Clifford Nass (1958–2013)

This book provides an entertaining peek at the intersection of technology, sociology, and psychology. It discusses various experiments which indicate how people anthropomorphize inanimate technology and how this tendency can be used to reduce confounding variables and better understand human behaviour. It wasn't quite what I had expected: given the title and blurb, I had thought that the book would focus more on the mathematics and technology than the sociology. Instead, I found that the book instead discussed how technology could be used to improve sociological experiments and how these experiments, in turn, could be used to improve interfaces for human and computer interaction.


I found the writing itself somewhat problematic. I thought that Nass's ego tended to get in the way of the story-telling. Basically every successful experiment discussed was Nass's, and he is quick to point out how he solved issues that had been plaguing other "lesser" scientists for years.At some point, I started to count how many times the phrases "I solved," "I discovered," "I showed," etc, popped up, but I quickly lost count. In addition, the layout of idea/anecdote-experiment-conclusion-reaction became somewhat repetitive. However, one benefit of this format and dearth of overall narrative structure is that if a reader was only interested in particular areas, it would be easy to skip to those sections.

The ideas ranged from the obvious to the surprising and unintuitive. One of my favourite examples was Nass's discussion of how he had been brought in as a consult for Microsoft about the disaster that was "Mr. Clippy, the Animated Assistant." I remember Mr. Clippy. The first advanced setting I learned in Word was how to turn the darned thing off. Nass's advice, backed up by experiments, was to make Mr. Clippy tell the user that he is stupid (thereby flattering the user's ego) and apologize and berate himself and Microsoft whenever he made mistakes. For some reason, Microsoft wasn't exactly happy with this...can't imagine why... Overall, definitely an interesting read from a renowned scientist.… (mer)

Clifford Nass has done a lot of consultant work on computer interfaces and how people respond to them, and in the course of this career, he came to the realization that people react to computers very much the same way they react to other people, so that if, for instance, a computer does something that would be considered rude coming from a human being, that makes people like the computer less. Eventually he realized that this could also be turned around: if people respond to computers the way they respond to other people, we can perhaps learn some things about human social interactions by conducting psychological experiments using computers. So in this book, he talks about various experiments that he and others have performed, what he thinks they mean, and how that knowledge can be applied. (E.g., he learned that people interacting with a computer in a driving simulator performed better and liked the computer more if its tone of voice matched their own mood. So he concludes that trying to motivate a sad person by being relentlessly cheerful at them is probably counter-productive.)

Some of these experiments are really interesting, and some of the thoughts Nass has about them seem both insightful and useful. But, as is often the case with this kind of thing, his conclusions often are often much stronger and more sweeping than the actual evidence seems to warrant. I'm also dubious about some of his basic claims about psychology, especially in his chapter on personality. And I'm not just saying that because he labels introverts like me as "submissive" and "cold," although that certainly didn't help endear him to me. In fact, it made me seriously wonder why on earth I should be listening to detailed advice on how to handle social interactions from a guy who clearly doesn't even have a clue when he's insulting people. Also, the rather creepily cold-blooded emphasis on how to manipulate people in corporate settings makes me think that I am really, really not the target audience for this book, despite my interest in both psychology and computers. And I am devoutly glad that I don't work for a company that subscribes to his painfully rah-rah ideas about "team-building."

To be honest, I think I would have been much more interested in how human social instincts need to be taken into account in software design. Maybe I should have read his previous books, instead, although considering how much he managed to rub me the wrong way with this one, I think I'll pass.… (mer)

This is a light-weight take on psychology (although the references to meatier and more statistically examined papers are there) and in particular the ways we treat computers. Even people who know better - computer programmers for example - treat computers that give them a tiny excuse to do so as if they're a person. A "social actor" as it's coyly phrased. They attribute personality, gender and the like to the computer or the TV, they react to bigger faces as if they're closer and therefore more important and so on.

Particularly interesting are the summaries and implications of the various chapters - although you will notice that some of the chapters contradict themselves: psychologists chop their experiments into little bits and try to isolate certain factors. Sadly people don't work like that, and so there are some contradictions that they highlight and come from that.

There's a couple of bits where I don't trust the conclusions too, but by and large it's compelling and interesting.… (mer)