ฝัง
- เผยแพร่เมื่อ 12 ธ.ค. 2017
- We all know that when we make decisions in groups, they don't always go right -- and sometimes they go very wrong. How can groups make good decisions? With his colleague Dan Ariely, neuroscientist Mariano Sigman has been inquiring into how we interact to reach decisions by performing experiments with live crowds around the world. In this fun, fact-filled explainer, he shares some intriguing results -- as well as some implications for how it might impact our political system. In a time when people seem to be more polarized than ever, Sigman says, better understanding how groups interact and reach conclusions might spark interesting new ways to construct a healthier democracy.
Check out more TED Talks: www.ted.com
The TED Talks channel features the best talks and performances from the TED Conference, where the world's leading thinkers and doers give the talk of their lives in 18 minutes (or less). Look for talks on Technology, Entertainment and Design -- plus science, business, global issues, the arts and more.
Follow TED on Twitter: / tedtalks
Like TED on Facebook: / ted
Subscribe to our channel: / ted
This is quite fascinating! Since I work in a team-driven environment, trying to reach group decisions is something I deal (and sometimes struggle :'D ) with daily.
This is the best use of graphics in a TED talk yet
DeoMachina You haven't seen TED Ed then.
a) A group of Ted talk guests are probably of above average intelligence, self reflective, critical thinking and eloquent, even though they are from different continents, they are most likely from the respective intellectual elite. 2) These experiments tested sponaneous, non-essential compromises. What happens when you add skilled politicians and stakes to your argument?
There's only one way to find out.
Are we bold enough as a species to search the space of possibilities in hopes for advancing our social order, or are we going to let the flawed current systems, which are not really working that well, to keep going forever?
That's the real question. We have plenty of theoretical, experimental and real-life evidence that we MIGHT do much better than what we do right now. We only have fear of change stopping us.
Or actual real questions with real and dire situations. This video was uninformative and only regurgitated advocating for a system for sheeps. Sometime we need outliers and that's what America was about in the first place, people crazy enough to state a fact of life.
Ka0s88: We need outliers in that sense yes. But that's not what they're on about in the video. There's a HUGE conceptual difference.
The system has to take into account the outliers, but that conceptual battle must be on the cultural level, not on the political power level, otherwise you're just defending the potential for tyranny from extremist outliers. You're just waiting for your side to be in charge. Sounds familiar?
The current system promotes extremist outliers above the average consensus, because it's based on majoritarianism. That's why political discourse is polarized and discussion almost meaningless.
This system in the video promotes the consensus over outliers, which moves the discussion to the cultural level, with debate and conversation, as opposed to allowing extremist outliers to steer the conversation, as it happens right now.
Who said anything about extremism? Matters of fact that's why most people throughout history fled their country and travled to America to avoid extremism by a mob. So I agree huge difference.
Battle? Tyranny? What are you on about? LUL! We already live in a system that accepts you for who you are and what is a culture without politics? It's toxic politics and bad governance is the problem. Not politics in general.
I don't think you have a full grasp of how good you have it, that we can sit here and ALREADY have civil discourse. I look out my window and I see no tyranny, I see no battle of cultures. My good friend is muslim and he knows what I advocate for and we joke back and forth but never get extreme about it. That's an acceptance of culture in America. The system you advocate for is here already guy, to a larger degree more than other nations.
Me and the guys in the video are saying it, because you don't seem to understand the difference between outliers and extremists, the origin of polarization, or how this sort of system handles that differently than the systems we have right now all over the world, which you seem to think are just perfect.
You also don't seem to understand American history, and have an extremely romanticized idea of it. I'm not American, don't live in the US and I can see how bad you seem to get it. That's very odd.
But you seem to be in this comment section to pick up a fight and mock others, not really discuss anything, so I'll leave you be. Cheers!
Love this format!
Subtitles out of sync. Please fix.
Yep. I couldn't understand some of the key sentences. But I'm not likely to re-watch it anyway.
This was incredibly interesting and diserves more views.
