141 - Michael Schwalbe and Geoff Cohen: When Politics Trumps Truths

Welcome to the Sanford Psychology Podcast where leading psychologists come to share their most recent work. I'm Angie, and for this week's episode, I'm very excited to share my conversations with two amazing psychologists, Dr. Michael Schrobb and Dr. Jeff Cohen. Michael is a postdoctoral fellow at Stanford University, from which he also received his PhD in Social Psychology. Jeff is a professor of psychology and of organizational studies in education and business at Stanford University. His research examines the processes that showed people's sense of belonging and self-concept, and the role that these processes play in various social problems. In this episode, Michael and Jeff share with us their most recent work on biases we have when consuming political news. People were more likely to believe and share news that are aligned with their political beliefs compared to news that were actually true. You know when the highlights were all landishly fake. Without further ado, here's our conversation. Thank you so much for joining our show today. I'm really glad to have the chance to speak to not just one but two amazing psychologists on the show. Michael and Jeff, welcome. Today we will discuss a very relevant and timely paper titled "When Politics Trumps Truth, Political Concordes vs. Moracity as a determinant of believing, sharing, and recalling the news." So there's a lot to unpack in the title here. Michael, you are the first author. I wonder if you can help us unpack the title a little bit. What do you mean by "political concordes" and "political veracity"? Great question. And just thank you so much for having us here. It was really exciting. Through researchers on misinformation, tend to look primarily at two attributes of the news. And the first is, is the news true or is it fake? And by looking at the differences in how much people believe true news versus fake news, they can look at the effect of something they call headline truths. The other main attribute is whether the news is aligned with people's political views or not, but for researchers call the effective political concordance. So they can look there within subjects design. What's the difference in how much people believe or are willing to share headlines that are politically concordant, i.e. "allying with people's political views" or "discordant" in that they're not aligned people's political views. God, that is very helpful. So basically, for political concordance, is it consistent with what I believe and "political veracity" is a real or fake? Exactly. God, that is very fascinating. And the next question is directed to both of you because you guys are collaborators on this project. Where did you get the idea of studying this? How was this project created? Thanks so much, Angie. Thanks again for inviting us and Michael. It's so awesome to be here with you too. I don't know, Michael, for you, for me, I'll say I've been really interested in the problem of democracy and what it means to have a free pluralistic society and how everyone talks about what the government can do, what the media should do to promote a democratic society. But I'm also interested in just what it means to be a citizen in a diverse democracy, a participant. And to me, it seems, and this isn't really an original thought, but it does seem like there's so many skills that are required the ability to disagree without being disagreeable, the ability to cooperate across lines of difference. But another one really critical one is the ability to discern truth from falsehood. Because an ill-informed electorate cannot make informed decisions. And so that democracy power to the people, which is what it means. How do we empower people? I think one key criteria or key source of empowerment is the ability to really discern or differentiate true news from fake news. And to be open to information that might be threatening to your police, but that may never the less be true. So that was it for me. Just to add to that, when we were looking at the research on misinformation, I think Jeff and I were both struck by the existence of the time was claiming that the effective political concordance, which researchers also call partisan bias, the researchers at the time are saying that it wasn't really a meaningful contributor to the problem, that the bigger problem between true and fake news, but that the extent to which partisan bias was influencing people's perceptions of the news wasn't very big. And I think intuitively we thought that seems that kind of counterintuitive to us. And so we were going to segregate further. So we looked through the research and we thought, okay, this seems really right for running another study to try to build on what's already been done. Got it. That is real fascinating. It just got a curiosity. When did it start? You see, when did the study start? We started it. We launched the study nine months before 2020 election. And we did some piloting beforehand. And we worked with Katie Joseph, who was a master student at the time and kind of a little pilot project for her. So I get to probably started a year before the 2020 election. Probably just in time for the next one. Yeah. Our audience probably is really excited to learn about what exactly did you study? Or how did you study it? Michael, I'm wondering if you can see more about the method that you took to study this particular question. Yeah. So we did it within subjects design where we exposed participants to 16 headlines of those half were filler headlines to try to maintain a cover story that it wasn't just about politics and that it wasn't just about their accuracy. And we took a lot of care to make sure we had an environment that seems as realistic as possible. So we tried to hide our hypotheses by telling participants that the study was about memory and can be for each headline that we showed them. We had I think five questions of which there was two that were our core questions, but we tried to include other filler questions to ensure that they didn't think that our study was just about how accurate they were or how much our partisan bias they had. And we even have them do a