Confirmation Bias (also known as Positive Test Strategy) refers to the tendency amongst people to take greater notice of information that confirms and reinforces their beliefs whilst ignoring or diminishing the significance of any information that contradicts those beliefs.
Confirmation Bias is therefore a form of selective thinking that can strengthen prejudices despite empirical evidence against them.
If, for example, someone who believes in astrology reads their horoscope for the week and finds that one of the predictions 'comes true', they may take that as proof that horoscopes work, whilst neglecting to acknowledge that many of the other predictions made in the same week did not come true.
The concept pertains not only to how people process information but also to how people search for that information in the first place. When testing hypotheses for example, it is possible to use 'positive test strategies', that is to say, to phrase questions in such a way as to encourage answers that confirm a working theory.
It is also possible to show bias when remembering information. This is known as 'selective recall', 'confirmatory memory' or 'access-biased memory'.
An example that illustrates both how memory can be skewed and also how language can be used to exploit this is one study where subjects were given information to read about a woman.
The information included examples of both introverted and extroverted behaviour. One group of subjects was then asked to assess the woman for the job of librarian and the other group to assess her for the job of sales person.
While the first group recalled more examples of introversion the second recalled more examples of extraversion.
The term 'Confirmation Bias' was first used by Peter Watson in 1960, however it was Francis Bacon (15611626) who wrote that the biased assessment of information was behind all superstitions, whether in astrology, dreams, omens, divine judgements or the like'.
For more information:
For more information on the introvert/extrovert study see:
Snyder, M.; N. Cantor (1979). "Testing hypotheses about other people: the use of historical knowledge". Journal of Experimental Social Psychology 15: 330342. via Goldacre, Ben (2008). Bad Science. London: Fourth Estate. p. 231. ISBN 9780007240197.
These two sites provide clear and simple explanations and examples.