Slashdot Mirror
Social Influence and the Wisdom of Crowd Effect
formfeed writes "A lot has been written lately on the crowd effect and the wisdom of crowds. But for those of us who are doubtful, the Proceedings of the National Academy of Science has published a study showing how masses can become dumber: social influence. While previous studies show how groups of people can come up with remarkably accurate results, it seems 'even mild social influence can undermine the wisdom of crowd effect in simple estimation tasks.' Social influence 'diminishes the diversity of the crowd without improvements of its collective error.' In short, crowd intelligence only works in cases where the opinion of others is hidden."
duh?
Just look at Facebook.
You got the touch!
Comment removed based on user account deletion
Does not!
We'll see if this works...
While most problems today are complex. We still try to cling to the groups ideology to try to solve the problem vs. realizing the ideology isn't the solution just a start of an approach which needs modifications. However political parties leader will not waver too far off their ideology core as the group in the hole still follows that ideology.
If something is so important that you feel the need to post it on the internet... It probably isn't that important.
In short, crowd intelligence only works in cases where the opinion of others is hidden.
Nonsense. Opinions do *not* need to be hidden, opinions are one source of information. What needs to be suppressed are cliques, groupthink, etc.
In short, crowd intelligence only works in cases where the opinion of others is considered but not blindly followed, where individuals think for themselves.
No wonder so many people like this talentless little sh--head.
Interesting. Maybe democracy would work better if we didn't know the opinions of others, have poll data, or hear media commentary other than candidates speaking and their records...
think for yourself.
The most important take away from this is that polls are a bad idea This study suggests to me that the aggressive reporting on the results of polls related to upcoming elections explains why the quality of our leading politicians has been declining over the last several decades.
The truth is that all men having power ought to be mistrusted. James Madison
I believe that society fails when context is taken to the extreme and not countered or debated. When a one sided argument is presented and no one is allowed to counter that argument, it inevitably causes ignorance.
Examples are all over our society, such as these recent examples. "Creationism" issue, what is "Fox News", and what happened during the Bush Jr era presidency. Jon Stewart had an issue just last week with Fox News presentation of a rapper and Fox News's inability to maintain a "status quo" in their arguments.
So, when it comes to social influence, if one were to put themselves in a bubble, such as Facebook can, and online games allow due to heavy handed administrators, then yes... ignorance can perpetuate itself.
Of course the counter to this is "trolls", those that make obviously inflamatory or ignorant counter arguments that defy belief and logic. But the reality is that some trolls are "for real" in their thinking. Is it really that far fetched to believe that some people in this world have no logical function what-so-ever? The proof is all around you.
So, of course, because of this conundrum that they present, we shield ourselves from others opinions and call them trolls.
Free flow of ideas was the internet, and never will be again.
Herd mentality implies a fear-based reaction to peer pressure which makes individuals act in order to avoid feeling "left behind" from the group.
Qualified as "fear-based" and a "reaction to peer pressure" already implies a negative force. It's always nice to have studies to back it up though.
Twinstiq, game news
The basis for most of what is believed in both realms.
Some of my colleagues recommended that I write a letter about how The Wisdom Of Crowds provides simplistic answers to complex problems. This is that letter. With this letter, I hope to help people break free of The Wisdom Of Crowds's cycle of oppression. But first, I would like to make the following introductory remark: The Wisdom Of Crowds likes thinking thoughts that aren't burdensome and that feel good. That's why it has always promoted the trendiest causes, the causes that all of the important people promote. But you knew that already. So let me add that sometime in the future it will feed information from sources inside the government to organizations with particularly deceitful agendas. Fortunately, that hasn't happened...yet. But it will decidedly happen if we don't improve the lot of humankind.
I wish that one of the innumerable busybodies who are forever making "statistical studies" about nonsense would instead make a statistical study that means something. For example, I'd like to see a statistical study of The Wisdom Of Crowds's capacity to learn the obvious. Also worthwhile would be a statistical study of how many infantile, libidinous rabble-rousers realize that The Wisdom Of Crowds likes to imply that it has mystical powers of divination and prophecy. This is what its quips amount to, although, of course, they're daubed over with the viscid slobber of narrow-minded drivel devised by its worshippers and mindlessly multiplied by unforgiving weasels. Other than that, The Wisdom Of Crowds has a vested interest in maintaining the myths that keep its polity loyal to it. Its principal myth is that power, politics, and privilege should prevail over the rule of law. The truth is that unless you define success using the sort of loosey-goosey standards by which The Wisdom Of Crowds abides you'll realize that true measures of success involve raising the quality of debate on issues surrounding The Wisdom Of Crowds's rummy expostulations. Success is getting the world to see that to someone whose eyes are open, The Wisdom Of Crowds's constantly repeated mantra that its complaints can give us deeper insights into the nature of reality is an insanely foul notion. By way of contrast, consider my personal mantra that if you don't think that The Wisdom Of Crowds exhibits a perverse talent for getting viscerally angry and staying angry long enough to subjugate persons of culture, refinement, and learning to abusive scalawags, then you've missed the whole point of this letter.
