The Man Who Lied to His Laptop: What Machines Teach Us About Human Relationships [Hardcover]

Lie to your computer?

Well, sure. In today's high tech world, lying to one's computer is little different than sweet talking your car, pleading with a slot machine to produce a winning combo or threatening a big garden boulder that refuses to move.

In other words, it's really about how people react to situations good and bad. The basic reasoning is simple: People have an instinctive "personal" commitment to the task at hand. We are hard-wired to cooperate with others, as deftly explained by Michael Tomasello in 'Why We Cooperate.' As Nass and Yen make delightfully clear in case after case, it's human nature to talk to machines.

Many years ago, Dale Carnegie wrote the classic 'How to Win Friends and Influence People.' Nass would have would have listened to car drivers and truck mechanics and written "How to Talk to Your Car and Influence Trucks.' Since computers are now ubiquitous, he listens to people talk to computers. The result won't make the computer any smarter, but it does a lot for people.

The result is a superb book about people. Computers are like cats, the gods of our society. Neither cats nor computers listen to humans, but people pay attention to both and are much the better for it. Look at a Neolithic effigy and think of the conversations Neanderthals had with it.

The chapter on teams and team building is wonderful. Most team building gimmicks are like watching the Dallas Cowboys cheerleaders -- they amuse the fans but don't teach players a thing about football. Fans want to see a winning team, more so than fancy pants dance routines.

In business, cheerleader events are "wilderness bonding" and other play-acting gimmicks. Instead of fantasy events, Nass and Yen describe how people react to situations good and bad. Yes, repeating that phrase is deliberate, because the book deals with dozens of real situations -- good and bad. It's an exercise on how to deal with people, using machines as a neutral example.

Consider, for example, a car that tells people how to drive safely. It sounds wonderful; but, if not done properly, it enrages drivers and causes accidents. Now, think of bosses who tell employees how to work efficiently and effectively; if not done properly, they enrage people and cause output to decline.

Having spent much time since the 1960s telling people what to do ... all I can say is I wish I had had this book in the 1960s. The premise is ingenious; namely, if you can learn to speak properly to your machines, you'll do fine when dealing with people.

People are social beings. It can be argued that the concept of self can only be defined in the context of our own interest in what other people are thinking and feeling. Nass realizes that these social behaviors may be so ingrained that they appear even when interacting with computers, and conducts his psychology experiments using machines as easily controllable partners.

The results are interesting. Many actual studies are described and explained, which I like better than a more prose-heavy argument. However, I disliked how few counterarguments were presented, and how simplistic people were at times made up to be. While there certainly are patterns in human behavior, I don't think situations are always as cut-and-dry as the authors make it sound.

Even if I don't think it's applicable to every situation, I learned a lot about social science from this book, and how to quantify or measure some abstract concepts. Things like retrograde interference, identification/interdependence, and valence/arousal are useful ways of thinking about how people behave, and they're explained very well. It is also particularly helpful that there is a focus on counter-intuitive findings, which end up making sense and forming an overarching consistent picture.

I can only echo Nass' praise of Corina Yen's writing, which must have made it able to transform a large quantity of data into a clearly presented argument, with the right emphasis and concision to make it an absorbing read. I recommend this book to anyone who wants to gain some insight into how people (yourself included) think and why they act the way they do. With practice I even think it will make me a better reviewer!

It's great when a book that is this full of serious research is also well written and highly entertaining. Nass and Yen bring together a collection of experimental results that repeatedly demonstrate how people treat computers like other people, and how we can draw conclusions about human social behavior from these human-computer experiments. The book is a great read--informative and funny, if a bit creepy when you think through its implications. I wrote a longer reaction on my blog:[...]
Disclaimer: I received a review copy from the publisher.

› Go to Amazon.com to see all 11 reviews  4.3 out of 5 stars