>> tonight we are thrilled to have t14 his book breaking the social media prism. professor of sociology and public policy at duke university when he directs the polarization lab a leader in the emerging field computational social science with the fundamental question of social psychology political polarization and technology using large-scale experiments of social media platforms. in-depth interviews and the latest advances of machine

learning. written for "the new york times" and "washington post" and cnn regularly lectures to audiences and government in the nonprofit sector to combat polarization. his award-winning book how anti- muslim french organizations became mainstream was among the first to explain how social media had political extremism. joining chris in conversation jonathan is a social psychologist at new york university at the school of business receiving his phd from the university of pennsylvania in 1982 anti- for 16 years at the department of psychology at the university of virginia. the author of the happiness hypothesis. and "the new york times" bestseller the righteous mind.

and the cobbling of american mind. one - - more than $100 academic articles with the american academy of arts and sciences and chosen by prospect magazine one of the top 50 thinkers. a few housekeeping notes before we begin. we will do audience q&a if you have questions, you can drop them in the chat or in the q&a feature at the bottom of your screen. with that please join me to welcome jonathan and chris to the state each. >> thank you so much for getting us through the technical difficulties. it is 7:20 p.m. on a friday night on the east coast.

i told chris i would be drinking and i invited him to have a drink. trying these new alcoholic lemonades that are good that are too sweet. >> now a very polarizing product. i wouldn't touch it. [laughter] >> what are you drinking? >> this is the finest local beer in north carolina. hold it up to the cameras of people can see and get free press. [laughter] >> i am pleased to be doing this with you. so we will hear from you about your stories and this whole thing. you write about things and everybody is wondering about and freaking out about in some cases. so for the audience, chris is studying one of the hardest

topics we are all facing which is basically what the hell is happening to us? doesn't social media has something to do with that? we feel it has something to do with that. so i read some of chris's research we been in correspondence for the great conversation by zoom. he asked me to check out his book i offered it so i will just read that because it is what i think why it is so important. t14 is been the best research on the polarizing effect of social media in his work he shows if you want to understand what is going on , online the don't focus on people's exposure to information. rather keep your eye on people's quest for status and group identity.

the book is rich with insight firm social media and essential reading for anyone who wants to improve it. going back to grad school how do you come to study this topic and write this book quick it's almost like buying bit coin and tesla one year ago. [laughter] so was there an event what you saw happening? >> thank you for doing this. it is a treat and thank you to the strand for arranging it. i got into this in a moment in us history which was pretty important. for the first time to out party hate surpassed in party live at the time i was in graduate school. now between 40 and 50 democrats according to some surveys are willing to use

violence to achieve political and. so we really reached a crisis point. you put your finger on it so what is the role of social media? when i set out to investigate the prevailing narratives what divides us on social media, i was shocked to discover there wasn't a lot of evidence to support the echo chamber or misinformation campaign. the important truth that i discovered in this book is it is us. social media users driving the problem even if we could get platforms to enact every single policy change, probably wouldn't move the needle a lot. host: i hope i can push back later. we will get to that.

i hope you're not letting them off the hooks. >> absolutely not. host: so these conversations are scattered. so some people make the country argument. so as it would help me if we are clear what exactly is social media? what are the kinds that matter and are polarizing and what are the features? tiktok and facebook and twitter but youtube? what matters? >> we are in the early days social media every big technological upheaval from printing press to cable and tv we see the surges. to some degree technology makes the idea of the public

possible because we can encounter each other in virtual settings. so with social media and allows us to have conversations to broaden the scale of the conversation. but certainly i'm old enough to remember platforms like myspace where younger people have never heard of. social media is evolving. it is premature but the unique characteristics for anyone to enter a conversation about politics and has potential for good. >> salon the political and social affect the key is that allows people to but to ask about. >> i've only then studying for a couple of years with the effect on teen mental health

simon focusing on driving people crazy is a feature of user generated content why did they like her picture more than mind? is that important is it just our product. >> the single greatest driver on social media is a rapidly growing gap of real life and social media. we all know that would say something online they would never say in person but the more we are isolated the more we come to rely on social media to understand each other the reason i call my book the social media prism it distorts the way we understand each other and ourselves. it is accelerating parts of

human nature like status seeking especially social outcast but it makes it seem completely invisible and that makes us feel more polarized and we are. >> that is something we have been observing which was spread into so many other institutions and creative industries but it's always been the case that then went social media spreads people keep their head down a student said to a friend of mine that silence is safer. that's horrible for college student that's her motto going through college.

