Sunday, January 10, 2010

Sway

One of my goals for 2010 is to read more non-fiction, and the first book was Sway: The Irresistible Pull of Irrational Behavior. Economics normally assumes that people are rationally self-interested; this book is a series of anecdotes showing that they're not.
In one experiment, the experimenters put a subject in a room with other people, showed them all three lines, and asked them whether line A, B or C was the longest. If all the other people in the room pick A (having been secretly instructed to do so), the subject will probably also pick A even if the right answer is obviously C. However, as long as at least one other person picks another answer--even if that person is blind and picks B--that dramatically improves the chance that the subject will be willing to go against the group and pick the right answer. There were several real-life examples from boardrooms and cockpits showing the same thing--in order to avoid groupthink blinding people, you need to have someone who's willing to challenge the consensus. (The authors also uncritically cite France arguing against the US in the debate about whether to invade Iraq; from my point of view, that shows that when you pick a devil's advocate, you should choose who is willing to argue with you but has your best interest in mind, instead of someone who's been bribed by your enemy).
In another experiment, the experimenter auctions a $20 bill, with one dollar bids and the condition that the top two bidders both pay their bid amount but only the high bidder gets the $20. Typically people get it the bidding up to around $16 and then realize "if I'm second, I lose my $16 and I get nothing, so I need to keep bidding"; a few minutes later the bid is at $20 and they're thinking "My bid was $19, if I quit now I lose that; whereas if I bid $21, I end up only losing a dollar", so they keep bidding it up--with a record bid of $204. Again, the authors present several real life examples showing how people will keep trying to avoid a loss and end up losing a fortune--the term is "pouring good money after bad" or "chasing a loss". You need to emotionally detach yourself from the past so you can see the situation you're in right now, and therefore see whether you should cut your losses.
As a third experiment, instructors were told that some trainees (chosen randomly by the experimenters) were highly rated and had good leadership potential. At the end of the course, those trainees had significantly higher scores even though there was no difference between them and the other trainees. In another example, an drink improved people's performance more when it was high-priced than it did when it was cheap or free. People tend to live up to expectations.
I have to say that I don't think the book was well organized or particularly deep. I'd like to have seen more discussion about other types of cognitive biases, and more discussion of "this is what you can do to avoid them". However, the book was an entertaining light read and it's always good to get a reminder to be rational. I'd recommend picking this up at the library.

6 comments:

MarmaladePam said...

Do you think you're more the type that will go against the group (and why would you, as in example A?) or are you a go-with-the-flow person (assuming you'd even bother to enter the room)? Don't be afraid to answer the second one if it's true... nothing wrong with wanting to get along or not be contrary, I suppose.

Laserlight said...

Cryptonomicon says something to the effect that a nerd is someone who cares more about each statement in the conversation being literally accurate, than he does about social graces. That's me. I've always been willing to argue with my teachers / professors /sales managers.

Lux Mentis said...

I've been out on a limb on so many things, group consensus isn't that important to me. Sometimes that gets you in socially oriented difficulties, but the truth is the truth, even if it is inconvenient.

"There are THREE lights." - J. L. Picard

Intimidating/Abusive Employer is a different scenario somewhat...

I'll bet if they did scenario 1 where you didn't know what the others had picked, results would differ. I assume in order to fear breaking the consensus, you must know what it is.

Scenario 3 seems very true to me. I have found in my life that 99% of the time, if I expect good from people, I get it, even from people whom others may write off or have negative opinions of. I've also found that if you expect negative, you tend to get that as well (seen too many friends who have negative outlooks on others then wonder why they get such nasty situations arising.... you get back what you put out).

Note Scenario 1 is one of the reasons a ship's XO (or military units XO) is supposed to play professional devil's advocate. It helps to have someone of sufficient rank always questioning a leader. It will encourage others to think and act sensibly.

It's only in cases where everyone is on the same page you get some silly scenarios (I could suspect the decision to go into Iraq by the Bush brain trust *may* have fit this category, but I can't prove it).

MarmaladePam said...

Haha, I totally KNOW that Picard reference.

Laserlight said...

"Intimidating/Abusive Employer is a different scenario somewhat..."

We had a new sales manager at my job who called us in to work on a Saturday, gathered us around, and had each guy announce how many sales he expected to make. When my turn came, I said "Zero", which was a reasonable number--I don't think anybody actually sold anything that morning--but it was not what he expected. After he got over the shock, he put down "one" for me and I said "If you don't want a truthful answer, you shouldn't have asked me the question." Shortly after that, I moved to another department which involved less "motivational thinking" and more "knowing what you're talking about".

Lux Mentis said...

I had a VP who thought 'management by intimidation' was somehow conducive to effective results.

The reality is that when the managers between him and the employees kept him furthest away, the better the overall team's results were.