Friday, November 07, 2008

Glasses for poor eyesight; sociometers for poor social sense

I recently finished Honest Signals, a book about social signals between humans across cultures and how it is the best predictor of outcomes of social situations from sales pitches to dating. Written by MIT Media Lab professor Sandy Pentland, this book discusses experiments that quantify what psychologists have thought for a while about the role of things like influence (how much effect one person has on another person's speech patterns and behavior) and mimicry (mirroring another person's behavior) in determining social outcomes. The authors have developed a sociometer, a device that measures things like change in vocal pitch, amount of gesturing, etc. This book provides "quantitative" evidence behind what many other psychology/business books say about the role of subjective judgment in important personal transactions. Pentland reports that data from the sociometer to predict which sales pitch will get chosen, which two people will exchange numbers during speed dating, etc. Pentland claims that the content of a presentation or proposal has to do with objective content, and sometimes it is the amount energy and confidence that a startup has (and not the content) that will determine whether it will succeed. It may be for this reason that a venture capitalist will almost always schedule a face-to-face meeting before agreeing to fund someone. Pentland also makes the interesting suggestion that these social, non-objective interactions are more productive than interactions that have tried to remove influence from social signalling. He gives the example that United Nations proceedings, which have speeches go through a translator, tend to be not as influenced by social signalling but are also infamously unproductive. Pentland suggests that the sociometer could be adapted to be used as an aid for people during negotiations, presentations, and other such things to enhance social sense.

Since I have been indoctrinated with this stuff already, what Pentland says it not particularly controversial or novel, but I found it to be quite interesting. I do, however, which the book discussed more methodology--it seemed to go from "we could used the sociometer to make predictions" to "this was our r-squared coefficient." I would have liked for there to be more discussion of the qualities that the sociometer measured, the theory behind why things work this way, etc.

* The most notable one I know is Influence (Robert Cialdini), which provides interesting insight into subtle psychological manipulation tactics that play on people's instincts.

1 comment:

Anonymous said...

I recently came accross your blog and have been reading along. I thought I would leave my first comment. I don't know what to say except that I have enjoyed reading. Nice blog. I will keep visiting this blog very often.