little commitment device in the beginning to make sure they were paying attention to the headlines. And then after going through all 16 headlines and for each headline, they would rate them on this five questions. We then included a number of survey questions to gauge their partisanship, their objectivity illusion and other demographic variables and other measures that we are interested in whether they would predict partisan bias or accuracy. And a couple of those ended up in the exploratory analysis bearing out in particular our measures of one side at use consumption and one side at use trust and object to do this. And we also included social media consumption for exploratory purposes as well. And just to build on what you're saying, Michael, the within subdued design, prids that to people with headlines that were true, some of which were true, some of which were fake and then some of them were aligned or misaligned with people's political beliefs. In this case, whether they supported or opposed Donald Trump for president. And Michael, can you, I can't remember, can you give some examples of just some of the headlines that were fake and true? I think they'll be helpful. Yeah, great question. Let me just put up one second. And so I guess I would add that the methodology allowed us to precisely gauge the impact of the news is like how true it is on belief versus the the concordance of the news with people's pre-existing politically. So we could pit these two independent variables against one and other to see which one really drove belief in the headlines, which one really drove sharing of the headlines by manipulating these two independent variables independently. We get a pretty precise gauge on the exercise of each. Yeah, just to continue that. Yes, there are half the headlines, half the political headlines were favorable towards Donald Trump and the other half were negative towards Donald Trump and then half were real, half are fake. But within the real and fake headlines, we tried to tear them. So we included two tears of true headlines, some that we thought were just so obviously true, some that were true, but maybe not like super clearly true. I just to give an example there like our true headlines were Trump's first year job numbers were very good or when it that was negative toward Trump, Donald Trump three percent of growth plan is only for the one percent or Trump's approval rating was it for the first nine months was a historic well. And then for the big headlines, we included big headlines that were that we designed to be fake but not like outrageously fake and then we also tried to design some outlandishly fake headlines and we intentionally wanted to create fake headlines that people would not have seen prior. So there's actually a literature showing that the more you're exposed to the headlines, the more you're likely to believe them. So we wanted to make sure they were seeing these for the first time and just to give an example of some of the most outlandish fake headlines and just contextualized people. This was back in 2020. So it might not see with our land is now but for back then positive side, we had headlines like Donald Trump series contender for the Nobel Prize in economics or Trump beats the Grandmaster Chess Champion Magnus Carlson and on the negative side, we had I don't know if this is appropriate for the listeners, but we had some sorted headlines like Trump attended a private Halloween dollar with sex or juice just as the poll worth Donald Trump killed a pedestrian while driving in 1973. We were just trying to push it as far as we could to test it and we weren't actually sure whether people would still exhibit partisan bias or inaccuracy. engaging this headline, which they ended up still doing, which that to us was one of the unexpected surprises of our findings. That is really fascinating. Thank you so much for sharing examples. I think these examples are just really helpful for our listeners to get a sense of what those headlines are. And one question that was wondering about about these headlines that I think, all of them are very specific. I don't know if it is just examples, but a lot of them are very specific to Donald Trump himself. And some of the headlines are not even that political in a way, for example, like beat the chess master one. That could be like anyone. So I'm wondering, what's the consideration behind focusing specifically on this one tentative versus say, oh, Republican party in general? In 2020, there was research showing that the primary cleavage amongst partisans was less Democrat Republican than it was Donald Trump, whether people supported or were Donald Trump. From a partisanship perspective, we thought that was the most clear delineation of where people stood. And Donald Trump was in the news, an immense amount. So we designed our headlines to be, as we mentioned, positive or negative about Trump. But then we also, in terms of how we categorize our participants, organize them as whether they supported or opposed Donald Trump to have that match as opposed to going across Democrat opposing Trump or Republican in favor of Donald Trump. Yeah. And I would add to that, it was an historic election as it is now. And it's very important when choosing between candidates to have a realistic assessment of each and to be able to parse the truth from the falsity. So pragmatically, just seemed at the time as it does now, a really important topic to focus in on at the same time. It is a limitation of our study. We don't know the extent to which our results would generalize to other political topics or non-political topics or at least topics even outside of Donald Trump. So that is a limitation. And with any kind of scientific study, you go deep in one thing to achieve certainty on that one thing. But then in achieving certainty on a specific thing, you lose some ability to comment on issues just outside of your focus. So we acknowledge as a limitation of our study. God, I thank you so much. Before moving on to the results, one last question about the method in process, where did you recruit the participants? Because I think we at a university in California, that has a very clear political leanings. And