Select text to annotate, Click play in YouTube to begin
00:00:00
our next speaker is damon he is a professor of communications sociology and engineering at the university of pennsylvania where he is the director of the network dynamics group
00:00:13
he is a leading world expert on social networks and behavior change and today he will talk about creating change with that said we have 50 copies of damon's book change how to make big
00:00:27
things happen to raffle on thursday so stay tuned damon stage is yours thank you i'm going to talk to you today about how
00:00:40
people change and how institutions change um and the way i want to start this is by asking you just for the basic question about individuals and collectives so when you look at people adopting a new technology
00:00:52
we tend to think of people as individuals much like these as individuals much like these fish interacting in a large social population
00:01:04
one thing to think about is if you were to try to study these fish one at a time could you ever figure out what they would do when you put them all together in this large group and what you see when they're in a group is this magnificent complex collective
00:01:17
pattern of behavior it would be very hard to predict if you just study these fish one at a time and so my view is that in order to understand society and social change and really the institutions that people form you have to study how people interact together
00:01:30
and what the forces are that create the collective coherence or that can undermine it and disband collective coherence and today the way we tend to think about this is in terms of networks what binds people together are the
00:01:42
patterns of connection between them and the people they know and the people they know and so forth and when we look at networks there are several key features that we think are the most important things for understanding human behavior and human behavior change
00:01:55
and typically we pick out the people who have very highly connected networks who are in the center of the network and they've got you know ties reaching out in all directions and then we look at people who are in the periphery of the network um shown in
00:02:09
the right who have many fewer connections but also those connections are sort of clustered together so it's not just the number of connections that differs between the center of the periphery but also the structure of those connections that's different
00:02:21
and when we think about any kind of technology diffusion process or innovation process we have a kind of common idea of how it works we say well typically someone who's highly connected adopts first and that person influences
00:02:34
the people around them and from that initial process then it spreads very effectively out towards the outer skirts of the network and then saturates the rest of the population so that's our classic model of spreading
00:02:47
what i want to suggest is that a problem has arisen in the last 20 years we've started to get really good data on how social movements spread how new technologies like twitter and facebook spread and really
00:03:00
what we tend to see is that when technology or new ideas new political movements take hold they tend to take hold out in the periphery instead of the center and then when they spread instead of jumping to the center what they tend
00:03:11
to do is they tend to spread around the edges of the social network and they only reached the center of the network at the very end of the process so now you've got a very deep uh scientific problem that actually means
00:03:25
something for every industry that works in technology and behavior change which is that our old model on the left that things spread from the center outward is in total contradiction with all of
00:03:37
our best data on the right which shows that new ideas innovations tend to spread from the outside in so how do we solve this problem what's going on here fundamentally is about how networks work and our
00:03:50
intuitions when we look at the network on the left a highly connected person located in the center of the population it just looks like us looks like a fast network we see one
00:04:02
person with lots of connections and we say if this was uh we can make a choice about who to target to spread our idea or to influence the population we would target that person at the center of that
00:04:13
cluster that fireworks explosion because they have lots of ties going in all directions but what about the network on the right well what i want to talk to you about today is that the network on the left is really good for spreading diseases
00:04:27
and many of our ideas come from that but the network on the right is actually the better network for spreading social change new technologies and new ideas and the reason for this is that almost all of our thinking about networks and
00:04:40
about behavior change is based on epidemiology it's based on the science of disease spreading and so when we think about the spread of something like covet 19 or the spread of measles we have a very clear intuition we think
00:04:52
well if someone gets sick and they come into a contact with a lot of people then all those people get sick too and again that works well for describing disease spreading but when it comes to social change or
00:05:04
changing the way people think we have to also take into account the social context in which people interact political boundaries social boundaries social groups social identity and the way that those things interact with peer
00:05:16
feedback and the dynamics of personal change interacting with the people around you and that process is much more complex in the way diseases spread and so i'm going to talk today about virals
00:05:28
spreading and why it fails us when we try to understand human behavior and then i'm going to focus on like the biggest concept for today the main concept that's in the book and the main concept of this talk
00:05:40
which is the difference between a simple contagion which is what we've talked about historically like diseases and complex contagions which is what we really care about when we think about social behavior change and institutional
00:05:53
change what this does is reveal several myths of diffusion and i'll show you how once we understand this new theory of behavior change it allows us to understand how to trigger tipping points in social behavior in institutions um
00:06:07
and in communities at large fundamentally one of the big ideas in the last couple decades has been nudges and i want to show you that nudges is actually a very small concept that just
00:06:19
applies to one individual at a time and we should really be thinking about our norms which is just like the schooling fish instead of thinking about how one fish might change its behavior we want to think about how whole schools of fish
