After you have finished reading this review, how do you foresee the rest of your day? Why do you expect an untroubled afternoon finishing off a few important tasks in time for catching the early train? Being constantly interrupted followed by an unforeseen near-disaster is just as likely. And why are you going to buy a lottery ticket on the way home, even though the odds of winning are 13 983 815 to 1?

Neuroscientist Tali Sharot thinks she has the answer. Her engaging book, The Optimism Bias, is all about how we expect the future to turn out rather better than it usually does. People tend to overestimate the possibility of positive events - for example, having a long-lasting relationship or professional success - and underestimate the likelihood of negative events such as illness or getting divorced. Sharot studies optimism at a neural level and knows her subject well. The book is written for a general audience and, as is voguish, mixes hard science with anecdote. So, waiting for a pint of Guinness to pour illustrates the power of anticipation and Sharot's friend Tim's holidays are used to explain how our choices can seem better after we have made them.

Why has optimism evolved? Sharot contends that when our brains predict positive outcomes for us, they are more likely to happen: self-fulfilling prophecies have a survival advantage. Optimism is also crucial for maintaining happiness. Depressed people are much more realistic than average; it seems that when people take off their rose-tinted glasses their equanimity goes too.

It is possible to dispute Sharot's conclusions; this is half the fun. It is a shame that there is little here which engages with Smile or Die, social critic Barbara Ehrenreich's recent examination of America's culture of positive thinking. Also, while I found Sharot's central thesis of a widespread human bias towards optimism persuasive, this theory can be over-stretched. For instance, she postulates that a bias towards optimism explains the current financial crisis, but I doubt whether such a bias is a suciently powerful explanation that all other theories can be disregarded.