comes to smartphones and social media and mental health? That's the question

that the NYU researcher Jay Van Bavel set out to answer with his collaborator Valerio Capraro. They took dozens of claims about smartphones, sent them to hundreds of experts in the field, and asked these experts to say on their own,

whether these claims were probably true, probably false, unknown, and why. Their work created this incredible document, a massive and frankly unprecedented survey of the field of psychology and sociology to answer this all-important question, what the hell is going on between rising mental health crises and rising screen use? Today's guest is J. Van Babel. We talk about how he designed this study, where experts agree, where they viciously disagree, why this paper was instantly controversial in the field, even though, ironically, it was trying

to reach a consensus. And maybe most importantly, what parents and teachers and all of us should do with this new treasure trove of information. I'm Derek Thompson. This is plain English. Jay Van Babel, welcome back to the show. Thanks for having me. Listeners might remember that you starred in one of my favorite episodes of this show,

The Four Dark Laws of Online Engagement. That's a show that I think about all the time, about the ways that the psychology of user behavior on the internet might be driving all of us crazy. Before we get to this, I thought, really fascinating piece of research. Just remind folks who you are and what you do. Yeah, my name is Jay Van Bavel. I'm a professor of psychology and neuroscience at New York University. And I run the Center for Conflict and Cooperation. And one of the main themes of the research that I do,

I've been doing for last decade, is the role of technology, especially social media, in society and how it fosters conflict or cooperation between individuals, all the way up to societies and between countries. So tell me about this consensus project. Jonathan Haidt, called it the largest ever expert survey. I wasn't entirely sure if he was talking about literally all expert surveys in the history of mankind or maybe largest ever expert survey in modern psychology. Maybe you can unpack some of that language. But what's the motivation for this thing and how did it come together? Yeah. So it's an interesting story. My colleague, who's the lead author in the paper, Valerio Caprero, I've published a couple other papers with him, including on technology and

AI. And we were both invited to write papers for this special issue on the impact and harms of social media. And he emailed me and asked me if you wanted to team up and write it together. And initially, he suggested maybe we could get a bunch, a panel of kind of like leading experts and make it a consensus of what we all believe. And when he initially suggested it, he had actually suggested John Haidt join us because he had, you know, the number one bestselling book in the country on the harms of social media. And I responded in my very first email to him saying, John also has a number of prominent critics in the literature. So if you're up for it, it might be worth bringing on some of those critics and seeing if we can find what the consensus is

between a really diverse group of experts who've been studying and thinking about this topic. Although I warned him, I said, this is going to be really stressful. So I said, it's up to you if you want to take the lead on this. Well, keep going. I mean, there are 229 experts here. I mean, was it just like the longest CC email from the get-go? Or how did you find a way to, like, bring in the right kind of experts to answer these questions?

Like, I assume that there was some filter process. Like, it was psychologists and maybe sociologists, but it wasn't like, you know, hey, if you're an astrophysicist and you have my email and you're a concerned parent, you can vote on the mental health of adolescents in the Nordics. Like, there must have been a curation process here. So without boring people, because we're going to get to the meat of this in just a second, like, what's the methodology? How do you find an expert cohort of more than 200 people to vote on dozens of claims

about the effect of smartphones on mental health? Yeah, I mean, that's a great question. And it's one of the challenges of the paper is we originally started by building kind of a small group to just figure out what were the claims that we were going to start with. And we decided to start with the claims from John's book. And we invited a number of people from various perspectives on this topic, including critics of John, to be part of that kind of curation of what were the key claims in this literature. And we added claims that weren't from John's book, but were kind of additional claims we were thinking about. And then the people that had kind of were this

core group that had built this set of claims, then we started like posting on list serves and reaching out to our colleagues who we knew to be experts in the field, inviting them to participate in the survey. And basically what we said is that you can be an author in this paper if you do this. And I will say that this is different than other surveys, because not only did we get this huge list of experts together and ask them what they thought about all these claims, but then we went through a really rigorous and painstaking process of kind of going back and forth and reviewing all the papers in the literature to figure out, could we get a, you know, over 90% consensus or better on every single claim. And that meant we kind of,

in many of the claims, we kind of had to adjust it and adapt it to make a bigger umbrella about what people could get behind. I love the idea of an expert survey. I think this is a really, really cool way to, you know, get beyond sort of the false certainties that I think can infect certain fields. We've done episodes on this show about how false certainty and false conventional wisdoms can really poison science. And this is a really cool way to essentially make transparent possible differences of opinion. But it's also interesting because like we would never say pull scientists on clearly understood truths, right? There's no interesting like expert survey on like, hey, when hydrogen and oxygen come together, can they make water? Like, that's a cheap

point in a way. But I think it's important to point out that the very fact that this expert survey had to be done is its own signal as to the certainty that people feel about some of the claims being made about smartphones and mental health and adolescents. So can you just maybe unpack that a bit and say, like, what does it mean about the state of research that we need a sort of gut check, a temperature check on the state of expert opinion on these issues? Yeah, I mean, that's a great question. It's part of the reason I was excited to do this in the first place. So as you said, you've had me on this podcast before. And one of the things that we talked about at that point was that when you tune into social media, what you often get is the most extreme people with the most extreme