00:06:31
behave and how we can change everyone at once so briefly i'll give you an overview of contagiousness i'll talk about the myth of virality the myth of stickiness and the myth of the influencer and then i'll
00:06:43
talk about how to trigger tipping points so contagiousness we can think about it in a very basic way you can say what is the difference between something like a social norm which is the kind of thing we want to change and something
00:06:55
that we're familiar with like a viral video or in many cases you know covet 19. say well for something like a disease or a viral video if one person has it they come into contact with somebody else and it's very easy to
00:07:07
spread it there's no reason why the person you know shouldn't get sick or shouldn't um check out the viral video and then spread it to the other people they know and then from there it spreads throughout the network from person to person now this is our classic model of
00:07:19
spreading which is that every link in the social network is equivalent to a transmission now again this works really well for diseases for gossip for viral videos but there's some uh very important assumptions here one is that every time
00:07:32
there's a contact there's a transmission and we know of course that's not true right we in fact we we know that in in certain cases with with behavior the opposite happens when you come into contact with someone who's doing it you
00:07:44
actually become less likely to do it particularly if that person like belongs to a different political party or has differences from you that make that behavior less salient as a result of that interaction also we assume that there's no resistance and the thing
00:07:56
that's spreading is easy familiar low cost whatever contact equals adoption but now imagine that the person who has adopted has wearing a face mask but all the other people shown in gray
00:08:09
surrounding that center person are not wearing face masks now because those people aren't wearing a face mask they exert a kind of countervailing influence on the central person but the central person is aware there's a norm in their
00:08:22
community and this person wearing a face mask is violating the community's norm so the non-adopters don't have to do anything they just have to sit there and not adopt and they put pressure that prevents the center person from adopting
00:08:35
and so even a second person can adopt and there's still enough pressure from the non-adopters to prevent the person from adopting finally a third person may adopt and then ultimately the center person may think okay this is
00:08:48
legitimate enough this is credible enough that i'll adopt it too and this is the process of complex contagion and so unlike simple contagion like the virus complex contagion relies fundamentally
00:09:00
on the relationship between the adopters and the non-adopters and feelings of legitimacy and credibility and so of course it's always the case that people require multiple contacts but more important than that they have
00:09:12
to feel like the resistance to change is overcome and this is the truth whenever we uncover um difficult or unfamiliar or costly behaviors whether they're technologies or changes in health
00:09:24
behavior or unfamiliar ideas and so what happens is once we appreciate the difference between a simple contagion and complex contagion it kind of pulls from our eyes some of the classic myths that we've always believed in
00:09:37
one is the myth of virality which is that everything spreads like a virus the second one is the myth of stickiness which is that you know all you have to do is design the perfect product with all the right features and everyone will want to use it and the third of course
00:09:49
is the myth of the influencer which is all you do is find that right one person and they'll spread your idea to everyone so let me disabuse you some of these myths and show you when these myths go away what is left is a new theory of how
00:10:03
to institute change and how effective it can be so the myth of morality really starts with this idea that if we can just introduce something into a population it should spread all by itself in the 1960s lots of different countries
00:10:15
were trying to introduce contraception which is a which is a difficult cell these were countries that were largely agrarian uh families routinely had five to six children and that was because there was high infant mortality so there
00:10:27
were large uh cultural and social norms built around having you know large families and having um very successful children and then having your children be successful and having lots of children and so forth and that was
00:10:42
considered to be a mark of honor so in the 1960s lots of these countries were going through the demographic transition there was an influx of vaccination of um health and sanitation of
00:10:54
industrialization right all of a sudden kids were living longer and infant mortality just dropped off precipitously which men that would if families were having five to six kids all those kids would survive and then all those kids
00:11:06
would have five to six kids so if you do the math within two generations you're gonna have massive overpopulation starvation disease it's extremely dangerous so these countries pakistan indonesia and korea among others tried to
00:11:19
institute these massive um contraception efforts to you know change the way everyone was thinking you can think of this in some ways as like a technology campaign like they were trying to get everyone to adopt a new technology that was you know unfamiliar and that was
00:11:32
going to change their lives but was you know ultimately for the good um the problem of course is that they were going up against long-standing cultural norms and so the lots of countries tried the standard thing that
00:11:45
you would try which is they tried like advertising campaigns mass media marketing they tried to shame people into having fewer kids you know it's bad to have too many kids and all these countries struggled um
00:11:58
with with uh with fertility rates for decades and the exception is korea and this is what we're going to focus on because korea um within 20 years of starting its contraception campaign
00:12:11
reached all of its policy goals so that's that's remarkable you can put on any national scale to change the entire culture of a country in 20 years is an impressive feat let alone for something
00:12:23
as entrenched and and difficult to address as you know family planning practices um if you compare that to the u.s like for example the u.s has a war on drugs it's been running since the 1970s
00:12:35