I imagine if you're just randomly recruit sample people around you, you might not have us at diverse political spectrum that you might want. We used a lucid marketplace and to recruit a set the US census matched a sample. So within, I think a certain margin, I think it's within four percentage points. I'm sure that all the key demographic characters across income, race, ethnicity, and so forth were matched the census. And we chose lucid in particular because we were trying to use a similar framework to the studies that had been done before us that we were in a sense looking to hold some of the variables constant and then try to achieve some other dynamics such as our study design. They sent me had a cover story. But we, yeah, we've got a penny cook and David Rand, published this paper, amazing, not biased. And they collect their study from lucid marketplace. We figured let's use a similar sample. God, that is really helpful. So let's dive right to the results. This is also a question for both of you. What is the most surprising thing that comes out from this study? Cardi trunce reality. What we found was that the effect of political concordance was, I think Michael the precise numbers, but almost depending on the analysis, but somewhere between one and a half to two times the effect of truth in terms of people's belief in sharing. So when it comes to what news you believe and what news you choose to share, the thing that people are using to determine whether they should believe and share isn't the actual veracity of the story as much as it is the story's ability to support what you already believe. So a huge confirmation by this people believe the news that concords with their political views, regardless of whether it's true or not. And that's true even for the most outweigh in the edge of headlines. So I think there was one. I think this is falls under what Lee Ross would call the demonstration tradition of social psychology. You show something. You show an effect that people already know to be true, but what's not so obvious, like the effect might be obvious. Yeah, people believe things that are consistent with their beliefs, but what's less obvious is the size and the impact of this seemingly obvious effect. So what's non-obvious is the effect size. And I think when one of the surprising results would just how much it mattered, how much it mattered in terms of people's acceptance of the news. And then I would also say that one of the things we often do when we hear results like this, we go even back to the Millgram study and classics and social psychology, people are always like not me. I would never do that or not my group, not people like me. And I think for us, one of the surprising effects was just the broad appeal of the broad applicability of the effect and the broad appeal of even of fake news that we found the same party over reality effect across the ideological spectrum. So for Trump supporters and for Trump opponents and up and down the educational level. So even people who had advanced degrees at PhD showed this the same sort of we could call party over reality or politics over truth effect. It wasn't something that was confined to one specific group, the ideological group or people who have less education. It was remarkably consistent. Just to add to that for myself, and Jeff is right, he has about one one at two times for belief in two to three times for sharing in terms of the effect of political concordance compared to the effective truth. And to further underscore that point, the existing research before paper came about found an effective truth that was about four times greater than the effect of political concordance. Whereas we found the opposite. So that was definitely surprising when we looked also at the effect of partisan bias or the effect of one thing that was very surprising to me was that it was actually stronger amongst our real heavens than our fake headlines. So another way of saying that is that our participants were more likely to resist inconvenient truths than they were to be susceptible to convenient falsehoods, which I think most counter to our intuition. I think we tend to think that misinformation is at the core of the problem. And the result suggested that there's actually it's not just people's perceptions of fake news. It's also their perceptions of real news. And then actually the more we thought about real life, I guess that does have consistency with the way people perceive elections, the way people perceive science coming out, pandemic, and right now, hurricanes. So that was also surprising to us. And to Jeff's point around the breadth of it, the that this effect held up amongst people who had the highest and the deepest form of cognitive processing even amongst people with advanced degrees was also a little bit humbling. As well. Yeah, that was really fascinating. So one question I had, I think both of you mentioned like the effect of shirring. So can you see more about that? I think for me, the shirring really feels like a very like social media error thing. Oh, I saw a news. I'm just sure to my friends. But is that a case or are you talking about sharing this news with like family, friend in general? To be a question, we framed it as how likely are you to share this with friends or family? And so the implication is as I think what would you be willing to share it on social media? And part because the question after that is how would you react to this new article on social media where we included a number of emojis? But the question was framed generally. So it could be shared on text, right? There's a lot of other channels to share information beyond just posting it on social media. One thing just to say is that so it's not just that people are believing the news or believing misinformation, for example, though Michael's right that we found even stronger effects of resistance to truth than we did on acceptance to falsehood. But there's like a kind of accumulative effects of these sorts of processes because people not only are believing the news, but but does we share it with friends and friends of friends? And so it's that old, old, Jewish story about the effects of gossip. It's like kind of