it is a prism it distorts of who can speak and who is being silent so let's be clear about the harm you mentioned polarization in your title what are the harms done by the social media platforms? >> it is that status seeking extremist. let me just blow your mind with one statistic 73 percent of tweets about politics are made by 6 percent of twitter users. and those 6 percent have very extreme views. if you encounter someone you don't see a typical person and

one of the things we did in this book is to trace people online using data science with massive social network but also with the experimental experiment some in the course of the research we discovered how profound between real life and social media can be. what sticks in my mind is a guy i call ray in the book. in real life he says things he is the most polite deferential to go out of his way to condemn racism all these people on social media are losers because they live with their parents. but then if you look at him online he is the most prolific political troll i have ever

seen after ten years of studying this. every night he does and to doctor jekyll mr. hyde transformation sharing meme after meme of unspeakable things of a oc or barack obama or nancy pelosi. so what about these profoundly different identities and what consequences does that have quick. >> you are sociologist i'm a social psychologist in the idea that people are one way in public and another way when they have anonymity makes a lot of sense and of course if you are freudian that is obvious. but we both share in common and you have a great quote we

use to live in one world now it is is mark zuckerberg world. >> goffman was way ahead of his time in the 19 sixties a sociologist who wanted to understand why we do this weird human thing we care about creating identity and each day he discovered each day we are experimenting with identities trying on a different version of ourselves and looking at heather other people react and cultivating those identities to make us feel good about ourselves. we have done this forever and social psychologist have taken this up to wonderful places but the interesting question is how does social media with

that process? the first allows us to be fully anonymous were to share selective accounts of who we are. we have a much broader palette to experiment with identity. the second point more efficient tools to monitor what other people are thinking. we can get the likes and the follows to create that endorphin values. we don't ask the important question. why am i doing this but it is the elephant. thank you for putting into psychological terms thinking about behaviors and psychology.

and then to explain we don't talk about them that much but social media really brought back with he trains that pigeon but all he does waits until it goes to the left and then here's the click and knows that there is a tiny grain of rice. then he waits until that pigeon turns more. so we can try things on. we both got to try things on and then within a few minutes you get lots of pellets of green or nothing or nasty comments. or silence. behaviors very powerful for training animals.

if in real life you don't get to award people that minutes later they don't work. that training and behavior in the first time in human history we put ourselves out there to say world, train me. world, shape my behavior. for teenagers this is completely disaster us. sorry. i am passionate and i love your book so much. we don't understand the many ways this is changing the basic fabric of society. in such a way that nothing works right anymore. tell us about other harmful outcomes. tell us about extremists. >> on the one hand i totally agree that the skinner analogy. we are looking for that instant.

but not everyone. this is a key point. i'm not arguing everyone has the potential to turn into doctor jekyll or hide. the average social media user hardly post at all they are basically visible and rarely post about politics. one i call sarah. she is an interesting character. she leans conservative that from new york city. half puerto rican. she went to the ivy league school very nuanced views and very open to liberal ideas. you will never hear her talk about them on social media. why? tell us about the last time she told us an amazing story. was up late putting my kids to sleep my husband owns a gun

the nra put up a post we support americans rights to own guns. she reach we that and replies and then within minutes she gets replies from people who look at her twitter profile. found out she had kids and are posting we hope your kids find your gun and shoot you. this is all too common. and the leading research firm put out a study that suggested the number one reason people are harassed online is for political views. so the story of jekyll and hyde character is concerning. even a more pernicious effect to make people like sarah who desperately need to bring the conversation back to the middle cannot. >> the harms are extremism, silence, making people afraid. democracy. let's talk about declining trust there has been a long