and you know in 2015 congress finally admitted that almost after 50 years basically nothing had happened and if anything the problem had gotten worse right so it's very very hard to institute large-scale uh change in such
00:12:49
a short period of time so what did korea do that was so effective that's what we want to understand and hopefully glean some lessons from and what they did was they focused on social norms so essentially there were three kinds of of networks or three kinds of village
00:13:02
structures that they saw one was that people were tightly clustered into these like tight-knit village communities with let's say a single tie or a single broker person who would go back and forth between one part of the community
00:13:14
or village and another part and what happens if you got a couple of early adopters of contraception within one cluster they could convince each other to uh you know participate in this in this idea
00:13:27
and then it would take hold within that community but um what would wind up happening is that it wouldn't spread across this sort of single weak tie um now the reason is because this is a
00:13:40
complex contagion that requires social reinforcement now over here in the urban network there is not enough social reinforcement at all to even get the innovation started it would just sort of be a non-starter in that community then
00:13:52
they discovered these villages that had this structure that wound up being one of the major discoveries of research during this era which is that they had these kinds of clusters in the community um that could
00:14:04
create reinforcement for this uh innovation then they also had wide overlapping bridges they could allow the innovation to kind of spill over to other communities and then when people adopted it would spill over and what's interesting is here the spreading
00:14:17
process wasn't like a viral spreading it didn't spread everywhere in all directions at once it just spread from part of the community to other part of the community it's very organic way but it was incredibly effective at saturating the entire population what's
00:14:30
so interesting is that lesson from korea translates directly into thinking about organizations and organizational change because organizations by and large are structured in a way so you've got different groups different divisions whether it's the engineering division in
00:14:43
the sales division or different divisions in terms of their regional location and they're often siloed they don't have a ton of communication back and forth so group a doesn't want to talk to group b or group b to group c
00:14:55
and so one of the major solutions that scientists came up with was to say well you should have someone who takes on the role of a broker and the idea was this person robert could show up in an organization and start
00:15:06
making ties over to different groups he could make a title elizabeth in group a and emily and group b and robert all of a sudden would be kind of the person or the conduit through which all the information spread across an organization so he would perform this
00:15:20
tremendous service for the organization by letting information spread across all these different groups in doing so robert would gain a lot of power right he'd get known as the guy who was the broker for the organization so more people would come to him and he'd get more ties and he'd have this
00:15:33
sort of spanning set of networks that reached across the organization very much like the fireworks network i showed you at the beginning with the influencer this is um kind of business management how to get ahead with networks 101 it's a very easy
00:15:46
way for robert to gain status and prestige and power in the organization while also helping the organization but here's the thing information transfer is not the same thing as knowledge transfer while robert
00:16:00
and his networks are very good at spreading information gossip new ideas from parts of the information other parts they're not very good at getting one group in the organization to start taking up the innovations that are that
00:16:13
are started by another group so if group a let's start let's say their depth of new management practice that's difficult to use but once they got a hold of it that actually allowed them to be much more productive and elizabeth tells robert about it and
00:16:25
robert tries to convince the people in group b well the people in group b haven't seen this and the only person they've heard about it from is robert now they trust robert they like robert but they also know that robert's main uh
00:16:37
advantage in the organization is that he's known as a broker as someone who spreads information so they're a little skeptical about this new management practice because they know that robert gains a certain amount
00:16:49
of social status because he's able to spread these things so what they want is like a more honest take or a more neutral take on this technology and whether it works now imagine the networks now look like this now there's wide bridges multiple
00:17:02
reinforcing ties between group a and group b so now the people in group b can observe the people in group a working with each other with this new technology and they can also talk to each other about what they are seeing whether they
00:17:15
agree that it's working well whether that technology and you know the way it works for people in group a would also work for them in the same way for group b right and they can start to come to consensus about whether or not they do want to adopt it and if they do then
00:17:27
they can coordinate much like we saw in the korean contraception case they can coordinate on adopting it and then spread it through their group and what's interesting about this is robert has less power than he did before right robert would prefer that the bridges
00:17:40
were narrow because he controls them now the bridge is wide there's all kinds of overlapping influences but this is good for the organization because it means that reinforcing ideas and new innovations can spread effectively
00:17:53
between groups it also means that if robert leaves the organization the networks don't disappear right because one thing that makes robert powerful is that when he's the only broker he controls all those networks
00:18:06
and if he threatens to leave he actually has a lot of value right because he's the exclusive holder of those networks so that makes those networks kind of fragile but if you have a wide bridge now robert can leave the organization
00:18:18
and you still have a structure that can support innovation exchange between different parts of the organization so this structure of wide bridges turns out to be one of those sort of more important concepts for thinking about how to structure an organization
00:18:31