The same result was observed in 12th grade and I always found it to be a powerful experiment.
FANTASTIC!
This is amazing
This video is 10/10 and I’m 10/10 confident with my choice
are these results going to be published somewhere?
Very nicely done! Thanks for sharing :)
Good info 👍👍👍
Thank you TED!This will help me.
Such experiments are very important, before we build our *Internet Democracy,* on the example of the Swiss council system.
Drinking the propoganda cool-aid much?
Sounds about right!
That was great
More this kind of videos
. 😀
Oh, man, good video!
Thanks
Simple logic and reason and evidence instead of emotion and dogmatic thought
Forget it.
Does age has any effect in group descussions?
5:49 Emotions of course
7:50 the thoughtful debate is the thing of the political parties which the people elect. Ok I agree that this is a little bit of a problem in the us because of only 2 parties.
2 main parties.
"as a society, we have to make decisions in groups"
Why? We should escape from our silly infatuation with democracy and realise that any group decision making will always leave some people out, and instead privatise as much as humanly possible so individuals can make the decisions that suit them instead of having to go with what someone else wants.
One reason is that Earth is not an infinite plane but a finite surface, on which we're having more and more impact every day (resources, climate, wildlife). We desperately need group decisions to counter our own natural individual behavior because it is destructive. Not that we are evil, but we evolved in an environment in which we had little impact so using all the available resources used to be the best choice. Passivity and selfishness were great to survive too, so it's not a surprise that we do so little despite being aware. Therefore we need reasonable collective decisions, and smart ways to make them emerge.
If we want our species and life on earth to survive, that is.
Subtitle timeline is wrong!
You can open Ted.com to watch this video with right timing subtitle. ..
/This is the way we, society, reach a better world.
this has great graphics. Pls do this more lol
I am disgusted so many people would consider restarting the AI, can someone explain that to me?
hawk0485 Most people view AI as a machine and do not have a empathetic connection to it because it is abstract to them.
@@keystothebox Or maybe because they fear a self-aware machine
So when it's a decision in the real world that involves money, time, energy, suffering etc and it's not just theoretical then people's answers would be different I imagine? And it depends on the specific circumstances, and whether the people involved have something to gain or lose? And just because people come to a consensus how do we know if consensus is actually the most useful tool to make a decision on? People think they know what they want and believe until a more advantageous alternative is given
Group decisions requires good personal qualities.... "....consultation is not an easy process. It requires love, kindliness, moral courage and humility." ~ Baha'i Faith
Interesting.
한글 자막 싱크가 맞지 않는군요. 확인해주시길 바랍니다.
I really don't get the proportion of people that think it's okay to make a likely conscious AI suffer. What am I missing?
I wonder if the value is not actually the "debate" but merely the "interaction". I say that because there is an existing line of research into Swarm Intelligence (also called "human swarms") that enables human groups to converge on solutions as real-time feedback loops. Small groups, as swarms, have been shown to significantly outperform crowds, without verbal debate. This TED TALK describes it: th-cam.com/video/Eu-RyZt_Uas/w-d-xo.html
Thanks Wise Jimmy Fallon
자막 싱크가 안 맞습니다. 고쳐주세요! ㅜㅜ
subtitle doesnt sync. plz fix it
robust avg can be good for question like height of tower or number yesterdays …. it is Not at all good for moral questions as it is equal to no decision and there exist right and wrong in morality with dire consequences for society usually.
What is a good decision?
one that maximizes the desired outcome
for who?
That's up to us to decide collectively as well. That's why this sort of system has to be put on a feedback look of self-correction.
But no, we started out by chance with one system, arguably the worst one, and now we're suddenly demanding unknowable knowledge in order to justify replacing it, instead of being bold and trying out alternatives and seeing how they work.
The status quo is above those same concerns, as always. When will we learn?
According to Mariano Sigman and Dan Ariely, a good decision is averaged yet diversified. They apparently discriminate the outliers.
Well, in a group that kind of question will make you an outlier. LUL!