views dominating the conversation. And so if I'm checking the conversation on anything, whether it's like Taylor Swift fans or people talking about the NBA finals or about the impact of social media, what you're going to get is people who have really strong views speaking up and repeatedly. And my view was that the same dynamic was happening on social media in conversations about the impact of social media, which was that I kept seeing like people with really strong views about social media is going to destroying a generation or concerns about social media are a complete moral panic. And when I would talk to people, you know, not only other scientists in my lab or at conferences, but other people I would see in the field passing through my department,

when we get to conversations about the impact of social media, their views were often more nuanced. And I felt like the consensus of what I was seeing in the literature and hearing from my conversations was often not reflected in these extreme views of mass destruction versus moral panic. It was a little bit more nuanced. And so I thought we need a better way of kind of surfacing

what the experts actually think. And so that's one of the reasons I was excited to do this, because I felt like what we're getting online is kind of a polarized, fun house, mirror, distorted version of what the debate is. And there's actually an underlying consensus that is pretty solid but more nuanced. And I wanted to see if that would emerge in a different type of forum for surfacing these voices. And so that was, at least for me personally, that was my goal. And then I'll just say that one of the reasons for this, just historically, you said there's

many scientific questions where we would not do this type of thing for. But I'll put it in context of climate change. So, you know, for the last 40 years, there's been a lot of debates over whether humans have caused climate change. And the debates used to take this, you know, pro versus con style where the media would bring on someone, some scientists would say, listen, humans got to change what they're doing. We're causing climate change. And then they bring on some critic. And what that does is it creates in the minds of viewers that the idea that this is divided, debate. it's 50-50, there's deep uncertainty.

And climate scientists realized, like, this was not working for getting out what the consensus was around climate science. And so they started to do surveys on it. And one of the first big, famous surveys found that 97% of climate scientists believe that humans are contributing to global warming. And so it kind of gets past, there might be some critics who don't agree with it. But just to be clear, they're in that 3%. And so I wanted to see also if we could get some consensus around what

we think the real impacts of social media are and then kind of quantify, you know, roughly how many people are on board of them and how many people disagree rather than just having one person from each side yell at each other. I was not going to follow up with this question, but I did not expect you to make this point, and I love it, that the same way that our language and our communication is berserked by the Internet, the same thing can be true not only for rando screaming at each other on on Twitter and Reddit, but also among academics, that they are often subject to the exact same forces

of recognizing that outrage will go viral, and in-group versus out-group dynamics go viral, and high arousal emotionality goes viral, negativity goes viral. I think I summarized all the four dark laws of online engagement that we covered a few months ago. So why don't you just give me right here an example of where some of the louder claims

in this debate, became softened a bit once you asked a bunch of experts to reach more of a consensus around the science rather than outfit their claims about smartphones and social media for sort of perfect online virality. Okay. So I will say I'm just going to try to break it down for listeners in two waves. So in the first wave, we sent out this set of claims. And there were, I believe, if I'm remembering, 26 claims that we sent out to our panel of experts. And what we found in the first wave is that there was already consensus about some of the claims. So the single claim where we had the most consensus wasn't even about social media, but it's related to it.

And it was that sleep deprivation can reduce mental health. And we had overwhelming consensus. Almost 100% of people thought that that was probably true. And then at the opposite end, a claim that was highly disputed was, this claim. One third of college students prefer social media to not exist. And that was based on a recent paper that had come out suggesting that college students like social media, like Instagram and TikTok, but they say they'd actually pay money for those apps not to exist in the world. Well, that's just a single paper. Not a lot of our experts had actually read it. It's only with two social media apps.

And so there was a lot of, you know, only about a third of people thought that that was probably true. another roughly third thought that they don't know. And then there was about 25% or 20%, sorry, 20% who thought that that was probably false. And so our claims in terms of certainty from the starting point ranged from, you know, about 99% of some people thought that it was true to that one,

where it was only about 30, 35% thought that it was true. When I was reviewing the claims that had the most support versus the claims that were most controversial, Among the claims with the most support were things like social isolation is bad for mental health, sleep deprivation is bad for mental health, behavioral addiction is bad for mental health. And one thing all those claims have in common is that you didn't need to start studying that question in 2015. You can study sleep deprivation and behavioral addiction in 1995. In 1925, you could study these questions if you wanted to.

And so we had a lot of time to build a body of research. to really understand the nature of the relationship between, say, social isolation and a negative mental health. On the other hand, many of the claims that were fresher, like, for example, removing smartphones from our children until they reach high school

will improve mental health. We're a little bit more controversial, even though even there, there was a strong ratio of people who said probably yes to people who said probably no. Going back to Jay, you're comparing this expert survey to the climate change surveys

where the headline the New York Times could throw on it was something like 97% of scientists agree that man-made climate change is real. Is there a similar New York Times style or Atlantic-style headline that someone could put on this paper to demonstrate the degree to which scientists do or do not have some level of consensus

about new claims being made about smartphones and mental health? Like, what's the headline here that you might submit to an editor at the Times or the Atlantic to say, write about this paper, this is your headline? Yeah. So I think there's a couple things that stuck out to me when I saw the data come in. The first thing is that there is not the degree of consensus with social media as there is with climate change. Our science is not as evolved on this topic, and therefore there's not as much consensus.