effectively to make it adaptive and innovative now it turns out that it's not just within an organization but within industries so if you think now about the connections between different firms in
00:18:42
silicon valley or different research teams working on the human genome project or different schools and universities departments working on the field of network science these lessons are all the same in every case the
00:18:54
success of these initiatives have come because um organizational boundaries that used to be very very strict and kept kind of people siloed the same way that you know in silicon valley different firms initially had very high boundaries like
00:19:08
ibm wouldn't let people from other firms come in and talk to their engineers about projects there was project secrecy there was intellectual property all these reasons to keep boundaries all of a sudden silicon valley
00:19:21
copying like this japanese model started letting people from competing firms come into their firm and work on teams together and they would allow them to be much more innovative because they could sort of harness the ideas from both teams but they weren't just spreading
00:19:34
information they were actually changing the way they were doing business as a result of these sort of co-teams these wide bridges across the organizations and this is one of the key things allowed silicon valley to be so
00:19:46
productive so fast and it's also part of what people know now as the open software initiative has been growing in the same way with allowing networks to form across organizational boundaries interestingly
00:19:58
it's also the same structure that allowed black lives matter to take hold so effectively and to spread from all different parts of the country from ferguson missouri all the way across the u.s and then ultimately all the way
00:20:10
across the world and it's one of those things that no one really saw coming but it's a direct function of the structure of the social media networks that formed around those original events in in ferguson in 2014
00:20:24
so this gives us kind of a big picture lesson about how networks operate both within organizations for the purpose of you know deciding efficient and effective and adaptive technology diffusion
00:20:36
and also you know within citizens and large populations for growing effective social change networks and the main shift in thinking is from individual brokers and simple contagions to
00:20:48
wide bridges between groups and complex contagions so one of the new questions that comes up then is to say okay well let's imagine you've got a structure that's our ideal structure it's going to work perfectly how do i as an organization or
00:21:01
i as an innovator target that structure how do i intervene to sort of seed my i knew i my new idea very effectively well if you look at these red nodes if you thought well i what i want to do
00:21:14
ultimately is get as much exposure as possible from my my idea and each red node is a change agent so wouldn't i locate my change agents as you know widely distributed as possible and that would get maximum exposure
00:21:27
that's simple contagion thinking and yes if you're talking about spreading covet this will be a very effective way to sort of seed covet in the population but what if you're talking about spreading face masks well now each person is
00:21:39
surrounded by people who aren't wearing face masks so not only are they not going to be very effective at convincing everyone around them to change their social norms ultimately you know if you wait long enough these people will give up on the social norm and go back to
00:21:51
doing what everyone else is doing so here's an idea which is to use the theory of complex contagion to take the same population with the same level of resistance in the same network and now initiate change just by changing
00:22:04
your strategy of where to put your change makers here i put all the change makers together in one cluster now from the classic model the viral model of diffusion this seems crazy because basically it's like having a bunch of telemarketers call each other what's the
00:22:18
point they all know they're trying to spread the product but in this case each of the change agents is able to talk to the other change agents and give each other support and confidence that they're doing the right thing and that this is going to work and then they can
00:22:30
coordinate together and talk to other people and provide reinforcing signals that this is a good idea and then what happens is they start to get some amount of take-offs and purchase with the within the small group where they started but then once this
00:22:43
group winds up converging then it spreads the nearby group and then to nearby group and you can see just like what happened in korea you can see this very effective change process taking hold with an organization and then spreading from each of the divisions
00:22:55
within an organization across the entire population so this means that this way of thinking about wide bridges doesn't just tell us what the structure of an organization is it really tells us how to target organizations and target
00:23:08
populations at large so one of the main lessons that came out of korea was that it wasn't just that they were very effective at spreading contraception it was that each of the villages became uh sort of coherent or
00:23:21
converged on a single method of contraception that's what you'd expect you'd say okay you know this village became an iud village and our classic thinking says well that's because the iud is a better technology right whatever technology is going to spread
00:23:34
it's going to be the stickier one so we should expect that to spread everywhere but that's not what happened in one village the iud took off everyone adopted the iud but in the next village the pill took off everyone adopted the
00:23:47
pill in the next village the condom took off everyone adopted the condom what this tells us is what's going on in these villages is a pro is a process of convergence on social norms irrespective of the particular
00:23:59
technology so the answer to change isn't to make a perfect technology the answer to change is to focus on how the social networks can create a norm and this brings us to the myth of stickiness which is the idea that if you just make the perfect product your change
00:24:13
problems will be solved now of course we know historically this is false we've got you know lots and lots of historical cases that you know the dustbins of economic history are littered with examples of better better marketed more
00:24:25
effective more efficient technologies that failed um and they had all the things they're supposed to have they were emotional they had triggers they were public they're practically all these kind of list of features we're supposed to give our product but it turns out that the inferior product