It has been proven time and time again that The Wisdom Of Crowds and I are as different as chalk and cheese. It, for instance, wants to prime the pump of sexism. I, on the other hand, want to look at our situation realistically and from a viewpoint that takes in the whole picture. That's why I need to tell you that I am sick of hearing it intone with an authority reminiscent of Moses descending Sinai that its sermons enhance performance standards, productivity, and competitiveness. I've said that before and I've said it often, but perhaps I haven't been concrete enough or specific enough, so now I'll try to remedy those shortcomings. I'll try to be a lot more specific and concrete when I explain that this is a lesson for those with eyes to see. It is a lesson not so much about its obdurate behavior but about the way that the next time it decides to turn a deaf ear to need and suffering, it should think to itself, cui bono?â"who benefits?
Over the years, I've enjoyed a number of genuinely pleasurable (and pleasurably genuine) conversations with a variety of people who understand that The Wisdom Of Crowds defines "truth" as "whatever promotes parasitism". In one such conversation, someone pointed out to me that I realize that some people may have trouble reading this letter. Granted, not everyone knows what "galvanocontractility" means, but it's nevertheless easy to understand that the unalterable law of biology has a corollary that is generally overlooked. Specifically, The Wisdom Of Crowds once said that we should cast our lots wi
after the weapons are collectded & recycled into being useful stuff, there well could be music in the air at all times, if we want that.
...that says something like "Persons are smart individually, but people are dumb." I didn't need a scientific study to reveal that factoid to me.
"In short, crowd intelligence only works in cases where the opinion of others is hidden." Hell yeah, that's why bad comments get moderated up, but moderating down Goatse posts works brilliantly!
I8-D
"Though we adore men individually, we agree that as a group they’re rather stupid.”
Mrs. Banks in Mary Poppins
These comments are my own and do not necessarily reflect the views or opinions of my employer or colleagues...
but more dangerous is a self-selecting subgroup who believes they know better than anyone else, and based on that, feel they have a right to impose their "wisdom" on everyone else. no, i'm not talking about math or hard science, i'm talking soft sciences or ideology. education isn't a protection, as "education" is often just indoctrination into a set of assumptions that cannot be doubted on fear of banishment from the group
so i cast my lot with the wisdom of the mob. i don't trust the mob, but at least its allegiances are simple and easy to discern. in other words, yes, the mob is dumb, but the mob is also honest. so-called experts meanwhile are more often just ideologues with a political agenda to promote
intellectual property law is philosophically incoherent. it is your moral duty to ignore it or sabotage it
Please watch Eli Pariser's talk at TED about Filter Bubbles: http://www.youtube.com/watch?v=Gbi2i_Y7gSE The "wisdom" of the crowds is "managed" via cherry picked search results, etc;
We play the game with the bravery of being out of range
somehow a debate came up about whether cats eyes glow in the dark or are highly reflective, making them appear to glow in the dark in low light situations. The teacher honestly did not know the answer (facepalm) and asked the class to vote on it. It was almost unanimous that they did in fact glow in the dark, aside from me and one other poor soul. I remember that day very vividly. It was the day I realized people are dumb and a general consensus means NOTHING in terms of accuracy.
This article shows that social media news aggregation sites that use crowd voting will constantly rise and then fall. Digg was a cool site that had tech oriented news. It became popular and then the diversity crowded out the reason what made digg cool.(Intelligent sci/tech oriented news) This led to its downfall. Reddit steps into the scene with the same sci/tech oriented aggregation. Masses came and the intelligence of it dropped. Hacker news springs up and has remained less popular and thus more "intelligent".
Their downfalls can also be attributed to other factors like the 'gaming' of the system. This will always happen whenever there is a large enough of an audience and there is a dollar to be made.
Why has /. maintained such high caliber sci/tech news? Editorialization. By users not seeing the dumbass submissions of everyone else; a proper decorum can be held amongst the users.
This was a favorite quote of mine from a fraternity I attended at an engineering school. "We're supposed to be smart kids, but collectively, we're dumb as fuck."
www.awkwardengineer.com
Democratic groups deciding something is more akin to what army do I join not what's correct and will work best. People simply select the group they think they will do best in or the group they think will win. So it's not really surprising that democracy is really bad at getting good results.
