Tipping points and beliefs – the 10% solution

From the Rennsselaer Polytechnic Institute

SCNARC visualization

In this visualization, we see the tipping point where minority opinion (shown in red) quickly becomes majority opinion. Over time, the minority opinion grows. Once the minority opinion reached 10 percent of the population, the network quickly changes as the minority opinion takes over the original majority opinion (shown in green). Image credit: SCNARC/Rensselaer Polytechnic Institute.

Minority Rules: Scientists Discover Tipping Point for the Spread of Ideas

Scientists at Rensselaer Polytechnic Institute have found that when just 10 percent of the population holds an unshakable belief, their belief will always be adopted by the majority of the society. The scientists, who are members of the Social Cognitive Networks Academic Research Center (SCNARC) at Rensselaer, used computational and analytical methods to discover the tipping point where a minority belief becomes the majority opinion. The finding has implications for the study and influence of societal interactions ranging from the spread of innovations to the movement of political ideals.

“When the number of committed opinion holders is below 10 percent, there is no visible progress in the spread of ideas. It would literally take the amount of time comparable to the age of the universe for this size group to reach the majority,” said SCNARC Director Boleslaw Szymanski, the Claire and Roland Schmitt Distinguished Professor at Rensselaer. “Once that number grows above 10 percent, the idea spreads like flame.”

As an example, the ongoing events in Tunisia and Egypt appear to exhibit a similar process, according to Szymanski. “In those countries, dictators who were in power for decades were suddenly overthrown in just a few weeks.”

The findings were published in the July 22, 2011, early online edition of the journal Physical Review E in an article titled “Social consensus through the influence of committed minorities.”

An important aspect of the finding is that the percent of committed opinion holders required to shift majority opinion does not change significantly regardless of the type of network in which the opinion holders are working. In other words, the percentage of committed opinion holders required to influence a society remains at approximately 10 percent, regardless of how or where that opinion starts and spreads in the society.

To reach their conclusion, the scientists developed computer models of various types of social networks. One of the networks had each person connect to every other person in the network. The second model included certain individuals who were connected to a large number of people, making them opinion hubs or leaders. The final model gave every person in the model roughly the same number of connections. The initial state of each of the models was a sea of traditional-view holders. Each of these individuals held a view, but were also, importantly, open minded to other views.

Once the networks were built, the scientists then “sprinkled” in some true believers throughout each of the networks. These people were completely set in their views and unflappable in modifying those beliefs. As those true believers began to converse with those who held the traditional belief system, the tides gradually and then very abruptly began to shift.

“In general, people do not like to have an unpopular opinion and are always seeking to try locally to come to consensus. We set up this dynamic in each of our models,” said SCNARC Research Associate and corresponding paper author Sameet Sreenivasan. To accomplish this, each of the individuals in the models “talked” to each other about their opinion. If the listener held the same opinions as the speaker, it reinforced the listener’s belief. If the opinion was different, the listener considered it and moved on to talk to another person. If that person also held this new belief, the listener then adopted that belief.

“As agents of change start to convince more and more people, the situation begins to change,” Sreenivasan said. “People begin to question their own views at first and then completely adopt the new view to spread it even further. If the true believers just influenced their neighbors, that wouldn’t change anything within the larger system, as we saw with percentages less than 10.”

The research has broad implications for understanding how opinion spreads. “There are clearly situations in which it helps to know how to efficiently spread some opinion or how to suppress a developing opinion,” said Associate Professor of Physics and co-author of the paper Gyorgy Korniss. “Some examples might be the need to quickly convince a town to move before a hurricane or spread new information on the prevention of disease in a rural village.”

The researchers are now looking for partners within the social sciences and other fields to compare their computational models to historical examples. They are also looking to study how the percentage might change when input into a model where the society is polarized. Instead of simply holding one traditional view, the society would instead hold two opposing viewpoints. An example of this polarization would be Democrat versus Republican.

The research was funded by the Army Research Laboratory (ARL) through SCNARC, part of the Network Science Collaborative Technology Alliance (NS-CTA), the Army Research Office (ARO), and the Office of Naval Research (ONR).