00:24:38
often wins over the superior one because it exploits networks better and so the key here is exploiting the structure of the network a really good example of this this is a um a current example
00:24:51
just prior to the pandemic one of the major challenges in sub-saharan africa was getting people to adopt hiv medication and the big breakthrough the insight was that the things they were trying to do were too hard you know getting people to get
00:25:04
circumcised it just was a cultural profanity or getting people to use condoms it was too much of a burden if they could just invent a sticky technology a pill that was free and accessible they people could take it one you know once a day
00:25:15
like a tylenol then um if that could be effective then that would solve the problem of hiv transmission in sub-saharan africa so they did this they invented this pill and it was thought it was like a miracle drug
00:25:28
and then they deployed this massive massive program across three nations in sub-saharan africa to deploy to you know to give people the pill for free and to show how effective it was at preventing hiv and what they had was like they had
00:25:40
women coming every day to get the pill every or every week to get the pill and then get a blood sample um and they were going to show how effective this was at preventing the infection rate from growing um and what they found was just
00:25:52
the opposite this is one of those major sort of historical backfire events in fact it was so it was so egregiously a failure such a fiasco that the directors of the program wrote an
00:26:04
article in the new england journal of medicine basically announcing what a colossal failure their project had been which is a rare it's a rare article to write at a high tier journal but it was because
00:26:15
there's so much faith in this model of like giving people a sticky product and what wound up happening was that the women would come in they'd give their blood they'd get the pill they'd go home and then they would throw it away right they wouldn't take it and then
00:26:28
that's mind-boggling because there's so much effort involved in being part of this process and this and this trial why would they go through all that and then not take the pill and the reason was because there were lots of stigmas around hiv in their
00:26:41
community and some people thought that taking the pill might give you hiv that sounds like an unusual thought but actually in the u.s lots of people think that getting a flu shot will give you the flu right a lot there's lots of misinformation or we call today fake
00:26:53
news about health communication about facts so people had some misinformation about this pill and that misinformation triggered social norms within the neighborhood
00:27:05
communities within these villages within these nations suggesting that if people were taking the preventative measure they might actually be more susceptible to hiv and then they'd be susceptible to the discrimination that comes from you know the way that people were treated once
00:27:18
they had hiv and so the social norms were much more influential at controlling people's behavior than any sticky product could be and this is where we say okay we kind of throw our hands up in the air we say the product doesn't work variety doesn't
00:27:31
work what can we do and then we all go and say okay let's get let's get the influencer if we can just get the influencer we'll solve all of our problems but that again is a myth and the myth of the influencer brings us back to the sort of central idea was
00:27:43
showing at the beginning which is that there are some people who are highly connected many people who are less connected and of course if you can find a highly connected person then that person is is able to detect
00:27:55
new things innovations in the population early on because they have so many network ties so they detect something and then they adopt it and they spread it to everyone right and that's our model of why we want to find an influencer and again that's a really
00:28:07
good model for how disease spreads right but what goes wrong with complex contagions well what goes wrong with complex contagion is that if you're trying to convince a highly connected person an influencer to adopt something
00:28:20
that goes against the norms that everyone else has that violates people's sense of normal behavior or their sense of legitimacy or their sense of credibility then what all of these ties that are
00:28:32
going out that we think of as like um accelerating diffusion these are actually also ties that are coming in and what that means is that someone who's highly connected is also highly watched they're being evaluated they're being
00:28:46
judged they're being commented on so if they do something that's uh unusual or goes against the norms particularly in their industry imagine this person is a highly connected ceo of a fortune 500 firm
00:28:59
well their reputation is enormous like they really worry about how they're perceived in their professional community so they can't just run around adopting any technology or any management practice without looking at how it's going to appear to their peers
00:29:12
because if it goes wrong they're going to be very embarrassed maybe even lose their job this is one reason why people who are highly connected tend to wait tend to see if more and more and more people adopt therefore it's more acceptable therefore they're all adopted
00:29:25
right so the question is well if these people have to wait how do you get something off the ground in the first place to ever get some kind of critical mass and this is where the network on the right comes in because these people are less connected it only takes a few
00:29:37
connections to create a sense of social influence they have fewer countervailing influences and also because they've got neighbors of neighbors and neighbors who are clustered together it's easy to gain kind of a critical mass around a new
00:29:50
idea that can they can then start to spread outward now this has been true for decades even since the 70s this idea that like major institutional change comes from the periphery or comes from these less
00:30:02
connected people has been studied but the problem has been that we've had this theory of influencers this theory of weak ties and it's prevented us from really seeing how these things work and so it's really just in the last decade with all the data from social media
00:30:15
we've been able to see really really clearly just how important the periphery of the network is for influencing everybody else and so this gives us i think a really important insight into tipping points this is kind of where we
00:30:27