decline the trust institutions and others. it may vary from country to country. tell me what you know about declining trust. briefly talk internationally is all western democracy having the same problem quick. >> that comparative question is fascinating but there are variations. a lot of countries are going down like in germany. one thing we have seen across country is dependency for social media to drive all polarization. this is the built-in human tendency to exaggerate the other side and minimize the underside. now with survey research and trusted outcomes, the number one predictor is false correlation the number one predictor is social media usage across ten countries so

one of the most pernicious effects are the ideas we have about each other. so we have an intervene we can correct and then another new line the research shows it does build trust and build animosity between people. >> with that polarization it is true parties. so we have a conversation with mark zuckerberg was reaching out to get competing views and i said but this causes all kinds of polarization and shortly after that one year ago. and his theory was look at

where political polarization is going up is mostly in the countries with rupert murdoch. fox news. he was saying it's not social media or facebook it is fox news and that sort of stuff. that did fit with the study. so is it social media or is it in sync with other things? let's broaden the lens. >> absolutely. everybody asks the question is social media driving polarization? that's not the question we know that was surging long before facebook and twitter. we can point to the southern strategy or market segmentation with perverse incentives for small markets

given point to the collapse of the soviet union and the cold war and provided a common enemy. there's a lot of drivers. it isn't too protective we cannot explore an alternative reality. but i do think that is contributing to the problem. no matter the answer to that question but given we will not have social media anytime soon hardly make that less polarizing quick. >> i see tobias is in the audience. i'm a social psychologist i just had a vague sense i met tobias and he clued me into the social media platforms changed radically between 2009 and 2012.

that wasn't radicalizing just presentation. >> that was and facebook copies each other and suddenly it's much more engaging and engagement with anger and that's when we get that and that timing. if you do agree with that, then what can we do? what is the segue? we should open to audience on the questions in about five minutes. >> it's a great question. i do think we need top-down solutions and how we have content in our feed. have a variety of ideas how to fix that. but to move the needle the area we need to work on is bottom-up. the data is clear.

even if we did make these changes we would be a very polarized place with incentives for people to be terrible to each other. and the structure of the current platform. we need to learn how to see the social media prism and the gap between social media and reality. learn how to avoid extremists that boost moderation. we need to find better ways. you have been such an inspiration to me. but what is important as we share this vision, it's easy to say we need to be more aware and reach across the aisle with that blue mean buzzing confusion, how do we create durable changes? that is where technology comes back in. we diagnose those drivers and

we created a suite of apps that users can use to identify and avoid extremist. get a sense of what other people see about their politics from your post. and we can use technology to focus on the small minority of actors which is where all of our attention is now and then to have minimal impact with counterproductive impact but we have done very little on boosting moderation.

so if we are looking at where we can create the most impact the bottom is the most important. >> . host: here is where i get to disagree with you. if we think about public health issues, smoking, overeating, pst have for a long time been saying people for life where sugar is prayer they gorge on it now super stimulus then junkfood now it's hard to get people not to overeat. of course we are limited we cannot ban junkfood. companies have a right to make things. cigarettes we can regulate. but it's very hard. so talking about the biological pleasure so the mind is divided like the

writer on the elephant which is conscious reasoning that the elephant is smart. you don't get change until you change the elephant. it sounds like you develop these great tools to show the writer what he is doing and helps them stay away from extremist the elephant was extremist and is attracted and it is exciting. and the other things were food and physical pleasure which is addicting but social reputation blows out anything else it is so much more powerful. if you say we have these platforms. they were not so bad before 2009. then they got bad. now we have to exert more self-control. i love your diagnosis. i will promote your book. but i am not putting any money on your solution.