The fittest are not necessarily the smartest? The dummer the person, the higher the psychological resilience, the more numerous the offspring.... :D it seems joining the crowd in undermining the wisdom could be the better survival strategy from evolutionary perspective :D...
A person is smart. People are dumb, panicky, dangerous animals.
- K
Nevermore.
Canadians collectively reelected harper - with a majority.
I'm confused, how is this new? I know that recent years have found ways in which crowds can sometimes be intelligent, but every first-year psych course begins by explaining how dumb ther are, and the seven reasons why. This is yet another observation done by a team who never cared to consider research already done and considered common sense.
How boring.
People are dumb (all of us in some regards including myself). A crowd of peoples even dumber as it panders to the lowest common denominator.
Okay where do I apply for a research grant?
sounds like they had a few trolls in their study. there's nothing better than having to debate common sense and logic with someone....
Facebook is just plain crowd: like a mob with everyone shouting.
Slashdot is a crowd moderated by randomly selected crowd-members, with multiple-moderation, meta-moderation, and karma-influence.
That's a huge difference. And you can see it. On Facebook, there's an endless stream of garbage. On slashdot, you can go back to an article that has matured and just read the 4s and 5s and get a pretty good sense of the best content.
No slashdot ain't close to perfect. What this shows however is that "wisdom of the crowds" is variable based on the system used. The more complex and well thought-out the system, the more wise the crowd gets.
...Terry Pratchett told me. In several of the Discworld books, he has stated the the intelligence of a mob equals the intelligence of the stupidest member divided by the number of people in the mob.
Sounds about right.
The most common way is for crowd members to transmit and receive most information from each other, instead of all doing their own research. This tends to amplify distortion (Remember the "telephone" game from your childhood). When intra-crowd communication is minimized, the crowd again becomes accurate.
This principle can be demonstrated by blogs that repeat stories from blogs that repeat stories from blogs.... Copy and paste, oddly enough, serves to minimize some of the distortion effects, however, additional commentary that accretes around the original story with each telling inevitably creates more distortion.
Another distortion is the "Fox News" effect, where erroneous information is repeated endlessly to a large portion of the crowd. Repeated enough times, this method too distorts the information picture by adding single-source bias. FYI, this happens with all MSM media outlets. Fox News is just the most obvious example.
So the wisdom of crowds exists and is useful, but the internet has provided a friction free means by which it can be distorted and become useless, or worse.
Please do not read this sig. Thank you.
This is the age-old problem with democracies: What to do when the majority is wrong? This can easily happen when even the "smartiest" are ill informed, or influenced by advertising, or don't care -- and now it seems that the more people share their thoughts and ideas, the less likely it is that those ideas are optimal. It seems like we are trading critical thinking for a feeling of belonging.
I actually had a go at this a while ago. Orange UK ran a competition where you had to guess the amount of money (single denomination coins) in a tube. People were allowed one guess a day, to be posted on the relevant Twitted feed with a certain hashtag. The correct guess, or the closest to it after a number of days won the contents of the tube. The amount was somewhere in the hundreds of pounds.
I tried two main methods to this, the first was to make an educated guess. Certain clues like the height of the tube (including base), etc were given so it was fairly easy to guess it was a standard 2m acrylic tube. The diameter was guessed based on the high res picture they gave - again, standard diamater found by Google. Packing efficiency I had to do manually with a glass and £30 worth of ten pence coins from the bank which got a ballpark estimate of £460.
Secondly I devised a way to farm people's guesses and then base a more educated guess from what they were saying. I wrote the script in PHP, automated with jQuery, running on a virtual server on my laptop which parsed the RSS feed of that hashtag, extracted each guess and stored them in a database with the timestamp so things weren't collected twice. A quick and dirty page worked out and displayed a number of averages and so on. I had a number of friends who were each given a guess by me to post each day; I offered a small fraction of the bounty, you understand. After I located where I thought the average was, I worked out where to guess (either side of other people) such that if the total was within a certain range, we would definitely win. The strategy was to pick prices spiralling outwards from the chosen average, so that over time the range stayed continuous, but increased steadily. Obviously some guesses were made filling in gaps where other annoying members of the public had chosen!
The results were pretty interesting. People guessed over an extremely wide spread, guesses ranging from the thousands to the hundreds. Each day a new clue was given, including two clues that provided upper and lower bounds on the total. These were the results:
http://img215.imageshack.us/img215/2407/screenshot20110518at220.png, x-axis is # guesses, y-axis is in pounds. Error bars were estimated (not sure why they're on there tbh), the blue line is the simple average and the black line is a moving average of 10 guesses. Removed are guesses that were clearly mistakes, £2221 reposted as £222.1 for instance and duplicate guesses.