close the talk i want to show you that if you have a network that looks like this and you say well what if we could trigger a tipping point and this is very much the you know the idea of most of our brand marketing strategies we say well you know maybe
00:30:40
hard to convince the influencer they may resist it but if we could give them enough money then they should be able to spread our our product you could say well look if what you're spreading is like a new brand of coconut water something trivial and easy a simple
00:30:53
contagion sure then you know give it to the influencer and they'll just spread it and it will explode virally across the network and that's fine but the problem is when you move to accomplish contagion it doesn't just
00:31:05
fail this strategy can backfire and i mean really hurt your organization and the famous case of this is google glass it's something i talk about in the book right you know kind of walk through the story of you know made sense to google to create this exciting memorable sticky
00:31:19
product this cyborg technology where you could you know browse the web and record the visual field just with a pair of glasses it seemed really cool and then the idea was not only to make it sticky but also to use
00:31:32
the influencers to spread it so they would target these elites in the silicon valley community and made it a very expensive technology 1500 glasses and you know the idea was that everyone
00:31:44
was going to see these techie elites wearing this cool sticky product and they were going to want it too right i was going to create aspiration what happened was exactly the opposite it was that using a set of eyeglasses to record the
00:31:56
visual environment while you were talking to people was seen as a violation of privacy norms also is seen as just kind of rude to browse the web while you're in like a face-to-face conversation with someone and in fact the term glasshole was invented by
00:32:09
regular citizens as a way of characterizing the influencers who are wearing this technology and so you got this massive cultural backlash that turned the influencers from the people we're trying to imitate into the people that we
00:32:22
didn't want around us the people we were rejecting and so um the entire product line was shut down but worse than that the reputation of the company google actually took a hit as a result of this product right so this is an example of
00:32:34
you know enormous backfire so now we have a way of thinking about this that can change this outcome hopefully and change our way of approaching social norms if you try to influence a population where there's
00:32:46
resistance to a new idea the influencer may get one or two people around them but ultimately those people are also surrounded by countervailing influences there's just too much pressure pushing back but if you take the same technology the
00:32:59
same population with the same level of resistance and just change your targeting strategy and now target this group out here which seems like it has nothing to do with any any part of the population is relevant for change but what you can do is identify that
00:33:12
this cluster of people has lots of wide bridges around the periphery and what you'll notice that's very interesting is once this technology takes hold in the group it starts to spread but it doesn't just jump to the center right away it
00:33:24
starts to spread through these patterns of wide bridges along the structure of the network and then it builds a critical mass and once it gets a critical mass then it comes through the center of the network and takes over the population and this is really this new
00:33:36
theory of change how to target social norms to change populations as a whole implementation really means that the key concepts here are reinforcement and the main things you're trying to do are achieved legitimacy and credibility and
00:33:48
wide bridges are the way to do that ultimately you're also looking for relevant peers it matters whether you know if you're you know from one political party or another from one social economic group or another people are looking for people who are relevant to them when making these kinds of
00:34:01
decisions and finally when you're looking at technology diffusion or product diffusion you have to remember that you're also fundamentally looking at norm diffusion you have to think about it from that perspective of social norms
00:34:12
so ultimately i would say that i'm hoping that now that when you look at this you'll see something new you'll see these two networks but you'll also see the network on the left which initially looked like it was going to be fast for everything you can see that and
00:34:25
say okay that's going to be fast for a simple contagion it's going to be fast for disease but when it comes to social norm or something that requires reinforcement or something where countervailing influences matter the network on the
00:34:38
right is actually going to be the place where it's going to take hold better and this really changes our whole concept of spreading innovation because now we have to appreciate the innovations are often unfamiliar they require complementarity we have to do them together we require
00:34:51
coordination and ultimately people are looking for these bridges that can help to sort of convey these ideas or innovations from group to group and this also changes our thinking about nudges because nudges are things we do one at a time we give
00:35:03
everyone a nudge but if people are paying attention to each other then an individual nudge isn't going to be that effective we have to think about the way that norms operate in the community and how we can shift the norm and in in many ways what we're doing is
00:35:16
developing a new theory of critical mass a new way of triggering tipping points for initiating large-scale change um so thank you very much and i look forward to taking your questions
00:35:33
thank you so much this was a very enlightening talk and i will surely have to watch it again to to make sure i got all of the information out of it
00:35:45
thank you very much with that said we have the first question here why do you think misinformation spreads so much more effectively than true facts in the same network how
00:35:57
can we utilize the spreading methods of fake news for spreading useful information yeah that's a great question i addressed that topic toward the end of the book the thing about misinformation is that
00:36:10
it tends to be a simple contagion why is there a simple contagion because misinformation can say anything all it has to do is be tailored to a norm or a bias or a set of beliefs that
00:36:23
a community already you know holds so if i develop a fake news story or piece of misinformation and it's completely consistent with what you already believe now it's not true
00:36:35