with the top down and platform changes. >> to be clear i am not naïve. these will not end themselves to create a seachange or polarization. but take your analogy further we've all ballot benefited from the analogy with think of the envelopment - - environment the elephant is navigating. there are other elephants. sources of food. that is where the top-down solution matters. think of the landscape. the platforms create the playing field that we all play it. if he want top-down solutions think about how we incentivize better behavior. the most concrete and immediately actionable idea i would like any other to take up tomorrow like facebook or twitter but boost content that appeals to different types of

people. . . . . we work a long time with loved wors and we experiment on these designing features because when you think about it the crazy thing about all this is that we never really considered in a concrete empirical way what would be the best designed on social media platforms to encourage stability and moderation.

we allow platforms that were created for harvard undergraduates to wage others physical attractiveness or arrange alcohol centric data in a default in the primary forms. what we would love to do is say they make facebook and hey twitter lettuce tweet something and tryout the algorithm idea that i just mentioned in the way that cambridge analytics was repurposed to serve arguably malicious political ads. everything has been shut down. there are glimmers of hope but the fundamental research on the guts of social media and the design principle is not possible so what we did in the polarization lab is. our on social media platforms and the whole thing about this is we can turn on and off in different pieces of the platforming control who interacts with two and we hope we can pin down what works and

what doesn't. >> he's are the kinds of tools we will need to do these things because we can experiment on the platform. and one of the solutions we talked a lot about anonymity and when we first talked you are suggesting one of the big problems with anonymity is there no consequences for the people do on line and that's the doctor -- on the other hand and there are some underappreciated positive dimensions for example if we are all trapped in this mentality identity is taking precedence over information, then when you think about it and the name of the in some ways drops identity out of the equation. we are no longer liable to our peers. i'm an democrat who is concerned about defunding the police. i post about twitter there could

be all sorts of negative consequences for me and if i'm a republican with a narrative about voter fraud that could be great trouble for me if i post that. anonymity gives us the time to explore popular ideas without pressure and a nice example is for example reddit change my view. reddit some past some of the most reprehensible qualities on the internet and this is an example where political compromises possible. we did this study where we created a made-up social media platform. unbeknownst to them we pair them with a member the other party to talk about issues. we were shocked to discover people in this anonymous condition be polarized much more than people who didn't especially republicans and that's particularly interesting. maybe it in my view the pier pressure on republicans with

that viral silence dynamics where people are scared of sharing their unpopular views may be anonymity for some carefully designed forms could be one way out. >> so you are referring to your most famous study that 2018 study. said this is a new study. it's a study that we just did last year. the early experiment was the one with the lots that we asked people that followed lots of exposed them to the other side for one month. >> that's right. that was an amazing study where you made up interactive people and it actually made things worse but let me be clear i've never criticized anonymity.

tobias and i understand of course you have two let people have anonymous accounts. our point is there's a difference between zero accountability and a tiny bit of accountability. what we want to what we are proposing is and this just makes sense is anyone can sign up for an account on these platforms to see what's going on but you have to at least be verify that you are a real person and you are over 18. and facebook is going to see your driver's license so your point is if you say terrible things -- but if you make death threats you could be found so that's our point. you agree but because trolls are kind of the fbi isn't going to

get a court warrant but if you are planning the shooting of a synagogue or a mosque on parler are making death threats to people or their kids and twitter -- it's an amazing example. i hope the audience will take this too hard heart in one day tell people about this. it i don't know if you've noticed the chat chris. just as we've been talking. >> are we being trolled john? >> yes, yes some made up the most outrageous the tedious alex sink he could think of them posted over and over and because of that day to disable disable that so this is becoming like terrorism. we have to learn to live with this. we know that it's a guy.

i would -- actually quite pleased that i did this. this is what anonymity does and look we had it right here and we had to shut down the chat because of mr. ray's. >> exactly but let's think about the incentive structure. obviously you and i are fun to troll right now. people and twitter are standing around looking for people to target to do exactly what the guys doing that if we. a different kind of playing field for example if you are rewarded on your own side it will be much less fun to troll because all these comments and negative stuff it won't get listed and resonating and trolls will sink to the bottom instead of rising to the top. >> that's what the platform