Average (all): £480.22
Average (£0-755.40): £406.12
Average (£100-755.40, narrowing averaging range to known limits): £419.18
So, going on that, I'd covered between about £450 and £520. Unfortunately the actual result was £380 - bummer. The problem is, of course, knowing where to take your average. I'm not a statistician, and while I deal with a lot of statistical distribution, error analysis and confidence intervals, this isn't my strong point. I chose to discard stupid entries (in the thousands of pounds) and that predicted the result to within £20; this works to a point, but how far do you go, do you discard incorrect low prices too? Anecdotal evidence says you should go for a total average - hence my bets around £480.
As to crowd wisdom, it was interesting being able to see the guesses where they came in and also how varied they were. The standard dev was huge, so I didn't pay much notice to it - something like £300. Clearly some people were being influenced, and as clues were given the results changed accordingly. TFA seems accurate then, but it was an interesting experiment and I nearly won 400 quid from Orange!
It's called the mob mentality. I read about it in 5th grade. "To Kill a Mockingbird" was an excellent book.
I'm going to see if I can get a grant to study "The Effects of Televised Sports as a Deterrent for Revolution". I just hope no one remembers anything before 1980 and I'll be all set.
*sigh*
"Helping to keep you two steps ahead of the Thought Police!"
Carl Gustav Jung said it long before. You might find his book, The Undiscovered Self interesting.
Could anyone explain the significance of this mysterious graph? http://www.wired.com/images_blogs/wiredscience/2011/05/crowd_estimates1.jpg
It's from the Wired article.
"A person is smart. People are dumb, panicky, dangerous animals and you know it."
C'mon, nothing new about this study, the poster for it came out years and years ago. Move along.
that would be an interesting comparison.
there was a lot less anonymity about 'voting' on kuro5hin, you could tell who voted your articles up and so forth and so on.
The Delphi Method was designed to avoid the "halo effect" and groupthink drawbacks... so this study basically re-affirms that when you want to improve the quality of conclusions from a group -- that the structure of the communication needs some anonymity or other means of removing groupthink.
http://en.wikipedia.org/wiki/Delphi_method
Kaczynski, is that you hiding behind an AC monkier?
Does it go on forever?
this study, and all the replys just reminds me of me of one of favorite movie quotes of all time: "A person is smart. People are dumb, panicky dangerous animals and you know it."
The Nexus Effect: people follow whatever others are doing to be: cool, happy, health, w/e.
Jean Jacques Roussea once famously argued (famously amongst political philosophers at least) that it is necessarily rational to follow the majority decision, because the larger a group you pose a question to, the more likely the majority answer is going to be.
He argued this roughly as follows:
(1) The average person is at least a tiny, tiny bit more likely to be right than to be wrong on any given question.
(2) Any bias in a set of figures will tend to be more pronounced in a larger set [e.g. a coin weighted 51% toward heads will be more likely to show a greater proportion of heads to tails in a series of a million flips than it would in say, two flips, where you might very reasonably expect one heads and one tails].
Therefore:
(3) The larger a crowd you pose a question to, the more likely the majority answer is to be correct.
I am fond of inverting that argument against his position, showing that if we deny the first premise and instead adopt the inverse, that the average person is most likely going to give you an incorrect question to a random question (which I find much more plausible than Rousseau's assumption), then by the same statistical reasoning, the larger a crowd you pose a question to, the more likely the majority answer is to be incorrect.
Voila, statistical proof that people are stupider in crowds than they are on their own.
-Forrest Cameranesi, Geek of all Trades
"I am Sam. Sam I am. I do not like trolls, flames, or spam."
George Carlin said it best: "Think of how stupid the average person is. Now realize that half of the population is dumber than that."
http://en.wikipedia.org/wiki/Stockholm_syndrome
Slashdot = Sarcasm
So when you get large groups together and mislead them they convince themselves and each other to act on bad ideas? Great...culture as a whole, is explained as one long, big, bad idea.
"We live as though the world were as it should be, to show it what it can be." - Joss Whedon via Angel
Free speech is important in society, but even today everybody knows it has its limit: it's either abused, or censored, or moderated for incoherent reasons. And EVEN THOUGH, some countries don't have free speech.
From here try to understand how single individuals can wisely interpret society as a whole: it's just impossible.
You could also try to separate the cognitive patterns of opinions from other thoughts. Good luck with that.