The research is part of a much larger body of work taking place under SCNARC at Rensselaer. The center joins researchers from a broad spectrum of fields – including sociology, physics, computer science, and engineering – in exploring social cognitive networks. The center studies the fundamentals of network structures and how those structures are altered by technology. The goal of the center is to develop a deeper understanding of networks and a firm scientific basis for the newly arising field of network science. More information on the launch of SCNARC can be found at http://news.rpi.edu/update.do?artcenterkey=2721&setappvar=page(1)

Szymanski, Sreenivasan, and Korniss were joined in the research by Professor of Mathematics Chjan Lim, and graduate students Jierui Xie (first author) and Weituo Zhang.

Advertisements

  Subscribe  
newest oldest most voted
Notify of
TimG

So what happens with two groups of 10% with mutuall exclusive ideas exist?
That is the normal scenario.

Bob Layson

Seems to me that ten percent leaves room for other equally committed, and opposed, minorities of ten percent or more. Hmm?

richard verney

If true, in the case of political and/or ideological idea, this is very worrying.

Allanj

There is something odd here, or perhaps I miss the point. If an opinion below ten percent popularity has no chance of growing to a majority how does it ever get to the ten percent (tipping point) level where it inevitably become a majority opinion?
To suggest that every opinion that reached ten percent will inevitably become a majority opinion is counterintuitive. There are many political parties in European parliaments that seem to be stuck in the ten to twenty percent range. Like Andy Warhol’s 15 minutes of fame, will each of these splinter parties have their 15 minutes of majority?
Population growth computer models have been around for decades. Maybe it is just too early in the morning, but I just don’t see what this model adds.

“There are clearly situations in which it helps to know how to efficiently spread some opinion or how to suppress a developing opinion,”
Um… we’ve noticed…
Now; is there really an UNDO?

John Brisbin

It is pretty bold, even for a press release, to say, “… when just 10 percent of the population holds an unshakable belief, their belief will _always_ be adopted by the majority of the society.”
Well, at least they are interested in comparing their shiny new computer model with some nasty old historical data.
And they have a pronounceable acronym: Snark.

Blade

“In general, people do not like to have an unpopular opinion and are always seeking to try locally to come to consensus. We set up this dynamic in each of our models,” said SCNARC Research Associate and corresponding paper author Sameet Sreenivasan. To accomplish this, each of the individuals in the models “talked” to each other about their opinion. If the listener held the same opinions as the speaker, it reinforced the listener’s belief. If the opinion was different, the listener considered it and moved on to talk to another person. If that person also held this new belief, the listener then adopted that belief.”

I’d say this part is absolutely correct. Related phenomena include peer pressure and appeal to authority.
In a nutshell it is exactly why the AGW cult is always, without fail, approaching the subject from the starting point of a ‘consensus’. Whether by accident or by design, they are trying to exploit this human foible to its advantage.
Needless to say, but I’ll say it anyway, this ‘argument advantage’ was also prominent in several very famous totalitarian regimes. It was the cornerstone of their ability to maintain order and control the masses. This is the first thing about AGW politics that caught my intention.

Gary Pearse

Ok, similar interactions probably lead to the direction a flock of birds or sheep go in. So the real mechanism arises from the fact most people don’t do their own thinking. In the agw situation when the name of the hypothesis kept changing as predictions failed and the inner workings of the movement was revealed in the climategate affair, the momentum began to swing the other way. Yeah, it works, global warming fell off the radar of more than 50% of the pop. No chance of tipping back now.

Jeff Wiita

Now we know why Social Networking is so important. And, who controls that?

Patrick Davis

Apprently the “tipping” point was 350ppm/v…and atmospheric concentrations of CO2 when the org with the same name was setup, was about 380ppm/v…so…so…we’re doomed, right? I dunno, doom, was a great game, cheat codes and all, in the ’90’s.

