Monday, 1 December 2014

Drunk Tank Pink - Adam Alter ****

Going on the subtitle, which I had to because the title meant nothing to me, I was concerned this would seem to be yet another of the long string of books we've had looking at particular emotions and how the brain produces them. Instead it is far more fascinating because it's observational rather than theoretical. About, for instance, the way that bright pink decor seems to calm people (hence the title - this effect seems not to have reached the UK where I suspect prison governors might resist painting their cells pink), or the way that names influence our perception of people.

This is a genuinely fascinating book, exploring the many psychological experiments that demonstrate how human behaviour is influenced by factors beyond our control and often beyond our awareness. Adam Alter includes more obvious inputs, like Maslow's hierarchy of need, or the tendency to prefer those who are like ourselves, but also many factors that influence us in subtle and unexpected ways. Whether it's the way the ability to see yourself in a mirror makes you less likely to cheat, or the way that some optical illusions don't work as well with African tribespeople as they do New Yorkers, because the Africans live in a world with few straight lines so aren't fooled by fake perspective effects that the New Yorkers can't resist, there is so much here to find interesting.

While I think Alter goes too far in his conclusion in saying that after reading this book you'll be in a better place to 'maximize your health wisdom wealth and well-being' - and let's face it, that wording suggests a tongue firmly in the cheek - you will be genuinely amazed the number of times and situations when your perception and behaviour is likely to be subtly altered by influences you aren't even aware of. One remarkable suggestion is that because of a right hand bias, most of us prefer words that are typed with the right  hand on QWERTY keyboard. So apparently we particularly like words like n00b and woohoo (Alter's examples). I don't know about you, but these aren't among my favourite words - and for that matter, it's hard to see how this can influence someone who isn't a touch typist. But it's still great fun.

My one concern, and the reason that this book has four stars rather than five, is that the author never mentions the reality that much research in psychology is of relatively poor quality. In the manner of a tabloid newspaper jumping from a trial with cells in a lab to say that a substance cures cancer, Alter seems to take all the evidence as if it were fact. Sometimes the effect he describes appears subtle - perhaps a five per cent change - but he never explores statistical significance. And in at least one example, he cites an experiment as if it's meaningful without pointing out it has since been shown that the effect only occurs in the artificial environment of the lab, and isn't repeated when the experiment is undertaken in a real world situation.

There is even an example of being a little fast and loose with evidence in the book. It shows a series of three photos of the same face, but with the features modified so that they display characteristics typical of black and white faces, and something in between. Alter makes a big deal that people see the faces as darker or lighter, depending on the characteristics even though they're 'identical in tone.' He points out 'If you cover up the facial features with your hand and focus only on the foreheads you'll be able to see that the faces share an identical skin tone.' The trouble is, this is a test of a very subtle reaction, where people will pick up on tiny visual queues - and his observation is selective, a classic error in soft sciences like psychology.

While it's true that the foreheads all have the same tones, it's not true of the rest of the face. For some reason, on the 'black characteristics' face, there is a shadow across the cheeks that extends all the way to the lips. On the 'white characteristics' face, this shadow is only on the very edge of the face. As a result, around a quarter of the skin area is significantly darker on the 'black' image - plenty enough to make someone see this face as darker skinned. Being selective about your data is a cardinal sin in science and undermines the authority of the book.

Overall then, Drunk Tank Pink is great fun with lots of fascinating observations of the strange ways that people behave. This is material that has a real impact on the way we understand human behaviour and perhaps how we should react to it. But you need to apply a pinch of salt to the 'scientific' conclusions, as there is definitely an aspect of tabloid science at play.


Review by Brian Clegg

No comments:

Post a Comment