Hey look its US politics. Confident on opposite ends, even if either is incorrect.
Deliberation rarely happens. Many are too strongly wedded to the opinion of their choice, even if it is not supported by the available evidence.
Excellent video, but it could have been more explicit about its real-life implications if it mentioned the importance of the voting system used.
A major problem with the public perception of "democracy" and collective decision making is that we think "majority rule" is the only legitimate form of aggregating opinions. But majority rule is fundamentally against the principles of collective decision making.
Wisdom of the crowds and the experiments this guy (and many others) has shown, *only* work when you take some type of average of opinion (arithmetic mean, NOT median!), and when your opinion can be placed in a continuum so that this average can be computed.
None of our voting and election methods work this way! Usually, we only get to pick one option, and that's it. This is equivalent to being a high-confident extremist every time you vote, and then the most popular extreme wins.
Is it really a surprise that politics is so polarized and oppressive, then? Is it so surprising that partisanship and see-saw politics happens? It just creates a vicious cycle.
This is why alternative ranked voting systems like instant-runoff voting (AKA ranked-choice voting) won't be much better than what we have now. They don't allow for averages, they are just majority rule with extra steps. The entire premise is flawed. The problem isn't just lack of third parties, it's lack of real representativity.
A better system would be something like STAR voting, which satisfies a weaker version of majority rule and intrinsically accounts for a robust average. That system is basically the same as what he's using with the audiences.
Bottom line: a proper collective decision system has to favor the average opinion, not extremist outliers. The REAL majority is "around the average", not "near an extreme". (Notice that this doesn't mean centrism. The average may very well be an extreme position.)
*None* of our social or political orders are based on this. This is why our social order is so unrepresentative and inefficient.
Want to enact real change? Change the voting systems we use to one based on robust averages in any type of collective decision making.
1ucasvb but how would such a system work? If I understood all of it (wich is difficult for me as a non english native) you are saying that the majority is around a certain opinion not exactly "the opinion" right? but isn't "the around" something we can't really work with? I am really wondering right now, hope you can explain it to me. how would the solution to a problem look like if we could vote like the crowd in the talk they made? for example: we got a problem because we got too many old people in our country (hypothetically or however it's spelled) so there is a need for different solutions (or an idea of a potential solution) how would the vote work then? hope you understand all my questions, my english isn't really that advanced for such a complexe topic...
Sure, I'll be glad to explain. Are you trying to understand the details of the voting process, is that it? If so, that's very simple:
Given many options (ideally more than 2), each person gets a ballot where they can independently rate each option a score, say 0 to 5. For instance, options A B C and D, and a ballot could be A=5, B=3, C=4, D=0.
The ballots are collected and the scores are added for every option. The top two options with the best total scores say (A and B) are considered "the best options". These are the two options closest to the "average opinion".
To choose between the two, we look again at the ballots and see which of these is preferred by the most people. (If you scored option A a 5, and option B a 3, you prefer A over B.)
The most preferred of the two best options is chosen as the winner. This is called STAR Voting.
> majority is around a certain opinion not exactly "the opinion"
I'm saying most people are closer to an "average opinion" than to any other opinion, so no real individual opinion (say, of a political candidate) is truly representative of a population as a whole.
So, by imagining an "average person" with that "average opinion" making the decision, instead of any arbitrarily-chosen particular group of individuals (which is what a majority-based system does), we can extract a better representation of the collective opinion.
Voting is then about influencing this "imaginary average person" by giving it your weighted opinion, and fundamentally, you get things to be your way by influencing everyone else in the population by means of political debate, discussion and conversation, because you cannot expect to simply get a majority and impose your will onto others.
The "averaging" process is extremely effective at preventing majorities of abusing minorities. At its worst, when complete polarization (50%/50%) exists, it's apathetic, as it should be.
The typical criticism against this is that "the average person is stupid", or something of the sort. But unless these people are outright advocating for their extermination, instead of their education and raising the level of awareness of society, then they are justifying tyranny just because they think they or will be on the right side of it. These people are treating fellow citizens as enemies, and the current system as a tool to oppress them, not a system of collective decision making.