terry open a pillow and all the little strands go out into the air and start to affect people in ways that you can't take back and have a thousand little effects that you're even unaware of. So I feel like the sharing variable is really important in our NIH because when people share fake news, it's just going out there into the ether and polluting more minds than we can count. Yeah, I agree. So the thing that is really surprising to me when I was reading this paper and I think Michael, you mentioned it is that people with stronger analytic reasoning abilities were more accurate only when the headlines align with their ideology. And that is really surprising because we were thinking that having a lot of trainings in like critical thinking, we're really helping this. So I'm wondering, what's retake on what does this suggest about the limitations of rational thinking in political context? If people with the highest abilities to, critically, I don't know, to analyze this scenario is do susceptible? What does this tell us about what we can do as regular people? And to be clear, the reasoning abilities that we were measuring were from the cognitive reflection test. So these are the type of questions where you might want to respond in a certain way that seems like an obvious answer if somebody passes the person in second place with place today and so they require it's more differentiated between type one or type one. to thinking, but it's not capturing the extent to which people have, say, intellectual humility or a sense of their own scalability or who an awareness of their own biases and work by the Ross and Emily Pounin has shown that the biggest predictors of whether people are able to overcome a biased blind spot is whether they're aware of their introspection processes or whether they're aware of how their biases work and the foulabilded their own cognitive faculty. I think it's a little tricky to say, these are some of the smartest people in the room and they're still suffering from these biases. It's had these people are ones who were cognitively engaging in a deeper way, but it doesn't reveal anything about their self-awareness or intellectual humility. Yeah, just to add on that, I think there's a bias out there in the academic literature that stronger reasoning, more depth of processing, can dig you out of these biases. And that's a belief. And to some extent, a past finding that's out there suggesting that when people are more habitually engaged and deep or systematic processing assessed by this test that Michael and I gave to our participants, they should be less susceptible to these biases, less susceptible to fake news, less susceptible to being carried away by your beliefs and just believing whatever you want. And so there's hope. So that was the kind of belief that's out there. And we didn't find any evidence of that. Instead, what we found was something we called selective reasoning, or you call it selective reasoning, that it seemed as though people would apply, people who scored high on this test, that what we found is that they were more likely to discern fake news, but only when that fake news challenged something they already believed. So it was as if they turned on their reasoning powers when their reasoning powers would lead them to a conclusion that they liked. But when the their reasoning, when deep reasoning would lead them to a conclusion they didn't like, it was maybe it was fake news that shined favorably on the preferred presidential candidate, it seemed like they turned off their reasoning abilities. Now I'm speaking loosely and more simplistically, but that then the actual, I don't really know if actually they're turning off their reasoning, but the pattern of results suggests that reasoning, this are people who scored high on on this measure. That was an advantage in terms of your accuracy. Only when the news was congenial, when the fake news was congenial to, to what you wanted to believe. Does that make sense? Yeah, that is really fascinating. So the next question is probably a little bit hard to answer right now because we are recording this episode on October 24th, still about two weeks away from Election Day and this episode is scheduled to be released right after Election Day. So I think a lot of what happened in this period. So in your opinion, what are the most concerning implications of the spies that you have found, especially in light of the studies findings about media consumptions and partisan beliefs? Because I think this topic is really timely. For me, my biggest concern is that it's going to get worse over time instead of better. Given the key drivers we found for prioritizing or being more influenced by political concordance than truth were one sided news consumption, extreme partisanship, and then also believing that your side is objective and less biased than the opposing side, those factors seem to be super present. And I think with increased democratization of news consumption, people are able to consume more of the news that they want to that confirms their beliefs or doesn't disconfirm them. And so those factors seem to potentially also work in a recursive cycle where the more one sided news you consume, the more you tend to think that news is actually the real objective truth. And also further is the objective illusion and might even influence the extent which you believe that your sample of news that you're exposed to is not biased. So my biggest concern is that I think a lot of the attempts to address it are also just focusing on fact checking or pre-bonding which the research has shown can help in people in getting people to be able to discern between true and fake news, but doesn't address people's underlying confirmation or his confirmation bias. And thus isn't going to impact our people or perceiving true headlights, which as mentioned is where we find some of the greater partisan bias. Yeah, living in a potentiac than a post-truth world where there's a less shared epistemic commons, I feel like further polarizes us. And that's my biggest concern. Yeah. Yeah, it's beautifully said. I would get this bias is a real threat to our democracy. In so many ways, I think, right? If we can't agree