Sunday, March 27, 2011

To what extent can groups of people self-organize?

Continuing on with the themes from last week, I continue to think about how people behave in groups, because what else is social media management but the attempt to get them to behave well, and toward some mutual benefit?

I think it is probably true that in pure anarchy, human behavior tends to the lowest denominator.  But what is the minimum required to get people to behave well?

I saw an experiment conducted at a SXSW keynote address that gave me some ideas.  The talk was by Seth Priebatsch, an internet super-genius with a gaming background (there's a video version of the same presentation as a TED talk).  The key factor of his genius, to me, was an understanding of human motivation, and a strategic ability to set up situations and reward structures to elicit the behavior you want.

He had the audience play a group game, toward the end of his presentation.  I didn't participate directly because I was watching a simulcast in another ballroom, but I keep thinking back to it and what it demonstrated.  In the ballroom where he was speaking, he had distributed paper cards in four different colors among all the seats.  For this game, the rules were that no one could leave their seat and there would be no talking.  The objective of the game was for each row of seats to get all the same color of cards.  The game started with them scrambled, so to get each row with only blue, or only green, or etc. would require some trading.  When a row was finished, they were to hold up their cards.  He gave the group three minutes to complete the game.

But they needed far less than three minutes.  The first group held up their cards after about 30 seconds, and row by row they followed quickly after.  We weren't playing, but we could watch footage of what was happening.  Rows decided really quickly what color they would be.  I didn't directly see but I imagine the people sitting in a row all looked side to side to each other, and one person kind of emerged as a natural leader, held up a card and waggled it, and everyone else nodded their heads.  You've been in team situations yourself where this happens, I'm sure.  Someone just steps up and makes a proposal, and if everyone else follows you quickly become a team with a common purpose.

What happened next I did see - row members who had the card of the right color would hold it up and waggle it to people around in other rows.  "I need blue", this meant.  Blue was sent that way.  People couldn't walk around (some did, but they weren't supposed to), so the blue cards had to make their way to the rows where they were needed through a series of small, local interactions.  As he pointed out in the wrap-up after the game concluded, since people couldn't move around, all the interactions were restricted to about eight people sitting right around each person, in a room with something like 2000 people in it.

The cards got sorted in an amazingly short amount of time, I think it was just over two minutes.  Seth himself looked amazed - he said he'd never tried the game before, and had tested it only with about 25 people.  But it demonstrated the point he was trying to make - he could have given elaborate instructions from the stage, saying, "Okay, Row 1, you're blue, and Row 2, you're going to be orange," and so on.  But by not doing that, by letting the room organize themselves, the whole thing got done many, many times quicker.

So what makes me continue to think of this experiment is the question, what does it take for groups to cooperate and self-organize like this, to act in harmony and also to achieve this kind of efficiency?  They needed an overall group objective, which he supplied.  They needed some rules - some of the rules were constraints to make the game harder and show how cool it all was, but some were guidelines that helped constrain and guide group behavior. 

Did they need a ruler?  That's the question I'm left with.  They had one in Seth, because he laid out the rules of the game itself and gave them their common objective.  They didn't have any preordained within the rows.  But probably some leaders naturally emerged, as in picking colors that their rows would be.  They didn't have anyone pre-mapping out the process how they were to get to the goal, they were free to improvise and sort of figure out the trading process as they went along.  Since it was a big crowd of people who had all probably had many years of schooling and all naturally fell into the pattern of obedience from an educator figure speaking from the front of the room, they didn't need enforcers to bring people who were breaking the rules back into line, or a system of punishments or banishment.  So is this all you need?  Or were the experiment and the crowd it was done in so artificial that you can't draw general conclusions?  Did the crowd bring in with them all the obedience and cooperation strategies and social hierarchies and etc. that made it work?

I'm almost worried that I'm thinking about these questions, because I know I'm right away going to get into some arguments with Political Science majors and macro-economists and other people I know, and my ignorance of the literature and major schools of thought in this area are going to be a problem very quickly.  But I'm thinking that social media management is management of big groups of people just like people have been trying to do throughout history, so it's worth it to see what smart people have had to say about it before.

Many more examples to consider, as well - Egypt, Madison, Japan, the Triangle Shirtwaist Factory in 1911.  Stay tuned, I'm sure I will have more to say soon.

1 comment:

  1. I think the structure of the game also has something to do with it - I have memories of another "self organize" game that went horribly, horribly wrong (memorialized in Would be fun to figure out what properties of a game (or society!) *do* encourage parallel self-organizing solutions...