Tricks of the mind: Confirmation bias

Posted on Updated on

In considering all the topics I’ve covered in this blog, such as pseudo-science, alternative medicines, medicine proper, and religion, there is one phenomenon which keeps appearing: confirmation bias. So let’s take a quick look at confirmation bias, and how it helps skew our view of reality.

Confirmation bias is what we all unwittingly do when we pay attention to things which confirm our beliefs, and forget those which don’t. We do this sort of thing because we are genetically programmed to make sense of the world. In his excellent book ‘The Greatest Show on Earth’, Richard Dawkins discusses this evolutionary trait extensively. He argues that man evolved the ability to detect and match patterns as a survival mechanism. Without this ability, each individual would be constantly analysing the environment before deciding on a course of action. The consequence of this would be greatly delayed reactions in sometimes life-threatening situations. (if any of you have suffered from ‘paralysis by analysis’ as I have from time to time, you’ll know what I mean). As a result, today we all possess a vast range of pre-programmed beliefs and behaviours, many of which are useful (for example, get out of the way of that speeding car, without needing to analyse the consequences of not doing so), and others not so useful (for example, that the position of the stars in the sky can forecast our future).

And so this bias manifests when we hold some strong belief, and try to fit the observed world to it. Of course, sometimes it’s not a genetic thing – it can also be self-delusion – wanting something to be true when it isn’t, or simply fraud, when test results are fiddled to achieve the required outcome.

The only way to overcome this bias, is to apply strict testing protocols such as double-blind testing, in which both the subject and the experimenter are unaware of how their choices will affect the outcome, and hence cannot influence the results accordingly to previously held beliefs.

The best way to demonstrate the extent of this problem is to assemble some common examples, so here we go.

  1. Astrology readings (also applies to palm reading, fortune telling, reading head bumps, so on and so forth): When believers read their horoscope, they inevitably feel that it is personalised for them. When things subsequently happen which confirm what they have read, like meeting a tall dark stranger, they remember those, but conveniently forget anything which doesn’t fit.
  2. Miracle cures – People with illness or injury, particularly chronic, desperately want to believe in miracle cures. See my earlier writings on Elmore. Although this oil is practically useless, except perhaps as a lubricant, many would report improvements which would have happened in any case. Since they expect/hope for this stuff to work, based on the advertising and testimonials, any improvements are ascribed to the oil. Conversely, if no benefit was received, this is often not taken as a problem with the treatment – it’s just one of those things that Elmore can’t cure. And of course if we feel like we’ve wasted our money on a quack cure, then we don’t rush out and tell other people about it.
  3. Religious miracles / prayer – this is the pinnacle of confirmation bias. That person who survives cancer when given a terminal prognosis is a miracle and hence evidence of the existence of god. The 100,000 people who don’t, are just ignored.
  4. ESP – A common fault with research into psychic phenomena is the retention of data which confirms the existence of psychic abilities, and discarding of data which does not. Amazingly, there are examples in which testers, when assessing the ability of people to psychically ‘see’ flashcards, have ignored a bad run of seeing as being a ‘warm up’ period, or a period in which the subject was beginning to tire. These bad runs have been discarded, leaving only the ‘good’ runs – which, not surprisingly, confirm their hypothesis.
  5. Superstition – Where do I start with this one? For example – ‘bad things come in threes’. We desperately seek to identify the three bad things, which we can generally do by being selective about the evidence (i.e. if there’s a 4th, then it’s just the start of the next batch of 3!) If there’s only two, then if we wait long enough, it will happen.
  6. Medical – the long running MMR-autism debate has a whole host of logical fallacies associated with it, but confirmation bias is definitely a player. Despite conclusive evidence that there is no causal link between MMR vaccines and autism, children who are diagnosed with autism around the time of their various vaccinations are used by the anti-vaccination lobby to confirm the incorrect belief that the vaccine caused the disorder. In fact, the original research by Andrew Wakefield, which started off this whole sorry story, suffered from intentional confirmation bias – he had a financial interest in finding a causal link, and, hey presto, the results showed it.

Anyway, I’m sure you can think of a million more examples.

Above all, try to recognise when confirmation bias is leading you to think or act inappropriately. Most of the time it doesn’t hurt, so have fun with your astrology, but sometimes, as is the case with the vaccination issue, the consequences can be disasterous.


5 thoughts on “Tricks of the mind: Confirmation bias

    reservoirdad said:
    February 28, 2011 at 12:58 pm

    Great article again. Thanks.

    […] some many people love it”: wishful thinking and self-delusion. Please read my earlier entry on confirmation bias – but beware, there are white coats […]

    Of Apes and Sci-fi « rationalbrain said:
    August 17, 2011 at 7:37 pm

    […] enough for me or anyone else to simply pick titles that fit in with my theory, for that is simply confirmation bias, which is a major source of error for unwary (or unprincipled) […]

    Middle child horoscope « rationalbrain said:
    September 2, 2011 at 9:29 am

    […] This smacks of reaching a conclusion, or holding and opinion, and then going out to find evidence which supports it (while at the same time probably discarding evidence which doesn’t). See this link to confirmation bias. […]

    […] I need to lecture you on confirmation bias again? No? […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s