And in fairness, we've only been studying this for like 16, 17 years, whereas climate science had been going on for a lot longer. We may get there. You know, we may do this. We may have a conversation again in 10 or 15 years, and we might reach that 97% consensus about social media, but I don't think we're there yet. So that's one conclusion.

The things we had strongest consensus about, we're actually not about social media. However, I don't want to suggest that there's no consensus about social media. even from our initial survey, there were far more experts who tended to agree with things that social media can impair sleep or can cause behavioral addiction

or can cause fragmented attention. There were vastly more people, experts who agreed with those, than disagreed with them. There were still a number of people who weren't sure, but the ratio of people who agreed to disagree on many of those claims

was roughly 10 to 1. And so there is a growing consensus among experts on some of those claims. But if we want to, and this is what we found in the next phase of our study, once we started getting everybody in conversation to see if we could get people to that 97% consensus level for most of the claims, it required much more nuanced claims that had a lot of caveats and were very balanced and nuanced. And so the final claims that we came to that had like 97% consensus or better were along the lines of, there is evidence that various measures of adolescent girls mental health have been declining since the early 2010s.

Well, that's not about social media. If you want to get one about social media, there is some evidence that screen time has increased and time with friends in person has decreased, especially in the U.S. That eventually got 95%. Or this has 76, 77, 78%. Well, there is evidence that heavy daily use of smartphones and social media can cause some sleep problems. The extent to which it causes sleep deprivation specifically remains unclear.

And so we were able to get a broader segment of our experts to agree, but it required those caveats and nuance for them to kind of sign off on that claim. I think what we should do now is talk about some very specific claims. Like, I think people have the big picture that there's overwhelming consensus about some issues. And then on other topics that are drawn from Heights book, for example, that if we delay smartphone use until high school, yes, there was still like a two to one margin of probably true versus probably false. But, you know, two to one margin is not the, you know, 33 to one margin of climate change is. is a man-made phenomenon, right? And so it's important to say all of these claims

had a positive ratio of probably true to probably false, but that ratio was really different across claims. We're going to get into some in just a second. But first, what do we make of the fact that some of the most prominent critics of Jonathan Heid's book are not named in the final list here, right? I think a critic could point at that

and say, why, Jay, should I take your paper seriously if you put together an expert panel on these questions that doesn't include feedback for the most prominent critics? Did you just not even ask them to participate in this project from the get-go? Talk about cooking the books. What is your response? Yeah, first of all, that's a great question. And I wrote actually in my news, the Center for Conflict and Cooperation newsletter. I went back once this paper came out because people were asking that. And I went back to my original emails with Valerio. And literally in the very first email, I sent Valerio when he invited me involved.

I suggested that we add people that are key critics. And then within the second email, we were brainstorming critics like Candice Audjors, Andrew Prisbalisky, Stuart Ritchie, and others. And we reached out and invited many of them. And in my newsletter, we listed kind of the timeline, A.B. Orbin as well. and the blow-by-blow of our emails to them in the very earliest stages, even before we had really settled on any of the claims. And so I just want to be clear that from the various origins of the project,

we had them in mind and explicitly discussed them and reached out to them and invited them from the earliest stages, even in the construction of the claims. So there was a conscious effort at every stage of the project to include critics. And I will also note that actually if you look at every claim, there are some critics who disagree with every single claim. And so we actually did include a lot of critics. And the final thing is that the Delphi process that leads to the final consensus

is a process that, by its very nature, requires heavy concessions to the critics. So we started with a bunch from the get-go of inviting critics. we were able to include a lot of them, as you can see by figure two of our paper for every claim. There were critics. And then we had a process that required all of our experts to share ideas and share papers back and forth and discuss them and converge on a statement that could get, like I said, 96, 95, 97 percent, 99 percent consensus. And so we really embedded criticism throughout this. And then I will also say in our preprint we have online, we have almost 100. pages of supplement where it shows the discussion in the paper shared with the critics,

and the lead author, Valerio, tried to address every single comment from critics wrote. So we bent over backwards to include critics. I would actually say that an alternative critique of our papers that once you see the consensus statements, some people might feel like they're overly watered down as concessions to the critics. And in fact, I will also say that we had some people who were strongly on the side of social media causing a lot of harm, and they took themselves off our paper at the last stage because they thought we had conceded too much to the critics. So, but this is probably the part of the challenge of any time you reach a consensus on a polarized topic, that you're going to lose some people

from both sides. And if you read my newsletter on the sausage making of this, that's exactly what happened, unfortunately. It's hard for us to know, did we lose an equal amount of both sides, or is the ratio of people who lost from both sides equivalent to what the underlying ratio distribution would be of people's genuine beliefs of experts? Unfortunately, I have no way of knowing that. But privately, we did lose people from both sides because we were not as extreme in the direction

that they would want. The nature of coalition building. I'm in Washington, D.C. right now. So this is certainly a city that understands that you lose things from both sides when you try to build consensus. I just want to really quick answer to this question. because I think it's really important.

Critics like Odgers aren't listed on this paper. Did you invite them and they dropped out? Did you not invite them? Or did you invite them and they participated but did so anonymously? It kind of has to be one of those three categories. So do you feel comfortable saying, did you fail to invite them? Did they say no or did they participate but insist on anonymity? Yeah.