needs to do. i want you to give me a reliever onto things and one is stop the troll-ish behavior and the other is if you sometimes say on the one hand this on the other hand that if you sometimes say that i want to hear from you if you've ever said that. i want to make you i can't see you and you can't see me and that incentivizes lot of people and if you want followers and you want to be accepted you have to stop being so and sometimes have some nuances. >> exactly. and i wouldn't add one they wanted that and that is not everybody's going to do this. but this is a feature. where's wintering social media all of the ways. we have instagram guerra pictures a one out a form for politics? do we know most people could

care less about politics. something like 6% of posts are about politics but we also know opinions trickle down from opinion leaders. there's a small segment of the population like influencers and opinion leaders who want to share this stuff. let's. a place for them to have productive debate and get rewarded. we even have models for this and technology. there's a site that computer programmers use that help each other solve problems and i get awarded if they solve a problem. imagine solving public problems. meme chris there are a number people from silicon valley who would like to do it. if have you thought of or have you developed your idea here but it seems like if you would team up with some of these people these venture capitalists you could. a different plan.

is there anything like that in the works for you now? >> not with us but what we are trying to do is build the scientific infrastructure so that people can experiment. our goal is to make the platform that we have told something that everyone can use and not just scientists like us but also entrepreneurs. nobody like social media. this is the funny thing. nobody says oh i had such a good time on twitter last night in nobody says you know about facebook -- there are some exceptions but for the most part the public dissatisfaction is rife. there's an appetite for something new and if we take the long view every two or three years a year ago with tik tok and what's going to be next? the this is where nobody is taking a scientific evidence-based approach to figure it out.

>> okay, the rate. we started 15 to 20 minutes late and the chat was interrupted but we asked people to e-mail their questions and i do see one or two that made it through. i encourage -- if you have one ready to go white why don't you come in and read it to us or something. in the meantime i will look at these two or three questions here. go ahead. see thanks and my apologies and thank you for being flexible and taking e-mail questions. our first question was i think from our league who says please address the social media prism

and identity and power on the microand macroscale. she's asking you to clarify the differences between the individual user and how we talk about it generally. >> the problem is we have the wrong model. we want social media to be a competition of ideas but it's really become a competition of our identity. that was probably predictable. we could argue about whether we have had and directed debates in the media but the key thing is when we go on line, think about the last time you encountered someone -- on social media. did you calmly say oh that person has a point calling alexander o. causey of cortez the communists are calling president trump a racist?

its competing for better evidence. we are learning increasingly that to defend their side in this is what social media is doing to us. the micro level we are learning these tactics and those are detrimental. we have stories in the book and this is what we do in the first two chapters of the book. if we focus on the moderate david french or the centrist people it's about ted cruz and the crazy thing that ted cruz said yesterday in the unenthusiastic part may become polarized in a demonstrative exposure. need to think about not, there's no like now we are getting to the macro. we don't just flip a switch. that would degrade but we need to get this two-part process.

these arm microinteracting with the macroso the macrois creating a structure we are also participating in it. even though we all know we are not changing each other's minds which the vast number surveyed says they are changing people's minds so what are we doing? >> i really want to underline this point in the book and it's a point that runs so much of her politics through the gradual change from ideas or interest to identity a theme of the 21st century. ezra cline has this great look with research from political scientists and sociologists. political political parties used to be coalitions of industries and industry groups and you can compromise. we will trade this for that but

beginning with the sorting of the parties and that after the civil rights act was passed once we get these purified parties the people in the other party are -- and that wasn't so true in the 1970s let's say so our national politics has shifted to be not even about money it's been more about identity and it said trump made it much more so and biden is has done a good job of trying to go back to a more democratic focus on the economy as opposed to the culture war stuff. this is something we are seeing on campus and now it's showing up in high schools and private high schools the emphasis on identity and encouraging people to think about identity, some are bad and some are good and this is just blowing up. i would urge our audience if you see this happening now you know

it's a ship identity is really bad for diverse liberal democracy. i don't know finance or zierlein's question but -- do we have another question? >> this question is from -- who says we don't often think of it as a social media platforms allow the ed does happen on google. we have seen promotions for black owned businesses etc.. it gives one a feeling that supporting such businesses identifies an unintended consequence. it's something that brings people together or the ability to identify specific ethnic

