The Wisdom Of Crowds
Most of us probably remember the television program Who Wants To Be A Millionaire—originally hosted in the United States by Regis Philbin—in which contestants were given three “lifelines” they could use to get help answering questions to earn money. What I found most striking about the show was that whenever a contestant used the “ask the audience” lifeline, the choice that the majority of the audience favored was usually the correct one. The questions didn’t require advanced problem-solving skills, just knowledge of facts, so perhaps I shouldn’t have been so impressed that getting a large enough crowd to weigh in would so consistently yield the right answer. But it got me thinking about the power inherent in groups to come to answers that often elude individuals, as well as the value individuals can gain from paying attention to a group’s viewpoint.
One interesting study suggests that when you group three or more people together to solve a complex problem that they’ll perform better than the best of an equivalent number of people working independently. (Interestingly, groups of two did no better at solving complex problems than individuals working alone.) The results also showed that each group performed better than the best person in the group would have performed alone. Group members were found to combine the best of their resources and abilities to achieve this result.
OTHER TYPES OF PROBLEMS
While the conclusions we can draw from this study are limited to a specific kind of problem, i.e., intellectual ones, it does open the door to the possibility that other types of problems might be more easily or quickly solved by groups as well. Certainly, we can see all around us the anecdotal evidence for the synergistic power inherent in people working together: few things in modern society would have come into existence without the cooperation of groups of people with different areas of expertise (not to mention all the standing-on-the-shoulders-of-others that’s brought us most of the modern conveniences we enjoy, such as antibiotics and computers).
I’ve been wondering, though, if groups don’t tend to arrive at the truth more accurately than individuals in general. Which is to say that if provided enough exposure to a person, place, or thing—enough data on which to form an opinion—I wonder if a group will be more likely to reach a collective conclusion that’s more accurate, encompassing, and consistently superior than the conclusion of any one individual even outside the realm of intellectual problem-solving. Though many observations fail to support this hypothesis at first glance, a closer look in each case might suggest otherwise. Naysayers could argue, for example, that large groups of people still make commercial successes out of dreadful movies and books. But, then again, mostly only great art survives the passage of time (reflective of an even larger group’s judgment). Naysayers might also point out how large groups of people have concluded in the past that certain types of other people (Jews, Gypsies, Tutsis) aren’t fit to live. But, then again, in response to the actions of such groups, even larger groups have disagreed.
I think it remains a tenable hypothesis for which several explanations might exist. One, put forth by the authors of the study, says essentially that a person’s creativity and therefore problem solving ability is enhanced when he or she is able to dialogue with others (one person’s good idea is often triggered by another’s, and wrong paths are more quickly identified and abandoned when individuals work together in groups). Another may be that if you put enough people together, you increase the odds of achieving what I call a cancellation of biases. That is, the biases that prevent people from being able to perceive or arrive at a particular truth will naturally vary from person to person (though many people dislike certain foods, for example, not everyone dislikes the same foods). If these biases vary enough within a group so that no one particular bias adheres to the majority, the effect of each bias becomes diluted in proportion to the size of the group (which may partly explain some of history’s worst genocides: too many people of like mind aggregated together without any mediating influences on their misguided thinking). Perhaps this is another reason why, in the study above, average problem solvers were able to make the best problem solvers even better at solving problems in a group than they were by themselves. Perhaps the average problem solvers, in addition to providing a helpful sounding board for the best, helped the best to free themselves from the influence of their negative biases, at least temporarily (this also strikes me as an eminently testable hypothesis).
We could, however, point out other important observations that fail to support the theory that crowds are wiser collectively than the individuals who make them up: populations have frequently elected the worse of two candidates for president. Also, if I offer a thorough work up of a patient with tingling in his feet to a hundred professional football players the likelihood that they’ll arrive at the correct diagnosis is pretty low. Then again, if I present it to a hundred doctors—or better yet, a hundred neurologists—that likelihood flips pretty dramatically.
So perhaps we need an obvious corollary to our theory: some groups are wiser than others about certain things. So if we want to get the best answers, our first task is to pre-select the group best suited to approach our particular question or problem. (We could argue, for example, that the general American public isn’t the “best” group to elect a president—and, in fact, our founding fathers struggled with that very issue—but that question lies beyond the scope of this post.)
Unfortunately, even with the power of social networks now at our disposal, the resources involved in getting large groups of experts together to attack certain questions or problems would seem beyond the capability of most of us most of the time. How, for example, would a patient assemble even ten doctors in a room to discuss his challenging case? (It turns out in direct primary care at ImagineMD, we’ve created a system that does just that). On the other hand, even if we can’t get the best group to talk to one another to help us with our question or problem, we can still derive benefit from talking to a large group of people one person at a time.
THE MOST IMPORTANT ARENA
The best use of this approach might relate, paradoxically, to questions we have about ourselves. After all, even the most introspective of us remains only one person examining the insides of one person. And examining a person from the inside isn’t always the best perspective from which to do it, given that we all have egos and biases. Our friends and family can often see important things about us to which we’re completely blind—and about which they exhibit a remarkable degree of clarity and agreement.
I would argue for this reason that when trying to make a difficult personal decision the wisdom of crowds can be exceptionally useful. Many of us already try to use it to a limited extent when we ask opinions of our closest friends and family members. Should we take this job? Should we move to that city? Should we marry this person? Should we have another child? Though most people might argue there are no right answers to many of these questions—only answers that are right for us as individuals and therefore answers that only we as individuals can find for ourselves—in asking these questions of others what we really have to gain is insight into our own thought processes and biases of which we’re not aware. Thought processes and biases that often affect our decision making adversely but over which we have no control because we simply aren’t aware of them. It would be quite a useful thing for us to know, for example, that many of our decisions are powerfully influenced by our desire to be liked.