Bloke down the pub

This is presumably how the CAGW meme spread in the first place and how the sceptic meme is spreading now. Little ice age aside, are we doomed to swing from one view to the other for evermore?

Tucker

I’m reminded by this article that Hitler came to power in this exact way, with his party starting as a fringe organization and only marginally successful for 15 years until…. So, while the example given was how dictators could fall, so too dictators may rise.
A thought provoking article …

John Marshall

Probably because the minority are so violent in putting over their beliefs the majority let them get on with it for a quiet life.

Hmmmmmm.
“when just 10 percent of the population holds an unshakable belief, their belief will always be adopted by the majority of the society.”
An interesting study on a trigger point for influence, indeed, but the quoted text seems a bit strong. Still, keep up the good work, WUWT, and maintain the polarisation. If the researchers are right, it is only blogs like yours that have prevented the adoption of CAGW as a world-wide religion!

TFN JOHNSON

When I was young, and firms had lunch breaks, I used to amuse myself by trying to ‘steer’ my group of colleagues during the usual stroll around after eating. I found that being on the outside of a group of 6-8 people had no effect when I peeled off towards, say, the canal. But if I had just one person outside me, then the whole group would follow once I edged the outsider left or right. It was not necessary to be in the middle of the group, let alone to try and shove them all.
But another aspect is that if in a debating situation it is likely that that there will be a call for compromise then it pays to set out one’s own view as extremely as possible: then the compromise should favour you.

Dr A Burns

Sounds like a quantification of the Asch Effect.

tobyglyn

“It would literally take the amount of time comparable to the age of the universe for this size group to reach the majority”
Wow, they sound confident, it must be true 🙂

Models. Yawn.
If they want to make such absolute statements, they need to show a pile of actual facts with no exceptions to the supposed rule.
A more likely rule: A 0.1% opinion will become the sole permitted opinion if the 0.1% controls the media. If the group doesn’t control the media, it can include 99.9% of the population without ever being mentioned, let alone adopted.

wermet

I see a problem with the premise of this study. If, as stated, “When the number of committed opinion holders is below 10 percent, there is no visible progress in the spread of ideas”, how can the population ever get the 10 percent or greater required to spread the new ideas to the majority?

Modeling human and social behavior is somewhat less reliable than modeling physical processes like the climate.
But this research is based on the influence of what it calls-
“committed agents who consistently proselytize the opposing opinion and are immune to influence.”
These results emerge from a population who are capable of changing their viewpoint being influenced by 10% who are incapable of responding to any outside influence.
In the field of climate science there are over 90% of scientists who are extremely unlikely to change their viewpoint without direct physical evidence that the AGW effect is being negated by something else.
The small number of scientists who proselytize the opposing opinion and are immune to the influence of the physical evidence that has convinced the >90% fail to reach the 10% tipping point.
It is unlikely that the climate contrarian viewpoint is going to capture the zeitgeist anytime soon.

Peter Plail

“When the number of committed opinion holders is below 10 percent, there is no visible progress in the spread of ideas. It would literally take the amount of time comparable to the age of the universe for this size group to reach the majority,” said SCNARC Director.
Is it me or is this nonsensical? If there is no visible spread in a time comparable to the age of the universe, how does it get to above 10% and then “take off like wildfire”?
Have these researchers come to the amazing conclusion that the more people who hold a view, the more the view is likely to spread? And they get paid for that?

Okay, so what happens when more than 10% of True Believers know the climate is still marching to its own tune and more than 10% of True Believers know that CO2 is evil and will cook us all?
I suspect subgroups will form that comprise less than 10% of one group and each will hold their banner high and throw sheep dung at the other.
Maybe we should offer the climate blogosubsphere as a convenient study object.

JimboW

“……If the opinion was different, the listener considered it and moved on to talk to another person. If that person also held this new belief, the listener then adopted that belief…..”
GIGO, as they say. How about something more like:
…If that person also held this new belief, the listener then thinks “Good grief, I’m surrounded by idiots, but they have the power of job/no job over me, so I’d better keep a low profile, or maybe pretend to go along with it if they put me on the spot”…
Climate models are oversimplifed to the point of uselessness, as, I fear, is this bit of network modelling as well.