but oftentimes the truth is more challenging than fiction because fiction can be a story that tells us what we already think what we already like tells us what we already want to believe
00:36:48
and so that's very easy to adopt and very easy to spread because you're basically just telling people what they already think it's confirmation it's it's it's confirmation bias but confirmation bias usually also means
00:37:00
that you're you're saying well i'm seeing something true and seeing something false and i kind of select the false thing but this is saying that the false thing actually just spreads faster it's easier to repeat it's easier to explain it's easier to promote because
00:37:13
it makes sense to us right we understand it it's a story that resonates the true story is often kind of subtle and complicated and difficult to understand and if we try to repeat it we kind of feel like we don't quite know
00:37:25
what we're talking about and so it's actually harder to get that kind it's a complex contagion we require kind of reinforcement from a couple of people to feel like we actually do know what we're talking about this is what it means and we need their help to be able
00:37:37
to spread it so a good example of this is like when the vaccination the vaccine first came out for covet 19 um there were all of these you know conspiracy theories but the conspiracy theories resonated with people's
00:37:49
existing beliefs so in the u.s there's this enormous historical tension with the african-american community in public health and the reason for that tension is that the public health community has committed like atrocities against the african-american community some of them
00:38:02
you know about they're they're famous like these experiments where they gave black men syphilis just to see what would happen but some of them are less well-known about and more pernicious like um involuntary sterilization of black women it's been going on for
00:38:15
decades in the u.s right so you've got this like really tortured and kind of lack of trust in the relationship between public health and the african-american community so then covet comes out the code vaccine
00:38:28
comes out and now the african-american community is very susceptible because of their fundamental mistrust of public health to misinformation to conspiracy theories about what's really going on and whether the vaccine is really safe
00:38:41
right and so you say okay well there's a history of mistrust so it's understandable that you wouldn't take the news story which is complicated and very specific at face value you're going to tell a
00:38:53
more complex story that resonates with your cultural norms and resonates with the history of your culture that's a true history now that's a very difficult challenge because how do you get true news into a community that has good reason for
00:39:04
resisting whatever the the sort of medical authorities are saying and the answer to that is is basically networks is that what you want to do is to change the network so you don't have influencers spreading stories that are
00:39:17
familiar to everyone you want to sort of get into the corner of the network the the periphery where you can sort of get some reinforcement for a story with some subtlety okay the vaccine works but it only works with people of a certain age and it works but only for a certain
00:39:30
amount of time so you have to get a second shot and you get a booster shot right these are nuances that are much harder to understand than some of the conspiracy theories like 5g towers cause coded right and
00:39:42
five but it sounds crazy to say that but the the cons that particular like fake news meme spread around the us and their people were talking about it everywhere 5g towers cox coveted and it's it's easy to say it's kind of fun to say because
00:39:56
it seems like there's a you know some underlying relationship between technology and disease which is mysterious but you know it it sort of um is consistent with some of the suspicions that a community may hold um
00:40:09
and so what you see is kind of different fake news stories spreading through different communities because the stories were kind of tailored to those communities whereas the truth is going to be the same for every community so
00:40:20
getting the truth to take hold in each community is a bigger challenge but again there are network strategies that can make that more effective right thank you so much i believe that was a really effective
00:40:32
answer the next question is can you resolve this with rogers slash moore's model of diffusion of innovation especially with regard to disruptive
00:40:45
technologies or should we retire that model yeah we we should retire that model and if i can um if you'll forgive me stopping my screen share i want to see if i have these slides here
00:40:59
all right sorry i don't think i do have them here so um there are several other slides that help to explain help to explain this this process of
00:41:10
change that specifically show why that model fails and oh you know i do have those slides i think i'll show them to you that explains why why that that process fails and how we can make it succeed
00:41:23
um i'm going to turn my uh my presentation back on in a second all right we will share it with your audience okay so here we go
00:41:37
okay and then if i can go back to here and slideshare i apologize it'll be worth it once this these slides come up all right
00:41:56
well um the basic story is that uh what rogers had talked about was basically there are sort of um different uh types of people and but if you could sort of identify that you know the early
00:42:09
adopters early on that you could um trigger a change process effectively here we go you could trigger a change process effectively and what i think is really
00:42:24
wrong about that is that is that ultimately uh change isn't really about getting certain people to adopt it's about getting certain parts of the social network to adopt and so what you want to do is very much like
00:42:38
what i was showing you before is figure out how to engage different parts of the social network there we go okay okay so the screen share okay so tipping points for change we have the story of
00:42:56
black lives matter we have the story of uh the growth of support for same-sex marriage but we also have this the story of twitter and turns out these three stories are extremely similar the way they operate in networks the way the
00:43:09
diffusion process worked and what we see in all of them is that there's this beginning slow growth curve then there's this sort of elbow where it hits what we call the tipping point it grows very very fast and then it slows off at the end and this sort of s-shaped
00:43:21
growth verb something we see in every kind of change process within organizations across organizations with social movements with all those technology adoption the same in every case and what rogers had said was look
