There are plenty of books that try to tell you how to be happy, but Stumbling on Happiness by Daniel Gilbert isn't meant to be one of them. As Gilbert says such books are 'in the self-help section two aisles over'. This is the book to read after you've read those books, as it explains why we often fail miserably in our quest for happiness. Some people say that to achieve happiness we should live each minute as if it were our last, but Gilbert questions whether such people would really spend their last 10 minutes giving such dumb advice to others. No, the search for happiness is not that simple.

The first thing to consider when trying to investigate happiness scientifically is how you are going to measure it. Gilbert points out that there are several types of happiness - there's the happiness of the moment and there's longer term satisfaction at doing what is right and progressing towards our goals. He thus identifies Emotional, Moral and Judgemental Happiness. But even if we know what type of happiness we are measuring, how can we put a number to such an intangible entity. Using the example of conjoined twins - not a life most of us would choose - Gilbert examines how different people might derive different amount of happiness from the same situation, and how we can't really know whether this is justified without having exactly the same experience as them. In the end though it seems that we have to accept the measures people assign to their own happiness, and trust that the law of large numbers will average out any distortions.

Much of the book concerns how our minds get things wrong. We remember things wrongly - for instance we fill in details which weren't necessarily there. We also remember unusual occurrences, but have a tendency to forget that they are not the norm. Most of all, we tend to try to reinterpret things in a more positive light. If this means seeing something good in a seemingly negative event then this may be considered beneficial, in particular if the negative event is unavoidable (this is the inescapability trigger). Interestingly, it is often easier to have a positive view of really bad events than of everyday misfortunes (the intensity trigger). However, part of this reinterpretation is to guard our own self image, which may not be so beneficial. The books tell of all sorts of self-deceptions which people use to convince themselves that negative things which they hear about themselves don't reflect reality. Also people feel worse if such rejection is from a group decision rather than one person (which is easier to explain away). Interestingly though, the subjects don't predict that they would make such a distinction beforehand.

This result reflects a central theme of the book - how bad we are at predicting how we will feel in the future. Several of the studies in the book show that factors which we didn't consider to be important at the start will actually have a big influence on how we feel later. This is a significant failing if we are trying to organise our lives to consist of what we think will make us happy. Although we're pretty good at predicting the immediate outcome of what we do, predictions about something that is more than a day or so ahead are a different matter entirely. We find it all to easy to base our decisions for the future on how we feel at present - an obvious example is how whether we are hungry or not when we walk round a supermarket will have a big influence on what we buy for our future meals.

Gilbert is a professor of psychology at Harvard, and draws upon plenty of psychological experiments to demonstrate how our minds work, and so to make his case. However, the book never reads like an experimental report. Rather it is full of witty comments, examples of illusions and even the odd card trick. My one minor criticim would be that the book tends to go for amusing anecdotes rather than memorable content - I'm not really sure what I learnt from the book, except never to believe experimental psychologists when they tell the subjects of their experiments what they are trying to test. On the other hand there is some useful advice - for example if you are considering whether to follow a certain path in your life, listen to others who have followed that path rather to your own imagination. I'm not sure whether this book will help you to find long term happiness - it may well provide a few pointers. But in the short term - well Gilbert is clearly a skilled writer and has produced a book that is certainly fun to read. I'd say that it has a very good chance of winning the General prize of the 2007 Royal Society Prizes for Science Books