H.R.

About that moon landing…
http://articles.cnn.com/2009-07-17/tech/moon.landing.hoax_1_moon-landing-apollo-astronauts-bill-kaysing?_s=PM:TECH
From that link…
“I do know the moon landings were faked,” said crusading filmmaker Bart Sibrel, whose aggressive interview tactics once provoked Aldrin to punch him in the face. “I’d bet my life on it.”
Sibrel may seem crazy, but he has company. A 1999 Gallup poll found that a scant 6 percent of Americans doubted the Apollo 11 moon landing happened, and there is anecdotal evidence that the ranks of such conspiracy theorists, fueled by innuendo-filled documentaries and the Internet, are growing.
Twenty-five percent of respondents to a survey in the British magazine Engineering & Technology said they do not believe humans landed on the moon. A handful of Web sites and blogs circulate suspicions about NASA’s “hoax.”
And a Google search this week for “Apollo moon landing hoax” yielded more than 1.5 billion results.

================================================================
So when will “moon landing is a fake” become the majority opinion? Personally I don’t think it will, but according to the models… (/SCNARC)

Matt

and since about 15% of the US population now openly admit to being atheists that means, according to the theory…..

nano pope

Australian Greens primary vote: a steady 11% We have passed the tipping point Down Under, if you couldn’t tell already. Anyone who isn’t in Australia should get your permits ready, we’re looking to buy $50 billion worth of hot air from overseas. I’m sure it will be as secure a trading environment as Brussels or Chicago.

Beth Cooper

Hmm, from 10% to a consensus. So that’s how our BBC is operating… Clever, huh?

wes george

It’s a bit of a worry that the Army Research Laboratory is interested in funding research on how to manipulate public opinion. Roll over Orwell. These are the same idiots who did LSD experiments and god knows what else to soldiers in the 1970’s.
“There are clearly situations in which it helps to know how to efficiently spread some opinion or how to suppress a developing opinion,” said Associate Professor of Physics and co-author of the paper Gyorgy Korniss. “Some examples might be the need to quickly convince a town to move before a hurricane or spread new information on the prevention of disease in a rural village.”
Yeah, right. The road to hell is paved with, well, intentions, good or otherwise. It might also be useful to help convince rural villagers to board trains bound for concentration camps. Or that we need to surrender our civl liberties to a global autocracy in order to “tackle carbon pollution.”
I hope useful idiot Gyorgy sleeps well at night.
Being a True Believer is the same as being a brown shirt. IMO.
http://en.wikipedia.org/wiki/The_True_Believer

But the scientists used *computer models*! And we know computers are always right! They can predict the climate for decades with total accuracy!
/sarc

Lonnie E. Schubert

Published the first of April? Perhaps the publication date is a typo. Anyway, absolutes nail it as poppycock. Anthropomorphising models seem over the top also. They seem to be saying they built a model to give them a predetermined solution, and then they found it interesting that the calculations ended up showing a tipping point of only 10%. Of course, it would seem arbitrary that they were ever able to reach the tipping point, given the astronomically low probability of reaching it. Of course, Army Intelligence funded it. 😉 I recall an Air Force study considering “true” teleportation.

JimboW 4:44 is on the right track. These researchers should run their models such that each person needs to hear the new idea at least four times before they even remember it is a new idea. then they need to discuss it with a “knowledgeable” source before adopting it.
Then the model needs idiots who will believe anything (5%) and stone heads that will believe nothing (5%).

Opinions are like a bathtub, when they are full of holes, they don’t hold water. 😉

“To reach their conclusion, the scientists developed computer models … ”
Nuff said.