If we listen carefully to the answers we hear from well-chosen crowds, we’ll often find important themes emerging, messages that often hold the key to our personal development. Of course, our close friends and family may not feel comfortable providing us insight into ourselves (or may feel too comfortable, their own biases making them unreliable observers). But if we’ve been careful to ask individuals whose judgment we trust, we must summon up the courage to help them give us their honest opinions. Because if they tell us the truth and we have the courage to listen without defensiveness—in essence, to leverage the wisdom of crowds—we’ll likely find ourselves with answers we need to hear.
[jetpack_subscription_form title=” subscribe_text=’Sign up to get notified when a new blog post has been published.’ subscribe_button=’Sign Me Up’ show_subscribers_total=’0′]
“If we listen carefully to the answers we hear from well-chosen crowds, we’ll often find important themes emerging, messages that often hold the key to our personal development. Of course, our close friends and family may not feel comfortable providing us insight into ourselves (or may feel too comfortable, their own biases making them unreliable observers). But if we’ve been careful to ask individuals whose judgment we trust, we must summon up the courage to help them give us their honest opinions.”
This is a most interesting post. And extremely important—although it needs to be hedged round with caveats (a bunch of undiversified bigots is worse than an individual bigot, etc). But I thought it also rang a bell with me about a Quaker process—which I’ve Googled, and come up with this lengthy exposition by Parker Palmer (a first rate mind and person, if you don’t know him):
https://www.couragerenewal.org/parker/writings/clearness-committee
Palmer is discussing the Quaker process of a “Clearness Committee” to aid an individual to make his or her own decisions about career, etc. Since the Quakers evidently came up with the process in the 1660’s, it’s been time-tested. And I think the explanation by Palmer here is worth the time also! I was very much impressed some years ago with his book Let Your Life Speak, and found things in it of enormous importance in my own life. I never had the nerve to try a somewhat formal approach to asking advice even if I had been a Quaker, which I am not; but I suspect this is really on to something. I’m sure you’ll be getting lots of comments on this post!
“We could, however, point out other important observations that fail to support the theory that crowds are wiser collectively than the individuals who make them up: populations have frequently elected the worse of two candidates for president (e.g., Nixon).”
I am not sure if you are talking about Humphrey or McGovern but as both lost their respective elections, we never got to see how they would have performed as President. Perhaps a Humphrey or McGovern Presidency could have resulted in nuclear war with the USSR (unlikely but we don't really know). It is therefore nonsensical to say that Nixon was the worse choice.
BTW, I was too young to vote in either of these elections.
There’s so much food for thought in this post, I’ve printed it out to read again in a few days. My tentative conclusion is that “group thought” has the potential for more wisdom to emerge but also the potential for “mob mentality” to emerge.
Personally, I have learned that when faced with a tough decision, despite my tendency to go it alone or to ask for input only from from someone I feel most comfortable with (often meaning someone who is like me), my decision is more likely to be a good one if I do neither and instead seek the advise of someone who is knowledgeable about or has personal experience with the issue at hand.
[…] This post was mentioned on Twitter by P. Mimi Poinsett MD, Alex Lickerman. Alex Lickerman said: Why more than two heads is better than one: https://bit.ly/dID63M […]
An interesting post, and I agree that what you describe can happen. I’ve experienced it myself many times; in particular I’ve learned that people less experienced or skilled than I am can be uniquely helpful to me in solving complex problems in my particular areas of expertise, in ways that other experts could not.
However, the opposite negative phenomenon of groupthink is well known and well studied. In many places you imply the absence of the conditions (directive leadership, lack of diversity, isolation from outside information) associated with groupthink, but I think your argument here would be more credible if you addressed the “groupthink” issue explicitly.
This is a hard concept for me.
Perhaps you are talking about the phenomenon where the whole is greater than the sum of its parts . . .
Perhaps, if what you say is true mathematically, it is “proof” that we are, as a species, social. And one of the demands of being social is that we submit to what a small or large social circle says is true and actionable.
I’m assuming you’ve read the book Wisdom of Crowds, since that is the name of your post. It’s very insightful. As I recall the wisdom of crowds only works when people are asked what they really think, not what they think other people want to hear, or what they think other people might say. That’s a problem when asking people to opine on our personal problems.
I agree with Toni that there is a lot in this post worth thinking about. While I don’t disagree with what you say about the wisdom of crowds there are definitely some instances where are group is not preferable.
From your examples and my thought it seems like in creative endeavors a group will do better than an individual. Intellectual problem solving is certainly one type of creative endeavor. However groups are very bad at accepting consequences or responsibility. One of the first things I have heard in every sort of first aid training or in training on how to deal with accidents is to assign people to do things. If you shout to a crowd of people to call an ambulance, as likely as not, no one will do so assuming everyone else will. However if you tell a specific person to do so then of course they will. I’m not sure where the wisdom of groups ends and the stupidity of mobs begins.
This is interesting and I have seen good things happen when a group of people “put their heads together.”
On the other hand, I wonder where the concept of “—by Committee” (fill in the blank) comes from? It implies that decisions or other end results of group efforts are not very good. Why is that such a widespread belief?
[…] The Wisdom Of Crowds […]
Just a quick insight. What you describe above reminds me of the interaction between the Doctors on the TV show “House.” I have always enjoyed that show in part due to the collaboration portrayed between Dr. House and his team.
[…] who know the person in whose character you’re interested. As I wrote in a previous post, The Wisdom Of Crowds, if multiple people independently return similar answers, the likelihood that their collective […]