Learning From Digital Measurements of Face-to-Face Interactions

Article tools

Printer friendly	E-mail article	Subscribe	Order reprints
	Discuss any Chronicle article in our forums
Latest Headlines 150 Campuses Win Recognition in Survey of Great Colleges to Work For More than 300 four-year and two-year colleges signed up for the second annual survey by The Chronicle, and 39 were named to a new Honor Roll for their workplace policies. New Top Republican on House Panel: Washington Doesn't Always Know Best They Thought Globally, but Now Colleges Push Online Programs Locally Jill Biden Shines Global Spotlight on U.S. Community Colleges A Town Rich in Colleges and Cows Gains a Crown College Towns as Baby Names? A Few May Resist Trendiness Commentary Kevin Carey: Stanford U., Duke U., Rice U., ... and Gates U.?

By JEFFREY R. YOUNG

Mr. Pentland, a professor at the Massachusetts Institute of Technology's Media Laboratory, recently published a book called Honest Signals, which argues that hidden social cues are key to understanding the productivity of researchers and workers.

Q. So you've developed a new kind of scientific device to measure social interaction? What are these gadgets, and how do they work?

A. We call them sociometers. If you look at all the organizational science, almost all of it is done by surveys, and sometimes people mine the e-mail of workers. But we all know that all of the really important stuff happens face to face, and there hasn't been any quantitative measurement of that. So for the first time we can give everyone in an organization a little name badge with sensors in them, sociometers, and based on that we can produce a map of who talked to who and when, and how did they talk to each other.

Q. What exactly do these electronic name badges measure?

A. They measure things like who's around you and are you facing them? Are you standing, sitting, walking, or gesticulating madly? And then they measure a lot about the tone of voice that you use. Are you speaking or are they speaking? Are you speaking a lot or a little? Are you speaking with great consistency and certainty? Who's setting the pace of the conversation?

Q. How do they measure that?

A. Well, they look at the pattern of speaking versus not speaking. You can think of the turn-taking we do in conversation as a sort of dance, right? I speak, then you speak, and so on. And you can measure the statistics of that to figure out what is, in essence, a measure of dominance. And all that without listening to the words. We never ever listen to the words because that freaks people out.

Q. What has your research using these devices found?

A. We've done things like look at sales and salary negotiations, and it turns out you can predict who will come out ahead just by looking at the tone of voice in the first few minutes of the interaction.

And we had a whole bunch of midcareer executives at the business school pitching real business plans and being rated by other midcareer people, and we found that you could predict which business plan would be picked without listening to the facts of the matter at all. What they said and who they were didn't matter. What mattered was, Were they enthusiastic about it and sounded like they knew what they were talking about?

Q. So more organizations should analyze this kind of data?

A. The main finding of the book is that humans have the conscious forms of communication that we all think about and plan our organizations around, but we have a much more ancient channel of communication — which is our social sense. The sense of, Is somebody interested in this? Are they paying attention? Are they on the same page? Do they know what they're talking about?

People have the ability to read each other for these social relationships. What I've done is develop a series of computer tools that can be built into tiny little name badges or even into cellphones that allow you to read this sort of social information.