Honest Signals: How They Shape Our World

by Alex Pentland

Book Overview

How understanding the signaling within social networks can change the way we make decisions, work with others, and manage organizations.

How can you know when someone is bluffing? Paying attention? Genuinely interested? The answer, writes Alex Pentland in Honest Signals, is that subtle patterns in how we interact with other people reveal our attitudes toward them. These unconscious social signals are not just a back channel or a complement...

• Format: Paperback
• Language: English
• ISBN: 0262515121
• ISBN13: 9780262515122
• Release Date: September 2010
• Publisher: MIT Press
• Length: 208 Pages
• Weight: 0.55 lbs.
• Dimensions: 0.6" x 5.2" x 7.7"
• Age Range: 18 years and up
• Grade Range: Postsecondary and higher

Customer Reviews

Concise and Compelling

This book makes it clear that verbal communication is a recent evolutionary development in humans which has only replaced a modest amount of the communication that our pre-linguistic ancestors used. The fact that we are much more aware of our verbal communication than our other forms of communication shouldn't cause us to underestimate those other forms. A good deal of the studies mentioned in the book consist of measures of nonverbal communication in, say, speed dating can predict results about as reliably as I'd expect from analyzing the words. These could be criticized for not ruling out the possibility that the nonverbal signals were merely responses to information communicated by words. But at least one study avoids this - entrepreneurs pitching business plans to VCs showed nonverbal signals that were excellent predictors of whether the VCs would accept the business plans, before getting any verbal feedback from the VCs. Even more surprising, investments made by VCs with nonverbal information about the entrepreneurs did better than those evaluated on written-only presentations. The sociometers used to measure these nonverbal signals have potential to be used in helping group decision making by automatically detecting the beginnings of groupthink or polarization, which should in principle allow leaders to stop those trends before they do much harm. But it's not obvious whether many people will want to admit that analyzing the words of a conversation has as little importance as this research implies. One of the more interesting methods of communication is for people to mimic each others body language. This is surprisingly effective at creating mutual interest and agreement. The sociometer data can be of some value for information aggregators by helping to distinguish independent pieces of information from redundant information by detecting which people are likely to have correlated ideas and which are likely to have independent ideas. I wish this book were mistaken, and that most of human interaction could be analyzed the way we analyze language. But it seems clear that unconscious parts of our minds contain a good deal of our intelligence.

Idiots and Gossip (Plus Other Tales from the Sociometer)

Idiots and gossip represent the biggest danger to idea markets and networked intelligence says MIT Media Lab Professor Alex Pentland in his findings, "Honest Signals." Of particular note is that in large groups behavioral dynamics can cause for less than stellar results via bad ideas introduced (idiots) and shared sources that repeat the same information over and over again (gossip). Anyone who has questioned the 2.0 echo chamber or wisdom of crowds can identify with these issues, yet Pentland demonstrates networked intelligence is superior to the individuals. Honest Signals reveals findings from a new technology called the Sociometer that measure human behavior, including overwhelming proof that humans do not make rational, logical decisions, instead opting for a base networked form of primal signaling amongst ourselves. This empirical evidence proves collaborative use of body language and other signals are more important in communications and decision making than theories of messaging and big man management. Further the findings bulwark the collaborative trends we are seeing in the social web, which brings us back to idiots and gossip. Anyone who has participated in Twitter or a highly engaged wiki environment can see this networked intelligence at work. And often the wisdom of the crowd can go astray in a bit of a frenzy or simply put, bad group-think. So the question becomes how to improve idea markets for better collaboration, performance and use, an activity the Media Lab, Intel and Hewlett-Packard are all actively trying to solve. The idiot factor -- introduction of bad ideas -- can easily be weeded out by performance. Someone who cannot deliver good intellectual capital simply loses credibility (idiot image by Geoff Greene). The gossip factor seems to be much tougher. While "me, too" may count as approval, the sociological problem lies in a variety of societal pressures (cliques, etc.) and subjective mental quirks. One idea spread across many is not many ideas, rather it's still only one alternative and its popularity may be temporal. For those who lament the echo chamber, we have to be discerning in large distributed environments and community idea markets like the blogosphere and Twitter, respectively. It's important to source ideas and understand which ones come from independent sources and which ones are simply, "me, too" theories. A couple of tips from Honest Thinking include 1) tight social groups rarely have multiple unique ideas and 2) make sure you use different sources of information than some other friends/acquaintances in the echo chamber. Number two is something I manage diligently in my Google Reader, quickly purging blogs which start miming other voices. You'd be surprised how many top bloggers actually present "unique" posts that in actuality seem to trailing other lesser known, more original thinkers. Other Findings Perhaps more relevant for the general communicator are the base sociometer findings, "that a great

A quantification solution for management practitioners

Dr. Pentland's research and the use of "sociometer" will offer a way to perform a quantitative analysis on data that till now was purely a "black art". For example, market researchers can use a sociometer to add a dimension of validity to answers given by interviewees. The potential of new applications using the sociometer can be a ground breaking tool for management consultants, trainers and auditors.

Beyond "Pop Science"

This book goes through the most recent (about last 5 years) research of Prof. Pentland's group in the MIT Media Lab. It's a quick but extremely engaging read, and in contrast to other pop science books like Freakanomics and Predictably Irrational (both of which are interesting reads), Honest Signals has the scientific details of the experiments that it talks about, in the form of a thorough 50-page appendix. For anyone interested in how sensing technology will change business and the sciences or who's interested in learning how people actually interact with each other, this is a must read.

sharpening stones, walking on coals

Business management has been ruled by mysticism and superstition for a long time. That ends now. Take a breathless ride with 'Sandy' on a whirlwind tour of the future of social engineering. He introduces new tools and methods that are bringing the stunning power of data-rich, observational science to bear on the heart of human endeavor: the search for influence, money, and a mate. The nerds have finally broken the code... and humanity may never be the same again.