If you are stuck in a two-party system like the USA, the only way to break out is to vote third party, even when it seems like a wasted vote. The more people who vote third-party and independent each time out, the more likely it is that a credible candidate will step up at some point in the future, knowing that the potential votes are there for him. What is more beneficial - helping a slightly less corrupt person in a broken system today, or slowly working towards a better system in the future?
That, was pretty interesting
I do not think this would work in Turkey :)) Do you know why? -- the group mental mapping to solve the problem who attend to TED talk ss omewhat similar -- they would like to solve the problem.. However the group that can not understand the problem or not will to take ownership of the solving process might not focus the alternative ideas. How about that?
Huh, I was a confident grey...I'm special!!!! Rofl fascinating, all jokes aside
The confident greys are the conformists. They already average others' opinions in their head before making their own decision.
Also, it's worth saying that average opinion is not the same as centrism.
Centrists are not the same as conformists.
English subtitle is not synchronized with the voice
ET = 324 metres (1,063 ft) tall, about the same height as an 81-storey building.
1:42 these two averages things are the same! what did you mean instead?
How can the 2 averages be the same? The values(figures) are different in both cases.
@@ishanrawat158 probably what i meant is that the average of the averages is exactly the same as the average of all the individual values. he should have raised a point that this is NOT the case in these kind of human interactions, probably due to individual people changing their minds due to group influence
in fact, there was a video where ppl would accept a clearly wrong decision if all or most of the group accepted it (like saying that something is a color when it clearly isnt)
Take care
each group has a leader, so i guess that the group's decision is influenced by the leader
Thats great
Will small groups still end up voting for charismatic populist demagogues? If so many of the problems of democracy will still persist. Voters seldom vote in referendums on policy matters - far more often they vote for people to make those policy decisions for them.
Evidence says probably not. Peer pressure plays a huge role in that process, and the small group of strangers reduces that effect.
A proper voting and election system based on robust averages is also very unlikely to elect extremist demagogues due to the fluidity of candidates and the self-correcting nature of averages.
Western democracy has its foundation on Liberalism, which believes in the autonomy of individuals. Voting in small groups is against Liberalism (not that it's wrong). But don't forget most extremists work in groups (KKK, Al-Qaeda, you name it). Teenagers get rowdy when they gather in groups because of peer pressure.
@@mhtinla I believe that small groups wouldn't have that issue because there wouldn't be more than like 5 people. Bigger groups? Absolutely, there's no way to prevent the group mentality that takes over and takes everyone's personal responsibility away.
Using average in statistics is sooo wrong... Should use median instead.
For example in the case of Eiffel Tower height (at 6:39), median would be 300m, which (in reality) is more accurate than this "robust average".
The next step to this experiment is the messages made here. To study what you guys have to say out side of a real face to face group so do em proud sheep's lol
The Eiffel Tower is 324 meters high. yw^^
Al Gore is in the audience 3:17
Come to me please 😍
This is applying statistics to guessing. Participants are reduced to sample points. On moral problems everyone is qualified. Try multidisciplinary team and real engineering problems where people actually have investment in the outcome . It does not work and has no right to work at least I have seen it failing badly over and over again. It is doomed due to lack of competence / failure of management to recognise skills and gaps of each team members What do you think will happen if you put 1 software 3 mechies and get them to vote on software architecture? The only competent person becomes an outlier!
내 생각을 다른사람을 의식하지 않고 내 의견을 낼수 있는 사람이 되도록 노력해야 겠다
There is specific KNOWLEDGE and accepted DOGMA !!!
The US uses the small-group theory to some degree in the way the Electoral college functions. Each 'small group" is one state.
You don't really get the small group dynamic where people with strong non-extreme views act as mediators, but it seems superior to direct election and at the very least better represents regional concerns when it comes to electing a US president.