So we originally started the original emails about this. launched this project in August 13th, 2012. We invited Candace Audgers on August 19th to 2024. So literally five days after the first email from Valerio to me. And Amy Orban as well.

But she declined. So we, Amy Orban declined and Amy Ogers had not replied. And so we followed up to try to by August 22nd. We were still waiting

for responses from her. And I don't think she ever accepted our offer, unfortunately. I'm sure some people are bored out of their minds by this back and forth on who got to be included on the list. But it's obviously incredibly, obvious to me at least, incredibly important to understand which experts finally made it. And it's very, it's interesting to hear from you that you lost both folks who thought, you know, maybe John Haidt doesn't even go far enough. Like, smartphones are basically cocaine on the one hand. And then you also had folks who think that too much is being made. of the screen effect on mental health.

And you lost people from both sides, and it's hard to know exactly how many you lost both sides. Jay, let's turn to some real hard claims here and find ways to communicate the expert consensus to parents, policymakers, writers. Let's start with Claim 14 in your analysis. And this is the claim that smartphones

and social media cause social deprivation. It's a claim that's very near and dear to my heart. I wrote this article earlier this year for the Atlantic called the Anti-Social Century that said that one of the most important problems in our society right now is that for a variety of reasons, not just related to smartphones, Americans spend a historically low amount of time in face-to-face interaction. Various measures of social interaction and social interaction are down and social isolation are up.

So, Jay, talk to me a little bit about how, what experts think, about the claim that smartphones and social media cause social deprivation? I would say this is one of the items where there is less consensus. So there was a ratio such that more experts thought this was probably true than probably false. I think the ratio was about four experts thought it was probably true for everyone who thought it was probably false at the outset of our study. However, once we dug into the data and we tried to find something where we could get 97% of people to agree on, it required a more nuanced statement that the vast majority people could sign on,

such that the statement changed to the strength and even the direction of the potential causal relationship between social media use and social deprivation likely depends on various individual and social factors. And one of the reasons I would say that is because some of these effects of social media are small, and when you're dealing with small effects, that means that there's a lot of variation. It means they depend on the type of kid that you're dealing with, the context that they're in. And so there's more variation,

and that means it's harder to have strong definitive claims that everybody can agree on. Yeah, I thought that the discussion here was really nuanced and wonderful. You know, experts who believed that smartphones and social media use are causing isolation, pointed to several meta-analyses that have shown pretty clearly this positive relationship

between social media use and loneliness. This research, by the way, did not appear in my original article. I try to be careful to not hang to the antisocial century too much exclusively on the phenomenon of smartphones, because as your paper indicated, I think a lot of this research is really,

young. And also, and I love this point as well, the truth is that smartphones are really complicated. Like, some people are clearly using their phones to connect with friends, to connect with family, to talk to their grandmother who lives 100, 300, 3,000 miles away. You have to use your phone for that. And so there's many ways in which we have relationships that are more connected than they've ever been because of our phones. You also pointed out in this paper, Another point I loved, which is that, you know, social media use can facilitate relationships that didn't exist before social media. You know, for some people, it's like, you know, you live in Providence, Rhode Island,

and you're a fan of the Cincinnati Bengals. And it's so much easier to follow the, you know, Bengals Nation from Providence, Rhode Island. This is a Mark Dunkelman, a classic. It's his story to me. It's so much easier to do this when you have social media, when you have group chats. you know, LGB-BtQ youth can more easily connect over their experience because of their smartphones. But at the same time, because these phones

take us out of the physical world and bring us into a virtual world that can give us both social connection but also social disconnection. It is a very, it's a naughty thing to think through because of the complicated relationship between smartphone use and social. isolation. So I really appreciated the nuance that you went to in this paper. Anything that you want to

clean up or add to in this category of social deprivation and smartphone use before we go to the second major claim here? Yeah, I mean, one of the things that I think that's cool, and I don't even really take credit for this, these were the discussions between the experts. And this was part of the interesting thing about putting them in this process of discussing with each other what the evidence was that was convincing to each side and then seeing what they could, if they would update their beliefs and the consensus statement. But I will add to the comments you just made by saying there's also research showing that how you use social media changes its effects on your well-being and the relationships you develop from it. So if you're using it passively and just doom scrolling,

that's the type of social media use that tends to be negative. If you're actively using it in a relationship-building sense and relationship maintenance sense, that's positive. And I even see this with my own kids. So I have a 13-year-old daughter when she's doomed. scrolling on TikTok, that's when she's just curled up in the fetal position with the light off in her room and it looks really sad. But a lot of the time where she's using her smartphone is like she's on a call with her friend and they're like, she's at breakfast, but she brings her friend in and they're talking the whole time or they're doing homework together from their own houses and they can help each other through homework and stay in touch and plan their social events.

And so I'm able to see it through also the eyes of teenagers. And I can see in each of them the ways that they do it that are antisocial and the ways that they do it that are pro-social and so a lot of my conversation with them is trying to nudge them to use it in healthy ways and get them to stop using it in unhealthy ways. And I think that that's the thing is that you're seeing a lot of times

what we talk about are the consequences of the negative. We don't often talk about the positive. And the negative are things are easy to see. It's like they can call their grandparents in Canada. But also when their grandparents come here and my kids are on their phone and not engaging with them, it's really demoralizing to the grandparents.

