[Jonathan Cainer] met a psychic poet called Charles John Quatro, who told him he would some day write an astrology column read by millions.
David Smith, the Observer , 20 June 2004

And would you believe it, many years later, Jonathan Cainer does write an astrology column read by millions! Incidentally, Cainer’s predictions grace the pages of “a newspaper dedicated to the subtle propagation of bigotry.” That description of the Daily Mail is by, ahem, Jonathan Cainer.

Are you impressed by the uncanny accuracy of Quatro’s prediction? Let me make my own predictions: if you already believe in astrology, your answer will be yes. If you don’t, your answer will be no. If you’re agnostic, you will probably find it somewhat impressive.

Was I right? Probably, though, not because I possess any psychic powers myself. Rather, I am simply aware of an effect psychologists call confirmation bias. This concerns how we filter out the mass of evidence for or against various theories and hypotheses. In short, what it means is that we consider evidence that supports what we already believe to be stronger or more significant than that which undermines it. Indeed, we may go so far as to pay little or no attention at all to contrary evidence and focus our attention almost exclusively on that which bolsters our prior convictions.

This explains why so many people are impressed by the claims of psychics and astrologers. If we are inclined to believe in the supernatural, then it is easy to focus on those examples where prediction comes true, or where psychics make accurate statements about the past or present. These “confirm” our beliefs that they really do have access to a source of knowledge beyond the physical world, or at least the world as science standardly understands it.

If we do not believe in the supernatural, however, we will focus on the countless times when predictions are wrong or when psychics make mistakes. Reading the article about Jonathan Cainer, for example – setting aside doubts about the truth of the story – we will think that this one accurate prediction doesn’t count for much, for the psychic probably also said many other things that were not true.

It should be clear, therefore, that people on both sides of the debate can fall victim to confirmation bias. However, it should also be clear that, in this case, confirmation bias works more to the benefit of believers than skeptics. This is because if we try to take a genuinely balanced look at the evidence, we will find that for every apparent instance of a true prediction by an astrologer there are many other false ones. What is more, many predictions are so vague that it is always possible to say that they came true in some sense. Confirmation bias is therefore more likely to lead the believer into error because the balance of evidence just does stack up against the truth of astrology. It is only by selecting the evidence to fit their beliefs that they could possibly come to the conclusion that astrology works. That is, unless I have failed to overcome my own confirmation bias against the paranormal.

Confirmation bias infects political discourse too. It is almost certainly the case that, once they were persuaded that Iraq had WMD, Blair and Bush placed more weight on evidence that supported their position than that which challenged it. They may have tried to keep open minds, but once you have committed yourself to what you see as the truth, it becomes very hard to assess all the evidence impartially.

On the other side, those who are persuaded that Bush and Blair are driven by purely selfish motives are much more impressed by evidence that supports this view than that which suggests that they might be sincere, even if mistaken. The genuinely open question of whether they lied or were mistaken about WMD becomes an open and shut case in the face of “clear” evidence that they lied, while any counter-evidence is dismissed.

Confirmation bias is a real impediment to good thinking, but unlike some errors in reasoning, it is very hard to root out. No one can expect to become totally immune to it: it requires constant fighting.