## Confirmation Bias ### Beware the Special Case: Gil wants to lose weight. He selects a particular diet and checks his progress on the scales every morning. If he has lost weight, he pats himself on the back and considers the diet a success. If he has gained weight, he writes it off as a normal fluctuation and forgets about it. For months, he lives under the illusion that the diet is working, even though his weight remains constant. Gil is a victim of the confirmation bias – albeit a harmless form of it. The confirmation bias is the mother of all misconceptions. It is the tendency to interpret new information so that it becomes compatible with our existing theories, beliefs and convictions. In other words, we filter out any new information that contradicts our existing views (‘disconfirming evidence’). This is a dangerous practice. ‘Facts do not cease to exist because they are ignored,’ said writer Aldous Huxley. However, we do exactly that, as super-investor Warren Buffett knows: ‘What the human being is best at doing, is interpreting all new information so that their prior conclusions remain intact.’ The confirmation bias is alive and well in the business world. One example: an executive team decides on a new strategy. The team enthusiastically celebrates any sign that the strategy is a success. Everywhere the executives look, they see plenty of confirming evidence, while indications to the contrary remain unseen or are quickly dismissed as ‘exceptions’ or ‘special cases’. They have become blind to disconfirming evidence. What can you do? If the word ‘exception’ crops up, prick up your ears. Often it hides the presence of disconfirming evidence. It pays to listen to Charles Darwin: from his youth, he set out systematically to fight the confirmation bias. Whenever observations contradicted his theory, he took them very seriously and noted them down immediately. He knew that the brain actively ‘forgets’ disconfirming evidence after a short time. The more correct he judged his theory to be, the more actively he looked for contradictions. The following experiment shows how much effort it takes to question your own theory. A professor presented his students with the number sequence 2–4–6. They had to calculate the underlying rule that the professor had written on the back of a sheet of paper. The students had to provide the next number in the sequence, to which the professor would reply ‘fits the rule’ or ‘does not fit the rule’. The students could guess as many numbers as they wanted, but could try to identify the rule only once. Most students suggested 8 as the next number, and the professor replied: ‘Fits the rule.’ To be sure, they tried 10, 12 and 14. The professor replied each time: ‘Fits the rule.’ The students concluded that: ‘The rule is to add two to the last number.’ The professor shook his head: ‘That is not the rule.’ One shrewd student tried a different approach. He tested out the number -2. The professor said ‘Does not fit the rule.’ ‘Seven?’ he asked. ‘Fits the rule.’ The student tried all sorts of numbers -24, 9, -43?. . .?Apparently he had an idea, and he was trying to find a flaw with it. Only when he could no longer find a counter- example, the student said: ‘The rule is this: the next number must be higher than the previous one.’ The professor turned over the sheet of paper, and this was exactly what he’d written down. What distinguished the resourceful student from the others? While the majority of students sought merely to confirm their theories, he tried to find fault with his, consciously looking for disconfirming evidence. You might think: ‘Good for him, but not the end of the world for the others.’ However, falling for the confirmation bias is not a petty intellectual offence. How it affects our lives will be revealed in the next chapter. See also [[Availability Bias]] (ch. 11); [[Feature-Positive Effect]] (ch. 95); [[Coincidence]] (ch. 24); [[Forer Effect]] (ch. 64); [[Illusion of Attention]] (ch. 88) ### Murder your Darlings In the previous chapter, we met the father of all fallacies, the confirmation bias. We are forced to establish beliefs about the world, our lives, the economy, investments, our careers and more. We deal mostly in assumptions, and the more nebulous these are, the stronger the confirmation bias. Whether you go through life believing that ‘people are inherently good’ or ‘people are inherently bad’, you will find daily proof to support your case. Both parties, the philanthropists and the misanthropes, simply filter disconfirming evidence (evidence to the contrary) and focus instead on the do-gooders and dictators who support their worldviews. Astrologers and economists operate on the same principle. They utter prophecies so vague that any event can substantiate them: ‘In the coming weeks you will experience sadness,’ or ‘in the medium term, the pressure on the dollar will increase.’ But what is the medium term? What will cause the dollar to depreciate? And, depreciation measured against what – gold, yen, pesos, wheat, residential property in Manhattan, the average price of a hot dog? Religious and philosophical beliefs represent an excellent breeding ground for the confirmation bias. Here, in soft, spongy terrain, it grows wild and free. For example, worshippers always find evidence for God’s existence, even though he never shows himself overtly – except to illiterates in the desert and in isolated mountain villages. It is never to the masses in, say, Frankfurt or New York. Counter-arguments are dismissed by the faithful, demonstrating just how powerful the confirmation bias is. No professionals suffer more from the confirmation bias than business journalists. Often, they formulate an easy theory, pad it out with two or three pieces of ‘evidence’ and call it a day. For example: ‘Google is so successful because the company nurtures a culture of creativity.’ Once this idea is on paper, the journalist corroborates it by mentioning a few other prosperous companies that foster ingenuity. Rarely does the writer seek out disconfirming evidence, which in this instance would be struggling businesses that live and breathe creativity or, conversely, flourishing firms that are utterly uncreative. Both groups have plenty of members, but the journalist simply ignores them. If he or she were to mention just one, the storyline would be ruined. Self-help and get-rich-quick books are further examples of blinkered storytelling. Their shrewd authors collect piles of proof to pump up the most banal of theories, such as ‘meditation is the key to happiness.’ Any reader seeking disconfirming evidence does so in vain: nowhere in these books do we see people who lead fulfilled lives without meditation, or those who, despite meditation, are still sad. The Internet is particularly fertile ground for the confirmation bias. To stay informed, we browse news sites and blogs, forgetting that our favoured pages mirror our existing values, be they liberal, conservative or somewhere in between. Moreover, a lot of sites now tailor content to personal interests and browsing history, causing new and divergent opinions to vanish from the radar altogether. We inevitably land in communities of like-minded people, further reinforcing our convictions – and the confirmation bias. Literary critic Arthur Quiller-Couch had a memorable motto: ‘Murder your darlings.’ This was his advice to writers who struggled with cutting cherished but redundant sentences. Quiller-Couch’s appeal is not just for hesitant hacks, but for all of us who suffer from the deafening silence of assent. To fight against the confirmation bias, try writing down your beliefs – whether in terms of worldview, investments, marriage, healthcare, diet, career strategies – and set out to find disconfirming evidence. Axeing beliefs that feel like old friends is hard work, but imperative. See also [[Introspection Illusion]] (ch. 67); [[Salience Effect]] (ch. 83); [[Cognitive Dissonance]] (ch. 50); [[Forer Effect]] (ch. 64); [[News Illusion]] (ch. 99)