So you can see, you know, it's the in-person relationships that are often eroded. Well, you're able to maintain the relationships that are more distant, and you feel a closer connection to those people through this technology. And so I think that there,

hopefully will be more nuance about it, but hopefully we can push for, you know, the development of technology that's like a little bit healthier. And it's like anything, you know, there's a period of journalism where there was yellow journalism. It was really unhealthy.

We developed professional, norms and standards. And we do this with all kinds of technologies. The car, incredibly dangerous. And we've, you know, we had to implement seatbelts in the 1980s to reduce mortality. Then airbags and all these, you know, drivers tests and so forth to optimize the use of technology in a healthy way and minimize the downsides. And so I think that's where the conversation hopefully will go as we get more and more evidence from research like this. And John Haidt wrote about this aspect, I thought, really beautifully in one of the later chapters of his book, The Anxious Generation, where he talks about how smartphones and digital technology are a disembodied, asynchronous experience. And that might sound like overly technical. But the idea is smartphone use often takes us out of our bodies and out of our locale and place us in a virtual community.

that we are not physically present in. And that's bad in many ways, I think. It's bad when you're not plunged into the physical reality of the present. There's a ton of clinical psychology that says that plunging yourself into the present, having really overwhelming sensory experiences can be really good for combating anxiety.

But it's important to say, just as you did, that being elsewhere, joining a virtual community, doesn't just mean leaving your body in a negative, way. It can mean joining a friend's group that is maybe more progressive than your family, more progressive than your church, more progressive than your local school. It gives you access to affinity networks for the Cincinnati Bengals or Taylor Swift that allow you to participate in a culture so that even if your parents don't care about football or Taylor Swift, you can still engage in a culture that privileges what you value. And there's something beautiful there. And so I think that it's

really lovely the way that you put it and the way that John put it as well, that like, there's pluses and minuses to this disembodied nature of digital technology. And we're, we're still, because we're in the middle of this relatively novel experiment, we're still trying to figure out how do we get the best of disembodied asynchronicity while, while not suffering the worst of disembodied asynchronicity, which I'm not trying to be overly technical, but that really is, that really is it, I think. These things take us out of ourselves, and sometimes that's freaking awesome, and sometimes it's absolutely terrible. Last thoughts there before we go to the second major claim? Yeah, I think the other thing, and this is one of the things John's

book has done, I think. I was in Canada over the holidays last summer, and I was at a friend's house, who's an academic, he was a former roommate of mine, and I was at the talking, his parents were there, and his wife was there, and all of them had read the book, and they were all talking about, it. And they were coming from very different levels of education. And it was a conversation starter, is that people are starting to think about this in a more formal way, about the impact of this technology and the tradeoffs it's caused. I think the problem with the technology a little bit is it's insidious. And roughly 5.5 billion people now have social media, more have smartphones. And I remember reading that we're on it for about two and a half hours a day per person,

which adds up to 5.3 years of our life, or 5.7 years of our life, I think, is going to be on this technology. And Jay, it's way more for teens. It's four, five, in some cases, six, seven, eight hours for teens. Yeah, and it's off the charts for teens. And it's only going up. And it's going to have bigger impacts on the next generation.

And so I think that's why we need to figure it out now because it is having all kinds of impacts. and we don't fully understand them yet. Well, you've perfectly on-ramed the next claim, which is claim 20 of your paper. Please turn to page 105, Jay, and let me know when you're there. The language that you polled is, quote, among adolescent girls,

social media increases exposure to other people displaying or discussing their mental disorders, end quote. I think this is such an interesting claim to put before an expert audience, Does smartphone social media use increase young people's exposure to mental disorders? Walk me through what you found here. Yeah, this was one of those claims that we had a huge amount of consensus with initially that there was roughly a 40 to one ratio of experts who have believed in this versus disagreed

with it. And then there was a small percentage of people like 10 to 15% who didn't know. And so this is one of those issues. that there's a lot of concern about how is this if we're exposed to this discussion about mental health disorders, what are the impacts going to be for society? And so this is one of those issues, I think, where I think there's a growing consensus that this is true that we're getting this exposure. There's less of a sense of what are the, less knowledge about what are the consequences of that. But this is one where our experts,

actually did tend to agree with it. And even after we went through the Delphi discussion, we ended up getting consensus around a statement that was pretty similar to our original claim, that our final consensus statement was social media may contribute to increased exposure to mental disorders. And 96% of our experts agreed with that. So that's, that is kind of a bit of a wide open area, but what the impacts of that are. But it seems there's a growing consensus that that's happening. Can you go one level deeper, Jay, to explain why this might be the case? Like, if you started the clock in 2007, 2008, and you said, we're going to take this little

screen and we're going to put it in people's pockets and we're going to put Facebook. Remember that thing that was like, you know, Facebook's for college campuses, but now everybody can kind of use it? We're going to put that thing on the pocket screen device. and it's going to increase young people's exposure to the world's mental disorders. I can see how that prediction might have been made by some people,

but it is kind of interesting that among the many things it increases people's exposure to very specifically. And here, as you said, there's enormous expert consensus. It increases people's exposure to the mental disorder of the world.

