Thinking, fast and slow – and thinking hard

The title of this book,

, captures the fundamental point that we have two systems of thinking – Kahneman labels them system 1 and system 2. S1 acts automatically, doing things such as recognizing stock phrases, responding to sudden noises or unpleasant pictures, driving in easy conditions. It is the source of intuitions and emotional responses. It kicks in first, without us having any capacity to stop it. S2 is the deliberative system, making calculations or doing any non-obvious thinking, and is a fall-back system because it requires effort and energy (although tasks become less energy-intensive as one becomes more skilled at them). Kahneman describes it as ‘lazy’ – S2 won’t bother to get motoring if S1 appears to have found an adequate answer. Exercising self-control is also a S2 task, and fatigue, or alcohol, as well as cognitive load can diminish our self-control. If S2 is busy, S1 is more likely to successfully demand a piece of chocolate cake rather than an apple. (There is a handy summary of S1 characteristics on p105.)

Much of the book sets out (with touching, frequent reference to Kahneman’s deceased co-researcher Amos Tversky) the biases systematically uncovered by a decades-long programme of research into how we think. Most of these are now well-known. For example, priming effects are ubiquitous – reminders of money make for more selfish responses and a greater preference for solitude or distance, for example (p55). People are more readily persuaded by anything that reduces cognitive strain, so clear bold fonts that are either blue or red and printed on high quality paper with sharp contrast are literally more persuasive. So too if you use simple language and use constructions like ‘one person in a hundred’ rather than ‘1% of respondents’; the very word ‘percentage’ seems to inhibit rational thought (see p163).

An important principle of thinking is ‘WYSIATI’ – what you see is all there is. Our mind focuses on activated information, and any other information might as well not exist. “System 1 excels at constructing the best possible story that incorporates ideas currently activated, but it does not (cannot) allow for information it does not have.” (p85). S1 measures its success by the coherence of the story it creates from what is immediately available. When this is relevant to the thinking task at hand, the result is an impressive feat of intuition; otherwise, the result is one of the biases Kahneman documents, such as over-confidence, over-estimating the probability of rare events. The S1 insistence on coherence also makes us highly vulnerable to causal explanations, and rubbish at assessing randomness. People will often see patterns in genuinely random data, and greatly under-estimate the likelihood of random clusters – something that is ominously apparent in a number of court cases one can call to mind. It also means we have a strong hindsight bias, imposing largely illusory histories on the course of events, which often owes more to chance than is commonly acknowledged (see pretty much any business biography or history book).

In addition, S1 deals readily with averages but very poorly with sums, because it works by example not by calculation. Calculation is a S2 task taking a lot of energy (especially chocolate cake). As is now familiar from behavioural economics, the upshot is that we have some strong biases in probability calculations, summed up in the famous prospect theory.

There is an amusing section on the stock market, recounting Kahneman’s demonstration to fund managers that their performance was worse than chance alone would have delivered – duly ignored by his audience. As also pointed out by Nassim Taleb in his

, most supposedly skilled investors have just been lucky. Kahneman comments, though, that: “The illusion of skill is not only an individual aberration; it is deeply ingrained in the culture of the industry. Facts that challenge such basic assumptions – and thereby threaten people’s livelihood and self-esteem – are simply not absorbed. The mind does not digest them.” (p216) I wrote in an earlier post about the consequences for expertise in any business where there are many ‘unknown unknowns’, including investing and economic forecasting.

The book does offer some good advice for dealing with S1. As well as the document design principles mentioned above, people’s behaviour can be changed by priming them – the instruction to ‘think like a trader’ shifts some people to S2 calculations. In groups, it is important to have someone who can offer an “outside view” given the natural tendency to under-weight the impact of extreme outcomes, and to fit stories to too little information (I have long done this in groups such as merger inquiries where lawyers will be scrutinising every step by asking one person in the group to play ‘devil’s advocate’ when we are near a conclusion.) The biases are not all bad, however. Business executives are almost delusionally over-optimistic, but if they were not there would be no enterprise and innovation.

The interplay between prospect theory and conventional economics is not, as Kahneman points out, a knock-out for the former. He describes the basic concepts of economics as ‘essential intellectual tools’, which in some circumstances predict behaviour better than prospect theory. What’s more, making decisions according to the pattern of risk-aversion for high-probability gains and low probability losses, and risk-seeking for high-probability losses and low probability gains will – over time – lead people to accumulate large expected losses compared to what their outcomes would be if S2 reigned supreme. S2 choices are more rational! Having said that, the existence of priming effects and reversals of preference do present a major intellectual challenge to economics.

It is, though, a challenge to which economists are responding with enthusiasm (see for example my post on this joint psychology and economics conference about scarce attention). Although the many critics of economics continue to perpetuate the myth that the subject depends on that chimera, the selfish hyper-rational individual, most economists I know no longer regard that set of assumptions as foundational to our subject. We are in the early days of a joint endeavour with psychologists and cognitive scientists to learn more about how real people make decisions and how that aggregates to social and economic outcomes.

is a brilliant contribution to that shared mission, and it will definitely be one of my top books of 2011.

[amazon_image id=”1846140552″ link=”true” target=”_blank” size=”medium” ]Thinking, Fast and Slow[/amazon_image]


One thought on “Thinking, fast and slow – and thinking hard

  1. Great review. Have just downloaded this book on my Kindle and looking forward to getting stuck in as soon as I have time. As a psychologist, I still remember how much I enjoyed studying Kahneman and Tversky’s work A an undergrad and how it seemed to bring psychology to life.


Comments are closed.