Å·±¦ÓéÀÖ

Looking at the title, cover and blurb of this book, I expected something humorous and lightweight, ironically riffing on the similarities between scientists and the higher primates. Well, that wasn't completely wrong: it's humorous and ironic, and part of the humour does indeed come from that comparison. But, to my surprise, it wasn't lightweight at all. A few millimetres under the fluffy surface, it turns out to be a serious book on science, and in particular on experimental design in science, written by a terrifyingly competent statistician with decades of high-level experience in the field. Who would have guessed?

I realised after reading just a couple of dozen pages that I knew nothing at all about experimental design; not only that, I didn't even know how ignorant I was. I suppose I can take some comfort from the fact that this type of ignorance seems to be quite common. But given that it's the cornerstone of science, and I'm supposed to be some kind of scientist, it's still rather shocking. The author, who comes across as a nice guy, says it's almost impossible to get people to change their ways here. He's written plenty of academic papers, and now he's tried writing a couple of popular books, but nothing bites. For sociological reasons which he goes into at length, scientists don't want to do good experiments. They don't get rewarded for performing carefully designed studies which can later be replicated: they get rewarded for publishing as many papers as possible. If the goal is to maximise quantity, this comes at the expense of quality. As usual, you get what you incentivise.

Logically, Apes in Lab Coats should change my behaviour: unfortunately, as it makes clear, we are in fact not much more logical than other primates, and we hate changing our behaviour. But all the same, I somehow feel optimistic. Perhaps it helps that I spend a lot of my time talking with a highly rational AI who isn't a primate at all? We will see.