00:43:33
what's going on here is that during this slow growth period really it's just that you have early adopters people who are eager to adopt then what you do is you get the regular adopters and that's a lot of people so that's when it gets
00:43:45
really you know accelerated and then you have the laggers everyone else who's kind of reluctant to adopt and they adopt at the end so this is a very compelling story because it kind of assigns types of people to this massive you know
00:43:58
change dynamic it's completely wrong and more importantly there's no empirical data to support it there's no study that's actually gone through a diffusion curve and identified who the adopters were at each stage this is just kind of
00:44:11
an anecdotal description of a diffusion curve based on a kind of you know casual psychology of different types of people it's compelling because it's intuitive but it's wrong right what's really going
00:44:22
on here is this slow growth part is a different part of the social network it's the structured periphery which is what i was showing you it takes hold in the outer part of the network and doesn't just jump across the center it kind of cascades slowly around the edge
00:44:35
edge building a critical mass this is key because the structured periphery and that sort of way of looking at networks actually is our is our key to understanding critical mass and tipping points because when the structured
00:44:47
periphery gets saturated that's when you hit a tipping point and then that's when you go across the center of the network this is when influencers get involved this is like you basically built enough of a critical mass that now you see spreading
00:44:59
and this is one of the reasons why we make the mistake where we say oh look we see an influencer and then very quickly after that everyone's adopted well you've sort of started looking at it late in the diffusion process because
00:45:12
the question is how does it grow to be so successful that the influencer actually takes notice that the influencer feels comfortable adopting it and this is true for every technology and in the book i have the story of twitter um which is also exactly like
00:45:25
this where twitter was like for years just growing growing growing until it started reaching this sort of exponential hyper-exponential growth and as it was at that point that like oprah adopted but she adopted when the
00:45:37
technology had already exploded it was in as fast as part of its growth curve and then after she adopted it actually slowed down because that was sort of the network process taking it's taking its sort of path to completion
00:45:50
and so if you look at it in this way this is the figure i showed you before what you're seeing here with this cascading around the edge is that kind of shallow growth part before you reach a critical mass this is the structured periphery of wide bridges and again
00:46:03
remember it doesn't just jump to the center you need to reach a critical mass before that happens once you reach a critical mass this is where you see it starts to speed up and go right through the center of the network and this is the this is the sort of elbow or the
00:46:14
tipping point and then ultimately reaches the unstructured periphery this is true for everything and in the book i talk about this you know for solar panels for bias medicine for political polarization for the metoo movement for vaccination and
00:46:26
so forth and so it becomes kind of a general story of understanding how networks operate in a way that the ever rogers story kind of overlooks thank you thank you very much for that
00:46:38
um i have one more question for you um how do you recognize the place in the organization where the idea or norm etc shall be injected
00:46:51
i mean how do you yeah i get it there's a second i mean how do you recognize the starting point of the spreading to reach the maximum speed yeah i understand um so we just
00:47:06
published a paper in nature uh in july that answers that exact question and also that paper was covered in um fast fast company so there's a bunch of news articles about it
00:47:19
basically we developed a mathematical formula where we can look at a network within an organization or within even an industry and pinpoint the exact little neighborhood of like five or six people
00:47:30
but if you're able to activate them it will effectively change the norms in the entire population and so it's essentially it's a group in the periphery but it's a very specially located group and what you're looking
00:47:43
for really is wide bridges you're looking for all the wide bridges across the the organization and there's a there's a place in every organization you can always find it there's a place where the wide bridges intersect now
00:47:55
what's so interesting about this and i talk about this towards the end of the book is what's what's interesting about it is that there's no one person that's special in that particular group it's not as if
00:48:07
they're more highly connected or they are more charismatic or anything like that the individual people in that group are like normal folks what makes them special is that their group is located at the intersection of
00:48:20
more wide bridges than any other group in the entire organization and so you basically do this mapping technique where you map all the wide bridges and it just reveals this spot and once you start with that spot then
00:48:33
it triggers change you know in this way that i just described throughout the organization ultimately hits a critical mass um so we we ran this on a series of empirical studies and we did 72 different empirical networks and then we
00:48:45
we compared it to data from historical studies of program management in india and the data are really striking because every single empirical network you look at you can always identify that specific spot and
00:48:59
show that specific spot is more effective than any other location and we compared it to all the traditional measures so classic measures of centrality you know degree centrality eigenvector centrality betweenness centrality they couldn't even compare but this
00:49:12
measure is just like head and shoulders above in terms of being able to initiate these kinds of change processes thank you thank you very much for answering those this is what we have time for right now but before we let you
00:49:24
go i would like to give you our present for you and we will send this to you and i hope you like it and thank you very much for joining us and thank you for this very very educational
00:49:38
talk i'll definitely have to uh watch it all over again um and viva vida said thank you so much for joining the stretch conference
End of transcript