on what happened with our candidates, how are we going to pick the better candidates? Misinformation is also a problem when it comes to dealing with these national and global problems where I think like the first step is just agreeing what reality is. And so as a society, there's a lot to worry about here, right? How do we how do we meet the challenges of these problems when we can't even agree on what's a problem or how much of a problem is or what the causes are? I think these biases, and that's even 90 minutes of mention just the way in which misinformation can spread and hate as we see with misinformation about immigrants in America today, and scapegoating. I think it's a huge, huge threat. And that's why I think it's really, I think I do believe it is a frightening thing. And I think that's why it kind of getting a good conceptualization of the problem is so important on what to do next. I believe that people think that the problem is too much misinformation, the lies that are being disseminated, and that for sure as a problem, it's a supply side problem in doing things like having fact checkers and people to create teaching people skills to creep out fake news. All those things are important, but that's not going to be the whole solution because as Michael has pointed out, it's that you're even in a media landscape where all the news were true. You would still have people given our results resisting true information if it contradicted what they wanted to believe or what they believe. So I think one way to put this is, I was thinking, the problem isn't just the lies. It's our willingness and desire maybe to believe the lies. It's the filters of our own minds that lead us to believe these deceptions, but also to turn away from truths that are unpolatable to us. I do feel like to some degree the enemy is within here. The problem is in our own minds and our readiness to just believe things because they agree with us. It's almost as if we're going through this world thinking like, if it agrees with that, would I already believe it must be true? I think teaching people, as Michael was saying, almost have a driver's manual for their own minds knowing that you can't trust, as the Dowager Jean puts it, you can't trust everything you think. It's like these thoughts flew through your mind and they might not necessarily be true. I think part of the solution here is to help people to be, as we'd say, in a lot of such intro psychology courses, to be critical consumers of the news and research, but that endeavor begins with teaching people to be critical of their own minds, because our minds trick us. And at the same time, they're in lie some hope because the people who did seem to be able to differentiate truths from falsehood a bit better were those conceded that, yeah, my side is in some ways just as bias as the other way, other side. It's not like I or my side have sole possession of the truth. I'm vulnerable to bias too. And it was those people who had that intellectual humility, as Michael has talked to you about, who provided a bit of glimmer of hope that they were better able to differentiate truth from falsehood, even when it wasn't something that concorded with their political beliefs. And so maybe that's part of the solution here is helping people to understand their own minds better and having a bit of humility, honesty, I heard that there's a new personality dimension out there. So no longer the big five, it's the big six and the sixth dimension is humility and honesty. And that's the kind of civic virtue that we can all aspire to now. And just one last thing to add there, I thought Carl Sagan had like really distilled it down in a really simple way that I thought captured this is in what his books he wrote his biggest fears of a vision for the future where people are unable to distinguish between what feels good and what's true. Yeah, that's such a nice quote. Just out of curiosities, have you, if I don't know if you have done this, but I'm just wondering if you just tell the participant who have participated in a study or finding and ask them to do it again, would you predict that this just this awareness, this knowing mitigate the effect a little bit? That's a great question. Good research question. I think Emily Prattin has shown that just telling people about the bi-spline spot alone doesn't reduce it as much as we love that to be true. You almost you have to experientially come to the awareness. And this is where I think it's a really cool area of psychology where you see different subdomains being able to help fertilize each other where she shows people like visual tricks of how the brain can trick itself with visual stimuli. And that experiential process of seeing it and fearing it firsthand is what seems to be most effective at getting people to understand the fallibility of their own minds and to learn to be able to be critical of their own minds. And on the other hand too, it doesn't feel good to feel like your brain is not something you could totally trust and rely on, right? There's some motivational some some self identity aspects of feeling like you can live in a world where what you think is true and so intellectual humility also takes some work on the other side of whether the longing or a sense of having a sense of confidence or self-esteem that the Tosco interesting to explore. Yeah. So I wish to running telling people the findings would help that it might improve their discernment between true and fake news because some of the past studies that didn't include a cover story just told people had his studies about accuracy and hear the headline how accurate you think it is and that's in part why they got effects of headline truth that were large us because they've shown that telling people were focused on accuracy surprisingly does not surprisingly get people to focus more on accuracy. To explain to that I think your question is really interesting. Tell me about me. I think there are these video games now where you catch yourself in the act of committing biases where you write you serve, oh I just committed a confirmation bias and you see it unfolding and you experience it introspectively. So it's experiential and I