Seriously, the recent advances in sciences and technologies are grains of salt versus what we should need in terms of society advances. Not social, just society.
Each time a security problem is discussed, I post the following comments:
a) the programming languages used to construct affected programs (C and C++ in most cases) are inadequate for the purpose of making secure apps.
b) operating systems could be designed in a better way, even coping with social engineering issues.
c) CPUs could be designed in a better way as well, coping with software modularization issues related to security.
Yet, each time, I meet the following responses:
a) it doesn't matter if the programming language at hand has problems, as long as the programmers are good. Well, the programming language does matter (long discussion, not suitable for this).
b) there is no way operating systems could be designed in a better way (they could, that's another long discussion).
c) CPUs are already optimally designed (they are not).
All people are subject to the mob thinking problem. Even the knowledgeable/informed ones as the slashdot members.
http://en.wikipedia.org/wiki/Delphi_method
"Anonymity of the participants
Their identity is not revealed, even after the completion of the final report. This prevents the authority, personality, or reputation of some participants from dominating others in the process. Arguably, it also frees participants (to some extent) from their personal biases, minimizes the "bandwagon effect" or "halo effect", allows free expression of opinions, encourages open critique, and facilitates admission of errors when revising earlier judgments."
What do you think politicians "consensus making" is about? It's about manufacturing and broadcasting an imaginary reality for the purpose of overcoming the crowds natural diversity and robustness.
"Biased Journalism sells more... magazines" - Adrian Veidt/Ozymandias, from "The Watchmen"
"What needs to be suppressed are cliques, groupthink, etc." - by perpenso (1613749) on Wednesday May 18, @03:38PM (#36169900) Homepage
See my subject-line: That's exactly the problem here on this website's forums - if/when you post material that shows Windows or Microsoft stuff does well, or even BETTER than *NIX variants? You get "down moderated", & often without any technical justifications
No... instead rather, solely based on "zealotry" alone, trying to hide what makes *NIX look poorly vs. Windows?
Instead, You get the "Pro-*NIX trolls" like tomhudson around here modding you down & trolling you via AC posts instead, AND TELLING OTHERS TO JOIN tomhudson in it also! Proof? Ok, I'll let tomhudson speak for himself on that very account:
"Wait until he starts on another kick, then reply to him as an AC. It's the new meme". - by tomhudson (43916) on Sunday May 09 2010, @08:29PM (#32150544) Homepage Journal
QUOTED VERBATIM FROM -> http://slashdot.org/comments.pl?sid=1646272&cid=32150544
AND, "True to AC STALKING TROLL FORM"?
Tomhudson did so again, repeatedly & again, here:
http://slashdot.org/comments.pl?sid=2086424&cid=35841122
and here also:
http://slashdot.org/comments.pl?sid=2086920&cid=35840680
Repeating his misdeeds in trolling & stalking myself via AC replies, and, telling OTHERS TO DO SO WITH HIM!
(Stupid, stupid, stupid...)
Also - please:
Don't even TRY to tell me it doesn't go on, you all KNOW it does!
E.G.-> Hell, even the editors won't post stories here that come from VALID/REPUTABLE SOURCES that do so!
Case in point/example?
Ok, this quote from Eugene Kaspersky where he even states Windows is as secure as Linux or moreso:
http://slashdot.org/submission/1568086/Windows-not-less-secure-than-LinuxOS-X
It was put up as a story for submission here, in the "recent section", but it never was put onto the main page... totally "blown off" & we ALL know why (the /. "Pro-*NIX slant" around here & the trolls that help promote it, knowing most folks are "sheeple" that 'follow the crowd' because they don't know enough about a tech topic to know better!)...
I've even had it to my posts here, & it's happened SO many times on security data I've put up in this regards that it's NOT funny!
E.G. #1 on ANDROID problems (a LINUX variant) ->
http://tech.slashdot.org/comments.pl?sid=2148646&cid=36106332
E.G. #2 on LINUX unpatched security vulnerabilities currently, having 3.5x as many as not only Windows 7 alone, but vs. nearly the ENTIRE suite of MS products for business & development as a platform ->
http://it.slashdot.org/comments.pl?sid=2077414&cid=35776848
(Not even a technical justification was given for the "mod down", just off topic trolling b.s. & the -1 rating, even though I used VALID data from a REPUTABLE SITE for security vulnerabilities remaining unpatched in both Linux latest & Windows 7 + all of MS' latest wares!)
APK
P.S.=> Quoting the article here now:
Social influence 'diminishes the diversity of the crowd without improvements of its collective error.' In short, crowd intelligence only works in cases where the opinion of others is hidd