What's going on? there. So now I'm speculating. Yeah, I'll speculate. But I'll say a few things. So there's a sliver of this that is people trying to advocate for greater awareness around mental health issues. And I see this all the time on pretty much every platform from Instagram to LinkedIn about how we should talk about this in workplaces. And so that is a sliver of this. So it's not all just negative discussion about this. Sometimes it's about connecting people with resources. There are often campaigns about awareness about certain mental health issues. I would say another thing, it might be that a lot of the users, as you said, are younger

people. And what you're seeing also is a rise in mental health issues among younger people at the same time. And there's a bit of an unclarity about what's causing what. But if younger people are experiencing this as a greater extent, they're going to be talking about it more and it's going to be getting more coverage on social media than it might be in traditional forms of media. The other thing is that I think it's also changed the norms around what is acceptable to talk about.

That it used to be taboo or stigmatizing to talk about your mental health issues. Now you can talk about them in a video on TikTok and get attention for it or get support for it. It can be both of those might be useful for you. And it can actually create incentives for other people to talk about it. If they see someone talking about it and they're going viral or getting a lot of support or growing, growing a followership, that provides a social signal to others that that is incentivized and rewarded and normative for them to do it. And so you can get cascades of trends of what people talk about based on those types of social media dynamics of rewards and incentives. And so I also would

argue that another thing is it is really well designed for social media and that it's often extreme and interesting. It's things that are outside of the norm. of our day-to-day experience. Mental illness is almost by definition. In psychology, we often fit that under abnormal psychology. So it's almost by definition counter-normative, even though it's quite common now. And the way people talk about are often talking about episodes

and experiences they have that are counter-normative. And so it's interesting to people to hear about it, and that is attention-grabbing in this attention economy. And so those types of messages and stories are often going to get more. engagement and capture more eyeballs and be privileged in the algorithms on these platforms. And so I think that might be part of it is just that this content is well designed to maximize attention in the attention economy.

That gave me a really weird thought that I never really had before, which is that if the fact that mental disorders and various kinds of mental illness are novel and interesting, and then algorithms promote the spread of novel and interesting ideas, thus maybe increasing the prevalence of mental illness through, among other things, social contagion. I don't think anxiety is mere social contagion, but I think maybe among other things we're looking at a social contagion here. Well, what happens with social contagion is that things that were, one,

new and interesting, become old and commonplace. And things that are old and commonplace don't go viral online, which suggests in a weird way that the social phenomenon of mental illness could theoretically hit some threshold point where it no longer goes viral in the same way that it used to, and therefore the sort of aerodynamic virality of anxiety content, mental illness content declines, this stuff goes less viral, and so the social contagion element

sort of burns itself out. Could something like that life cycle presumably happen? Could it presumably be happening now? Is the story that I told utterly unempirical, or is it like, yeah, it's a just-so story, it's stylized, maybe it's true, maybe it's not?

Yeah, I mean, the one way to look at it, at least on social media, is like mentions of mental health language. So let's say you could look at, say, language discussion about depression and see if it kind of rises to a point and then starts to plateau because it's no longer novel and interesting

and attention-grabbing and eventually becomes commonplace and boring. And then this is a little bit of what people are constantly doing is they're constantly fighting to create novel content that will capture interests and capture, be interesting, a capture tension. And so I would not be surprised if that happens. If you're seeing like a contagion effect of mental illness, you might be imagined that that could take two forms.

One is that, let's say everybody's talking about their anxiety, that could be something that increases everybody's anxiety. For example, there might be social information that the world is anxiety producing and maybe I should be more anxious. But another thing it could just be doing is not increasing individuals' anxiety, but increasing their self-disclosure of it as a way of fitting in or as just a norm of expression. And so this is why I think from a mental health perspective, there are concerns about like over-diagnosis of some of these issues. Just because people are talking about them doesn't mean they're actually observing all the symptoms that would normally have a psychiatrist to diagnose them according to like the DSM.

So I think that that's another thing we have to distinguish between when there are these like social contagion effects. Some of it might just be people expressing that they have this mental health issue. They might not actually have it in a true clinical sense. All right, Jay, final claim here. Let's deal with solutions. Jay, why don't you turn to claim 24? And this is one of the most famous claims about the dangers of smartphones.

and social media for young people. This is the claim that delaying smartphone use until high school will probably improve mental health. That's the claim. In the final analysis, 68% of the experts said this claim is probably true, delaying smartphone use until high school will improve mental health.

