How We Know What Isn't So: The Fallibility of Human Reason in Everyday Life [Hardcover]

I've been bumbling around in the area of behavioural economics/finance for a bit as it has some bearing on my day job. As a result I have developed a geeky interest in the area of mistakes in reasoning, biases etc. The book is a really good introduction to the area and very clearly-written.

It basically gives you a much better insight into some of the things you have probably already noticed (for example, people only seem to notice/value information that confirms what they already think). Aside from the fact that it's a genuinely fascinating area, you might also glean some information that actually helps you in the real world. I was struck by the point that partners tend to notice when their other half does (or doesn't do) something, that causes them pain, but not when the reverse. Hence I will notice if the Mrs doesn't stack the dishwasher, because it annoys me, but won't register when she does do it, so I will tend to take from this that she "never" stacks the dishwasher.

Usefully it also takes the insights from wonky reasoning and applies them to "questionable" beliefs - ESP, 'alternative medicine' etc. And you might learn a bit more about how to tell a good story!

Definitely worth a read. The other book recommended alongside this - Inevitable Illusions - is less good, but also worth a look. It gets into the interesting area of probability (and how bad we are at estimating and understanding it).

I can scarcely recommend this book highly enough to anyone interest in widespread errors of reasoning. Working in the same branch of social science as Thaler, Tversky and Kahneman, the author manages to explain very clearly how we fall into reasoning traps, probably because of the environment in which our brains evolved.

Although Gilovich's case studies are interesting (belief in ESP, and belief in extreme forms of "alternative" medicine), it's the theoretic part of the book which really stands out as being of benefit to the reader. Anyone who reads a newspaper, watches the television news, votes or invests in the financial markets ought to take the time to read this book.

In his introduction to this classic work on the fallibility of human reason in everyday life, Thomas Gilovich asks, "Why worry about erroneous beliefs?" It's tempting not to - after all, aren't we supposed to tolerate another person's beliefs, however wacky, and respect them too? Where's the harm in "a little superstition" or in someone visiting Lourdes in the hope of a miracle? Gilovich begins with some striking examples of harm being caused in the natural world: black rhinos, green-haired turtles and black bears have all been hunted to near extinction as a result of "mistaken beliefs about aphrodisiacs and cancer cures". If you are now congratulating yourself that you would never fall for the idea that a bear's gall bladder could cure your indigestion, Gilovich has plenty more cognitive biases for you to choose from. No matter how sophisticated we think we are, we're all suckers for some kind of wishful thinking.

Part I of the book opens with an important claim. "Human nature abhors a lack of predictability and the absence of meaning." We tend to see order in the "the often messy data of the real world" where there is none. Part II examines why we might want to hold questionable beliefs, and the role society plays in supporting or promoting those beliefs. Part III looks at several areas in which erroneous beliefs flourish, including the fertile ground of alternative medicine. Finally, in case you were beginning to lose all hope in humanity, Part IV shows "how we might improve the way we evaluate the evidence of everyday life".

Evidence of whatever provenance is important for most people, but one of the major themes of the book - indeed of this whole field of research - is to show how evidence cannot always be neatly bagged and labelled like an exhibit in a court case. "For nearly all complex issues, the evidence is fraught with ambiguity and open to alternative interpretation." We "often fail to recognize that a particular belief rests on inadequate evidence" and are "prone to self-serving assessments". People have even been found "to attribute their successes to themselves, and their failures to external circumstances." Who'd have thought it?

We humans are rightly proud of our ability to see patterns in nature, but we overreach ourselves when we extract "too much meaning from chance events". Professional basketball players and their fans fail to recognize randomness when they talk about winning or losing streaks, and so commit the "hot hand" fallacy. Gilovich and his colleagues discovered that the outcome of any given shot has no predictable influence on the outcome of the following shot. The powerful impression that there is some kind of connection between a sequence of similar events is "the clustering illusion" and the temptation is then to "explain" the phenomenon with "superfluous and often complicated causal theories." Ignoring regression to the mean has rather more serious consequences in education, for example, where spurious regimes of reward and punishment result in a lot of wasted time and energy.

Our best defence "against erroneous beliefs" is, basically, science. There is no magic formula, only easily understandable principles, such as insisting upon "replicability and the public presentation of results". Ideas and findings "that rest on a shaky foundation tend not to survive in the intellectual marketplace." Contrast that with our everyday lives, in which we tend to seek confirmation of our beliefs, not contradiction. It is as if we ask ourselves "Can I believe this?" for what we want to believe and "Must I believe this?" for what we don't want to believe.

"Many of our most bizarre and erroneous beliefs do not survive our interactions and discussions with others." Religious leaders understand this, which is why they fear the books by so-called militant atheists they would once have burned. It is not only the religious who are thankful that people "are generally reluctant to openly question another person's beliefs." Fringe medicine is also "plagued by questionable, erroneous, and often harmful beliefs". Few of us realize the extent to which the "body is a truly amazing machine with remarkable powers to set itself right" or are aware of the "common fluctuations in the course of most diseases". By falling for the post hoc ergo propter hoc fallacy, quacks claim any temporary improvement for themselves (while of course accepting no responsibility for any deterioration).

Confidence is one of those go-to words - especially at times of financial crisis or political change - and is thought to be a universal good. Yet we should remember that those holding false beliefs can be confident too, often more so than those who would question those beliefs. Reading Gilovich will reduce certainty and confidence where they are unwarranted, but will reward you with a surer footing in a complex and often unpredictable world. Be prepared for "conflict and disharmony" as you question beliefs that are unquestionable, and set the bar low. When we tie ourselves in irrational knots, don't expect the intellect to make any great leap.