think they're having some luck in in debiasing people with those sorts of video games. So the extent to which the effects are lasting it transfer to novel domains I think remains a little bit up in the air. But I think it's a fascinating area. I for me I often feel as though what's helpful is to write have that humility but also I think just see my mind as like a territory that people are trying to colonize and it's there's different media forces and government forces out there trying to like a virus. They're trying to colonize my mind and understanding that there are these forces at work trying to control me and to make me think things that may not be true. That's also a helpful too. So it's about humility but it's also about skepticism and being aware of my place in a larger system of influence that may not always have my best interests at heart which we know right from these social media companies that use algorithms that are designed to feed you information that make you feel good, make confirm what you believe, create outrage rather than actually teach you something. So understanding that yeah there are these forces at work that are trying to colonize or mind make you think things that may not be true always. I think that's also helpful. One of the other findings consistent with this in our study was that there were a couple of things that two things in particular that were predictive of less bias. One was that humility dimension understanding on my side and it is also susceptible to bias. The second one was not consuming so much one-sided media. So people who had a more diverse diet of media or who in the famous words of Howard Beall would just turn it off, they seem to be a bit more able to differentiate truth from false hood. So there is that. There's training your mind for bias having these kind of metaphors of mind that help you to navigate a complex informational world. And I guess I would also add, I feel like this work by Phil Telox also useful. Having a kind of admitting to the fallibility of your mind but also having a growth mindset about your ability to learn and prove your beliefs and make them more accurate so that you treat your beliefs not as treasured possessions but as hypotheses that you're continuing subject continually subjecting to testing. I know there's a kind of lot of things whirling about here but I think this is such a great endeavor, such a great problem space for social psychologists and a lot of different disciplines working together to help people to be more adaptive citizens in a pretty calm informationally complex world. Yeah, thank you so much for sharing. I know we are short on time but I'd really like to end our conversation with a forward looking note. Someone are you, what's on your mind right now? What's the future directions of this fly award that you tend to pursue? I think one question that came out of the research that would be fascinating to better understand is why partisan bias is stronger for true use and fake news. What's the driver for why people are resisting and convening? True news more than they are accepting or being susceptible to convening fake news. Is it the nature of the headlines or is it, is there a certain underlines human phenomena of lie for instance, disconfirmation bias might be stronger than confirmation bias? One other finding we got was that people recalled and remembered fake news more than true news and on the base of it that seemed obvious that when we tried to control for all the other features of the news, people's reactions to the news, they're surprised. The effect persisted and I think better understanding what is so sticky about fake news that causes people to remember it so much more than true news is also another fertile area for research that came out of it. I would add to both of those which are great solutions. What do we do with it about it? Now that we have a kind of understanding of the scope and severity of the bias and maybe some of its sources and a lack of intellectual humility over exposure to one side of media, what do we do about it? Are there interventions that we can deliver or make available to people so that they're able to be good consumers of media and be more open to admitting when they're wrong and learning things that they didn't really realize, learning to update their beliefs in the face of new information. So what are solutions? What are interventions? I think that's a really important new direction. Michael has done some really interesting work in showing or suggesting that helping people to be a bit more subjective in the way they talk about political issues. So when you give your opinion, you say things like I think or from my point of view, that may, he has some initial evidence that may help people to become more, have more humility about their own points of view and about points of view that may be different from their own. And so one potential intervention is maybe not just teaching people about biases, but getting people to act in ways that project humility. And when they act that way, they might act their way into a new way of thinking relating to their own minds. That's beautiful. And just to add, in addition to interventions for individuals, I think there's also a question about what kind of policy would be important to explore, given these findings. And there used to be something in the US called the US fairness doctrine, which got revoked in the 80s, that required news to at least provide time for the other side when they presented their sides of certain issues. And once the US fairness doctrine was revoked, you really saw the rise of hyperpartisan news. And I think that might have contributed to our polarization. So I think a fascinating question to be discussing is what is a contemporary version of a fairness doctrine even look like with social media and our current news environment, which would be great to see more discussion around that in the policy circles? Awesome. And I would like to thank you again for coming on the show today and I look forward to read more of your working in the future. Thank you so much Angie. Thanks for having us on the show. It was delightful. I thank you so much. We enjoyed it. (gentle music)