groups causing more separation? >> these are really fascinating questions. obviously we all use it and youtube uses 85% of americans now days that uses around 87% that but youtube is the dominant social media platform. but the interesting thing about this point is yes if the data question and a privacy question but it's also fundamentally about how do we discover each other and of course the enormous strength and benefit of technology is exposing to more diverse information. technology can help us do that and it makes us much more efficient. everybody complains about algorithms but can you imagine getting a -- the cat is out of the bag but we are scanning through all of this information.

technology when well-designed and that's a huge caveat so i think that's a different point but the second is geographically already the for covid we knew geographical polarization and segregation was happening. recent study out of harvard two weeks ago by a political scientists and using high-resolution voter data that mimics in some ways find resolution most democrats and republicans live in areas of the country where they will almost never interact with a member of the other party. that's a stunning statistic and unfortunate means for better or worse places like little are one of the few places. we are going bullying anymore together. we are going to have these conversations on line so for me it's one of the most urgent questions.

>> there is one more question we will take it otherwise we should wrap up. >> they are last one is from julie who says hi from omaha. she said looking forward long-term what we think is the best worst-case scenario? will polarization trend out and will it correct itself? >> i think all of us and i know john you've written about this too will hope that covid would be this check moment. in a way it was a common enemy. it became political and initially was a threat to all of us. there was coordinate a behavior for all of us to address it and i was finishing this book when covid had so i had this moment where my whole book was going to be irrelevant and one of the interesting things i was able to do is go back and revisit the characters in the book.

it was shocking to discover the social media prism was as powerful as ever. supreme -- they were talking about china and bleach and the moderates were completely invisible. even though the majorities of both parties were agreeing about social distancing closing the borders until the politicization happen happened in large part because of social media. the current trend i think is that we can't conquer covid together and social media was making that problem worse i'm a little pessimistic about the current -- but on the other hand we are only 10 or 20 years into this and we can learn a lot especially if we. opportunities for entrepreneurs to articulate will be the next thing and how we'll make it better and there is some evidence that what they are saying publicly is they realize it's in their long-term amp or

interest that they are losing advertising revenue with these boycotts. in my opinion out what your john thinks that my opinion is we are in a critical moment and there's a fork in the road which we'll even go off up the roads even more if that's possible or it will be the loss coming of a new kind of technology. >> no have to say that matt ridley in the long run things get better and with every technology there disruptions that 50 years and now things are going to be a lot better and we will have figured this out however for the next 10 or 20 years and probably for the rest of my lifetime not the younger people here i think what has happened to us is the tower of babil was destroyed between 2009 in 2012. in then the story of the bible god says they may not understand

one another and it's always difficult to find the truth. i think after 2012 we are in babil and social media has made it impossible or people that. alternate narratives within every university and everything is a battle with the narrative force. i think we will never again -- i shouldn't say never again in the next 20 years we will not yield to find shared truth and i'm also saying good research such as yours and their ability to agree on the facts and even social bias is being compromised by it. i would say to julie long-term optimism and you have to be an optimist long-term but short-term i am very pessimistic. so i guess -- short-term i'm very pessimistic that long-term, long-term it will be great. >> chris your book is wonderful

and very important and i urge everybody to buy it and read it. they think you have the last word. >> thank you jonathan for a fantastic event and rolling with the punches tonight. we had a few punches. and audience thank you for rolling with the punches. we really appreciated and if you missed any portion of the beginning of the event it's going to be available on our facebook page @strand book store and you can watch the portion you miss. we have a link to the chat to chris' book and if you haven't already purchased it. any final words before you we close chris? >> please buy the book and support independent bookstores as much as ever because this might be on c-span i have to say hi mom and thank you everybody

and thank you john it's been a pleasure and a real treat, thann

c-span2. >> today we continue to focus on ai innovation to fresh exploration of ai history. as well as the choices that we face of digital citizens. before ai began to change your world for better or for worse.