I'm not sure if breaking into small discussion groups is even practical on a large scale, but this is at the "proof of concept" stage. There's no proposal to roll this out and run a country with it.
Can someone politely tell me why it'd be wrong to give our kids a set of genes to make him/her better looking? Doesn't that give them better oportunities and better odds to reproduce?
Confident Grey propoganda.
Exactly
The guy is wearing grey... And he's pretty confident. Coincident?
Too many cooks spoil the broth! Age-old advice.
I don’t give a damn weather people are future oriented let me watch my damn Ted Talks
"Using science to help us understand ..." Aaaaaannd, there go the christian fundies.
I think science will make us go to ways powerful people wants. I think that you shouldn't continue with this: none you no one else. It's okay to know that works. Now I tell you: it's not okay to know why neither how it works.
can not understand what you mean
I mean that science has always been used to control people an it will be the easiest it'd ever been if you search the reasons such as why and how. It's a fact.
Gregory Schift
Science often emphasizes the ruling opinion but occasionally enables changes, we know now, smoking is unhealthy and how to operate against tyranny, science will continue to help develop (at a rate proportional to its freedom).
Pretty sure burning tens of thousands of political opponents and social outsiders at the stake ostensibly for religious reasons was *objectively worse* than the subprime mortgage crisis, MARIANO.
democracy 2.0
Oligarchy 1.0
People aren't predictive
Not individually. But on average, people are predictable.
You can't compare non-binary questions whose answers can be averaged (like the height of Eiffel Tower) with binary ones whose cannot (like voting for Clinton vs Trump).
"Either the planet explodes today or it doesn't. It's a 50% chance."
Not that simple. The opinions and decision processes that lead to a person choosing Hillary or Trump are not binary. This is about the decision process itself.
It doesn't really matter whether the outcome is discrete or continuous. To get to the consensous a group does internal weighting anyway so it won't just be a majority vote and even if it was like that, a group discussion will change some opinions thus it will be a slightly better vote as if just every republican votes for Trump, while every democrat votes for Hillary.
1ucasvb Nope. After today, it's again binary -- either the planet exploded yesterday or it didn't, regardless of your thought process.
Look, it really makes no sense that you are insisting on this. You are basically saying "all paths that lead from A to B are EXACTLY THE SAME", or "all functions f(x) obeying f(1) = 0 are exactly the same". It's complete nonsense.
The point isn't what goes in and what goes out. You're just looking at those and thinking everything is the same. It's about how the method processes what goes in (the voters) and how it produces an output (the decision).
You cannot consider ONE SINGLE (and pathologically polarized) SCENARIO like "Hillary vs Trump", let alone one with only two options, to understand this concept being presented here in the video, or what I'm talking about with voting systems.
You must consider all possible elections/decision events based on many different number of different types of candidates/options and voters in order to judge such systems. And you must do it with at least a bit of formality.
In social choice theory, we have ample experimental, theoretical and real-life evidence the current system (majority rule with plurality voting) is the worst voting system around, and the least representative. And we have ample evidence that we can do much better to maximize "quality of outcome" in many different measures.
But you don't seem to be very interested in understanding this, so I'm not sure why I even bother trying.
Certain decisions lead ONLY to binary outcome. In your own example, either the planet exploded or it didn't. You can have 1000 smart scientists to calculate the possibility, but in the end they all die or they all live.
OMG who makes up these groups the ****ing data miners who have a fixed outcome in mind and can manipulate group makeup?
If you guys like this video, you'll probably enjoy the videos I post!!! (Interviews with Bill Gates, Steve Jobs and other famous people!!!)
A “robust average” of people in 1700 would have happily supported slavery as morally acceptable. But wasn’t it as immoral then as it is now? So, uhm, the moral questions at the core of humanity are not always the stuff of “Family Fued” logic (‘survey says ...’)
Hereisrich, are you sure about that? Even in 1700 there were a lot of people who thought slavery was wrong. Even if we got it wrong in 1700, I bet a robust average would have gotten it right sooner than 1865.