Some Collective Intelligence Can’t Be Reduced to Thumbs-Ups and Star Ratings

In a podcast this week, I learned about Sermo, a private social network where doctors can share knowledge with each other. Seems like a good idea — let doctors submit and “rate” treatments for various diseases, Web 2.0-style*, like Digg or YouTube. (This is for fellow doctors only, not like WebMD.)

I suggested Sermo to my father (an ophthalmologist), but he was skeptical. He said each patient is different and many situations are unique. Sometimes patient comfort or reducing risk are more important than treating the disease. Sometimes “subjective” elements like fear or hearsay affect which treatments a patient will accept. How can these complexities be reduced to a simple, Web 2.0 “vote”?

Overview of Sermo
Overview of Sermo

That’s not to say doctors aren’t taking advantage of the Internet. My father subscribes to the American Glaucoma Society’s emailing list and has found it helpful. He said glaucoma specialists from around the world share stories and experiences. When a doctor tells a story he can share more detail, and the listening doctors can interpret and apply the story to their own patients. The collective intelligence in these stories can’t be reduced to a simple thumbs up/thumbs down vote or a star rating. (That’s not to say that this is Sermo’s model — I don’t know — or that my father won’t still try it.)

The book Made to Stick explains the importance of story-telling for transmitting information. A Xerox repairmen told his co-workers, over a game of cribbage and in precise detail, how he and his partner spent 4 hours repairing a photocopier that gave them a misleading “E053” error message. Here’s why:

Why do people talk shop? Part of the reason is simply Humanity 101–we want to talk to other people about the things that we have in common. Xerox repairmen work with photocopiers, so they talk about them. But that’s not the only factor at play here. For example, the storyteller above could have shared the general arc of the story without the details. “I had a real bear of a problem today–it took me four hours to get to the bottom of it. I’m glad that one’s over.” Or he could have leapt straight to the punch line: “After hours of hassle, I traced the problem back to a measly burned-out dicorotron. How was your morning?”

Instead, he tells a story that’s much more interesting to his lunch partners. It has built-in drama–a misleading code leads two men on a wild goose chase until they uncover, through lots of work and thought, that the problem is simpler than they initially thought. Why is this story format more interesting? Because it allows his lunch partners to play along. He’s giving them enough information so that they can mentally test out how they would have handled the situation. The people in the room who weren’t aware of the misleading E053 code have now had their “E053 schema” fixed. Before, there was only one way to respond to an E053 code. Now, repairmen know how to be aware of the “misleading E053” scenario.

In other words, this story is part entertainment and part instruction. Shop talk conveys important clues about how to respond to the world. It teaches nurses not to have blind faith in heart monitors. It teaches copy repairmen to beware of the misleading E053 code. (Made to Stick by Chip and Dan Heath. pp. 207-208, my emphasis.)

* I’m using Web 2.0 in the classical sense, the way Tim O’Reilly defined it to mean the aggregation of collective intelligence, not the popular connotation of brightly colored websites with rounded corners.

3 thoughts on “Some Collective Intelligence Can’t Be Reduced to Thumbs-Ups and Star Ratings”

  1. I think this example just points out that Collective Intelligence systems need to be expanded to handle both additional levels of acceptance as well as providing some “fuzzy logic” results. At our current level of machine intelligence, a good decision support system can provide perhaps 80-90% of the rationale, depending on the problem, with still some human leeway for the final decision. Anyway this system seems to be breaking some interesting ground.
  2. I strongly suggest that your father give Sermo a try. It’s a very unique experience to have almost instant access to thousands of colleagues in the country. In fact, about a year ago, I suggested to the Sermo physician community that we should write an open letter to the American public outlining the deficits of the current healthcare system. We wrote the letter with the input of over 1,000 physicians on Sermo and now we have over 12,000 signatures on it. Our efforts have attracted the eyes and ears of the new administration and can actually help pave the way for a new wave of healthcare reform.

    The open letter is a great example of tapping into the collective intelligence of a large group, web 2.0 style. You can read it here:

Comments are closed.