11% said probably false. And what you guys do in this paper, which is so cool, is you list the studies that give sucker to both sides. So, for example, there was a large-scale survey of more than 27,000 individuals between 18 and 24 that found that indeed, mental health outcomes are associated with later age of first smartphone

or tablet ownership, with that effect being most pronounced in females. So that seems like a really, really powerful suggestion that you want to delay smartphone use among your children. But then there's a 2021 study. that found that the age at which individuals obtained their first smartphone had little to know, there's a quote,

little to no predictive value for later well-being outcomes. Jay, what should parents make of this? Like, translate this expert outcome and the content that you amassed through this Delphi process to communicate to parents in clear language what they should make of the,

level of consensus around delaying smartphone use until high school? I would say that the area where we found some of the least consensus is in the policy prescriptions about what to do. And in part because the evidence is mixed or they haven't been tested very well, and therefore we don't yet know what works or what doesn't. If parents want, I think like this is where the science is not yet definitive enough. to tell parents what to do. I think parents and schools are going to have to figure out what to do in the meantime, because sometimes we have to make policy decisions before we have solid evidence, and then get as much data as we can. So for example, I'll say another area where we need more

evidence is whether like banning phones in schools is going to be helpful. This is actually a study that for the last six months I've been trying to get money to run because, like for example, in New York State, we're about to ban phones from schools. And so what we want to do is do an A-B test, like a clinical trial where we randomly assign this policy to schools and then measure the outcomes in kids. The problem is a lot of policymakers don't want to control condition. They either kind of want to let the status quo be and not interfere, or they want to get the phones out of schools and not test if it works.

So that's an unfortunate thing for us as scientists, because we really do need that evidence to get consensus. So that's a little bit. We're about a little bit of an impasse with some of these policies because of that. Big picture, Jay, what surprised you? What should surprise someone who follows this debate closely about the outcome of this expert consensus survey? One of the funniest things that surprised me was how worked up people got about this when we

shared it on social media. People got, it was so polarizing and people got so upset about it. And so that was something that was like pretty surprising to me. I thought once we have this consensus, it's going to kind of like lower the temperature of the conversation around these. And then it will kind of provide an anchor for future discussions because there's a ton of papers cited and nuance and tons of experts included on this. But it didn't seem to do that at all in the short run. It just people ranted about the paper and what they liked or didn't like. And so that was one thing that was pretty surprising to me.

and my other authors. We're so down there. You're not going to get away saying that there was this volcanic explosion of outrage and give me no details or context about the substance of that outrage. What were the most common objections? You can pick serious ones.

You can pick silly ones. But like what stuck to you? Okay. So the best objection that was substantive was just criticisms about did we have a representative sample of experts? and I got comments from people, again, from both sides, who one side felt like we did not include enough critics.

Other people were frustrated that we did not include enough people who think social media is harmful. And so both people from the extremes of their views on social media were seem to be equally mad about the conclusions. So that was one thing that I observed. Maybe I should have known that. And then the other thing, that people really got mad about and maybe I own this is I shared it and briefly on social media and really on blue sky in particular, there are a lot of, that seems to be where most of the

skeptics of the harms of social media live, which is ironic because most of them have migrated from X because they think it's harmful. But they now, a lot of the academics are there and that's where a lot of the really critical people are. And they thought I shared it. it in a way that was not sufficiently nuanced and got quite mad about that. And so that was something that I was very surprised by. And it got like media coverage in two science journals. And largely the media coverage was just about the social media backlash or reaction. Very little of it was about the substance of the actual paper consensus or process of the

study. So that was like pretty surprising to me. I've had a lot of papers and that's only happened. And I've had up, like, I've published like 150 papers. That's happened maybe like once or twice at that level of, like, backlash to a paper. So that was surprising. Well, it's, I mean, it's impossible to react to that story and not feel like there's a profound irony in the fact that you ran this consensus survey that achieved what you thought was a kind of local consensus within the bounds of the paper. And then when you published it and it made contact.

with the broader world, it might have actually further polarized the debate. I mean, does that, what does that tell you about either social media, the nature of the debate over the relationship between smartphones and social media and mental health? Is this just an area that is so highly politicized that any effort to sort of even identify and make transparent some kind of consensus is doomed? Or is there some other message here in the reaction? Oh, I think I have a lot of thoughts on this. I think, again, that there are challenges when, this is an ambitious project

to bring all these people together and find a consensus. And people outside that project can get frustrated for various reasons. They don't like the direction it went. Some of them might feel threatened because the consensus challenges their research. So I think that's always a risk. And again, on social media, the loudest people with the most extreme views are going to be the most vocal. There might be a huge group of normies with nuanced views who are actually like, actually, yeah, this kind of matches what I think from my reading of the research.

I actually suspect there's like a quiet majority who probably actually aligns with this. Just because of my research on social media tells me that's probably what's going on. But I do think it's going to be hard to get. get consensus around polarized issues on social media. And that includes whether you're talking about like what the next, the future of the Democratic party should be or whether you're talking about what the consensus on social media is. And it's a problem of this as like we talk about social media. You know, for example, Twitter at one point people talked about like the public square.

It's not structured to be a useful public square. It privileges people who are extreme and loud and aggressive. And so consensus is going to be hard to assess in platforms that do that. And so I think that that's intrinsically a problem about getting consensus. And I also think I saw this happen with the vaccine debate. The vast, overwhelming majority of scientists support vaccines, but you had a small number of anti-vaxxers, and they're in the minority of even citizens.

And they just got incredibly vocal on social media and built a following and changed the dynamic of the discourse around vaccines. And now we have radical changes at the highest levels of government, of vaccine policy, in part because of that, I think. And so I think that this is intrinsically a problem of having a platform that amplifies extreme views. And I think that is part of the problem. And we're trying to get consensus around what it does to people,

but it's hard when the platform itself is not designed to surface consensual or nuanced perspectives. Jay, you know, I don't know you that well. We've spoken for a total of like, I don't know, three hours in your life and my life. I want to believe that you're a good faith guy. You seem like a good faith guy to me.