People generally operate under two seemingly opposing dynamics: to get ahead, and to get along.
The subject that one person is inflexible about, if its in disagreement with his broader group, is generally the subject that doesn’t come up in lunchtime conversation, in the interest of “getting along.”. When it does come up, that person generally tries to be persuasive rather that dictatorial, because local group membership is usually more important than conversion of the locals to that person’s broader worldview.
It is when doubts arise in someone’s mind from outside, from elsewhere, that the sole contrarian is sought out (usually in private), and at that point the lone holdout often becomes two holdouts. And so it grows.
Once an opinion is changed from one side to the other, it doesn’t feel like a flip-flop between two equal and opposing views. Instead it feels like one has transcended from a somewhat naive and childish view to a broader and more maturer one. That person can acknowledge still-valid points from his past, but he sees them now from a larger perspective.
Which explains the commitment to that view. From the “higher” perspective, one can see where one’s opponents are coming from, one can understand their point of view, but they don’t yet understand yours. So you’re patient with them, tolerant even, but committed to your own knowing that sooner or later they’ll transcend those younger worldviews also.

… the percentage of committed opinion holders required to influence a society remains at approximately 10 percent, regardless of how or where that opinion starts and spreads in the society.

How does this differ from the (old school) observation: “Yhe aggressor sets the rules”?
.

Antony Jay and Jonanthan Lynn are doing their bit in influencing opinion about CAGW. Their new play, Yes Prime Minister, at the Apollo Theatre in London, UK is simply hilarious and takes a massive swipe at the global warming fiasco, with a wonderful poke at the machinations of Government to keep it going and be seen to “be doing something”.
If you are within striking distance of London – get yourself a ticket.

Richard S Courtney

Friends:
The research (sic) is merely a computerised formulation of demonstrably untrue assertions.
The report says;
“To reach their conclusion, the scientists developed computer models of various types of social networks.”
But an ability to describe an idea using a computer model is not an indication that the idea is correct.
And, as several have pointed out, this so-called research is demonstrated to be wrong by its own conclusions.
For example this;
““When the number of committed opinion holders is below 10 percent, there is no visible progress in the spread of ideas. It would literally take the amount of time comparable to the age of the universe for this size group to reach the majority,” said SCNARC Director Boleslaw Szymanski, the Claire and Roland Schmitt Distinguished Professor at Rensselaer.”
If that were true then, for example, Christianity, Bhudism and Islam would not exist because they each started with a single individual.
And this,
“Once that number grows above 10 percent, the idea spreads like flame.”
If that were always true then, for example, any political party that has more than 10% support should grow, but several do not and some decline.
Models which provide indications that are denied by empirical data are faulty models that provide no useful predictions. Of course, some of their predictions may agree with reality because chance works that way.
Science consists of attempts to find information which indicates faults in existing models of reality then refining or replacing a model so it better represents reality.
Pseudoscience consists of attempts to find information which concurs with predictions of existing models then assuming the found information supports the model. But some predictions of almost any model may agree with reality because chance works that way.
The report says;
“The researchers are now looking for partners within the social sciences and other fields to compare their computational models to historical examples.”
Such comparisons can only be pseudoscience because the predictions of their model do not concur with observed reality (e.g. movements founded by individuals have changed and do change the world, and political parties with more than 10% support do not always spread “like wildfire”).
In other words, if the “researchers” were scientists then they would be rejecting or amending their models and not “looking for partners within the social sciences and other fields to compare their computational models to historical examples.”
Richard

Fred

So this explains what is happening in Washington! I prefer the old way – you know, the one where people talked and reached consensus.

theBuckWheat

What I didn’t see in the research is where there is more than one group in a population that is > 10% but that holds an unchangeable belief. In any case, this groupthink effect can be good or bad, depending if the majority is being moved from truth or away from it. After all, phrenology used to be a serious subject.

This is nonsense for the reasons amply stated above–if less than 10% can’t get traction and over 10% must get traction, then there will never be crossover above and below 10% and we will forever be locked into one dominant unchanging system.
And what about where more than one group has over 10% (a situation that pretty much always prevails)? They both inevitably win? Does one win quicker and then cede the field? Or do they both achieve total unchallenged dominance at the same time? Whether the first situation or the second, how does it play out in real time?