I feel like when a group of social psychologists try to get 200 field experts together and reach something like a consensus on 20 to 26 complicated claims about human nature and its relationship to technology. I think that's really hard. And I bet you probably worked really hard and sent a lot of emails and you've made a lot of people very upset about the outcome. How does that make you feel about either the intersection between science and social media

today or the nature of this field itself and how deeply politicized this issue so clearly is that, you know, as opposed to another issue, I said, you know, hydrogen, oxygen, that's very apolitical, maybe even unlike something like climate change and climate scientists, this is just an issue where people just fundamentally don't agree about some of the most important claims being presented and questions being asked. Does it make you despair, or do you feel like it helps you see more clearly the shape of a debate where there fundamentally is no consensus? Let me say that when I saw the results, I was actually really optimistic because they aligned very closely with what I believe in my own reading of the literature. And that is that there's

actually an emerging consensus on a lot of issues at a, you know, at a 10 to one ratio of experts believing versus disbelieving something. And that's always reassuring that when you're aligned with like 90 to 95 percent of your colleagues versus the ones who are opposing you, it kind of means maybe you're reading the literature the same way. There's a shared reality there. And that was the case with like over half of the claims. So I think in that sense, I was hopeful when I saw it. And the. And the. claims where there was less consensus were areas where I am actually trying to do research myself, because I don't think there is knowledge, which is a lot of the policy claims about like what can we do

to intervene and what are the impacts. And then I was also optimistic because we got so many people and the conversations actually that they had with one another were so constructive and so nuanced. In some sense, that's like the best of academia, all these brilliant people reading this research, sharing it and trying to come to some consensus. And so all of those things made me incredibly optimistic. The thing that just made me demoralized was the discussion around it on social media. And I should have known better because I study the dynamics of that. Why I got into studying social media, I'll tell you story, was it was about 12 years ago and I was studying academics, mainly psychologists, talk about the replication crisis. And I noticed how extreme and moralized

they got. And people just turned into us versus them in two camps. And that was what led me to study polarization and politics online and how it polarized us. Because I'm like, if it can do it to academics, talking about research methods. And in a way where I would think this person was dead set against me, a colleague, by the way they talked online, then I would talk to them over email at a conference and realized they were totally reasonable, and I agreed with them on 94% of what the issue was. And it was like the dynamics of the medium created this, us versus them,

and polarized rhetoric. And so I think I just see this over and over again, whether we're talking about politics or scientists, talking about social media now, which is just an incredible irony. And that is demoralizing. And it makes me want to spend less time on social media. And social media, it feels like getting more and more broken as it goes along. So the reason I'm studying it is because I'm fascinated by it and I'm a participant in it. And I get sucked into the same terrible behaviors that I observe other people doing. And I look back on it. I don't know how I did it.

but I think it's because of the design and the algorithms and the norms guide us to do certain things that might not be our natural way of engaging with other people. And I'm not immune to it, even though I have studied it for 10 years. And so I think that's the challenge of this technology. And we're all five billion of us are stuck in it now. Well, I have three takeaways from this conversation and the paper. Number one is that to the extent that there is a New York Times Atlantic headline that we're trying to write. This one might be too long, but the headline that occurs to me is, quote,

guy who studies how social media is an anger and outrage machine that promotes self-righteous, us versus them, dynamics. Surprised when social media turns out to be an anger and outraged machine that promotes self-righteous, us versus them dynamics. You know, this is one of those physicians take their headlines and I do not. That is prevention. Take their own medicine situation. Number two, you know, something can be depressing and interesting at the same time. Like a ton of things that I write about and talk about are depressing and interesting at the same time.

And it's just okay to recognize the reality, to see this reality clearly that the dynamics of social media and science communications can be depressing and interesting. I mean, both that layer of science comms is depressing and interesting

and I think the effect of smartphones on our life can be depressing and interesting at the same time. And then number three, one reason why I am optimistic, at least at the level that we're going to learn stuff,

is that it's not as if smartphone research is something that started in 2015, ends in 2025, and now we're just going to move on to something else. Like, we need a lot more research, and we're getting a lot more research. We need a lot more natural experiments and randomized controlled trials.

I think we're going to get them. We're certainly going to get a ton of research on, let's say, you know, one school district in a Texas city banned smartphones, another school district that has, has the exact same, you know, social determinants of health and demographic data doesn't have

the same policy. And then it's a natural experiment. We can say, you know, after a certain number of years, what effect did that smartphone and social media ban have on District A versus District B when everything else seemed more or less similar? And yeah, we're just going to, we're running a ton of policy experiments. And it's to John's credit that he wrote a book that wasn't just a chin stroker. He tried to change policy. and he might be wrong, right? Like, I might be wrong. I think John's right.

I might be wrong about the effect of smartphone and social media delays and bans on adolescent mental health. But I am curious to know that I'm wrong because ultimately I want to be right. And I think you feel the same way. I feel completely the same way. Yeah, thank you, Derek.

That sums it up for me. I feel like this ended with a little bit of person-in-person therapy. But honestly, I think all of my podcasts would ideally end with a little bit of a clinical psych, even for the social psychs among us. So thank you, Jay. Yeah, even the psychologists need it.