SABR Matt

I’ll echo the obvious question…the normal mode in American politics (and all over the world I presume) is for one issue with two mutually exclusive resolutions to have 30+% of the population believing each alternative with hopeless commitment.
On economics, when the US National debt becomes a problem, 1/3 of the country believes, and is unswayable in that belief, that the only solution is to tax the rich into oblivion to pay for social justice programs. Another third believe and are unshakable in that belief that the only solution is to cut spending for all those bleeding heart social programs and tighten our belts, plus lower taxes to help business. Neither idea has a super-minority base and neither idea can be reconciled with the other. What does the model say about that?

scp

Seems like it all boils down to their model for persuasion. If a listener talks to two consecutive “true believers”, then their persuaded? How accurate is that scenario? Any evidence given to indicate that it mimics real life?

Hector M.

It is important to note that these are not experimental results but only MODEL results. It is about a computer model designed in a certain way. For instance, people are all alike (no ethnic or social differences among them), the proportions of the various opinions and degrees of belief are set in advance, and (perhaps the most important) a person changes from one opinion to the opposite after encountering one true believer and just another one person both with the opposite opinion.
It is quite probable that slightly modified details in the model would change the results, perhaps dramatically. And also, in all this the “opinion” is completely subjective and freely adopted, such as liking or not liking a pop singer, using or not using a new fashion, and so on. Objective facts apparently do not count: if the opinion is about something objective, would new facts alter the situation? Would the situation be altered if “true believers” or others are or are not able to produce verifiable and replicable evidence in favor of their views?
Experiments can be conducted on these issues, but the report is all about computer models. No reality is implied.

pat

psyops research perhaps. now that only around 10% of people are unshakeable believers in CAGW, perhaps we are to believe they will win the day!
27 July: El Pais Spain: (HEADLINE TRANSLATION: THIS JULY WILL A LESS WARM CENTURY)
http://www.elpais.com/articulo/sociedad/julio/sera/calidos/siglo/elpepisoc/20110727elpepisoc_5/Tes

Kip Hansen

The ‘science’ behind social networking theory has been seriously called into question. Other studies have claimed that obesity and divorce are contagious (like smallpox), based on the same types of erroneous logic displayed in this piece.
Anyone who played around with automaton ‘bugs’ in computer programming back in the 1970’s will recognize the technique they are reporting….often used in graphic studies of population dynamics. The results are entirely dependent on the ‘rules’ set by the programmer, even though the results can be quite surprising sometimes.

JSkelley

Allenj has an excellent point: what if multiple (and opposing) subgroups which have reached unshakable 10%? I suppose the issue rests on the definition of “unshakable.” May I suggest the term ‘cult?’

Madman2001

Too many appearances of “always” and “never” for me to put too much faith (>%10 faith) in their modelling.

steveta_uk

Oh dear, guys, learn to read!
There is nothing contradictory in the fact that any opinion held by under 10% of the population would take approx. the age of the universe to reach the tipping point, and the fact that opinions held by more than 10% of the population spread throughout the population.
It simply means that all opinions currently held by a majority of the population were formed in the big bang.
This is obviously the first evidence for a “conservation of opinion” law.

DonS

So, with just a $16.75m grant from the US Army, which probably posed the question: “How can we get these Iraqis to stop blowing themselves up in our presence?; this group with two laptops, one female, one WASP-like man, two young men and no documentation in sight have achieved this great feat since Oct 2009. Not bad. Imagine what they’d have done if they had also received some stimulus money. My opinion is formed from clicking the link above and observing a photograph. Pass it on.

t stone

The whole thing smells of:
a). Appeal to authority
b). Determinism
c). Group Think
d). Collectivism
e). All of the above
And not to mention the headline: “Minority Rules: Scientists Discover Tipping Point for the Spread of Ideas”. Scientists “discover” a tipping point. Really? Where? Oh…in a computer model.
So let’s announce our “discovery” and then confer with folks who might know something about human interaction to “confirm” the “facts” we discovered with our model.
Sorry, but puhleeeeease.