Motivation and emotion/Book/2018/Confirmation bias motivation

Overview
"When men wish to construct or support a theory, how they torture facts into their service!" (John Mackay, Extraordinary Popular Delusions and the Madness of Crowds, 1852) Have you noticed that some people seem to read the news, and come to a conclusion that not even the stories they reference seem to endorse?

People avert their eyes, and dismiss the facts that lay before them, because they seem convinced that something else is at the centre of the problem - for instance, their idea of science seems to lack mainstream evidence, or their interpretation of the world requires that you ignore many things widely known about it.

These peculiar behaviours result from a phenomenon known as Confirmation Bias (Nickerson, 1998) - shortened to CB hereafter - and this chapter seeks to answer a few key questions about it, using psychological theories and experimental research:


 * What is CB and is it intentional?
 * What are some of CB's underlying mechanisms?
 * What causes CB to persist?
 * Are their ways we accidentally accentuate CB?
 * And are there practices which minimise the effects of CB?

Throughout this chapter, boxes like these will discuss and

explain little scenarios from daily life which exemplify the

confirmation bias phenomena described.

What is CB, and is it intentional?
CB is generally accepted as an unwitting selection and manipulation of information as it is collected: it does not typically comprise the behaviour of lawyers or debaters, where their goal from the outset is to artificially select information in order to make the strongest case for their side. In these situations, CB compounds a person’s capacity to confirm their prior bias, but that does not define CB per se. What defines CB is that we have a general tendency – a bias – to confirm information we have previously collected, even when we do not intend to (Nickerson, 1988).

CB becomes increasingly common when people or problems pertaining to us are involved in some way, because our family, friends, work or our beliefs can be affected by the conclusions the facts produce. Once we have made our first conclusion, then even if we intended to make that conclusion as objectively as possible, we are no longer able to behave as objectively from that point onward (Nickerson, 1998).

Stages where CB can occur:

·        You might start overconfident in your original belief, and you will likely remain confident in it with new information

·        When parsing information, you may see what you expect or know to look for easier than less well-known details

·        In searching for evidence, your biases may favour testing for hypothesis-supportive data, for confirmation purposes, instead of, or by avoiding disconfirming tests

·        Your interpretation of evidence may be based on previous conclusions, such that congruent information is considered trustworthy and incongruent info is unreliable

·        In finding new information, including incongruent arguments, you may not adjust your beliefs the appropriate amount

·        Even if you start to lose trust in your original position, you may not be able to think of any feasible replacements to it (Busemeyer, Medin & Hastie, 1995).

EXAMPLE BOX…

…BOX END

In research that tested the presence of CB, participants claimed that they would not be affected by their previous knowledge when inspecting new information, and also that they shouldn’t be affected by it: but from the results of those studies, it is evident that they were wrong on both accounts – people are affected by their previous knowledge, and there is now a wealth of evidence to suggest that they will definitely should be (Busemeyer, Medin & Hastie, 1995).

Tables
Table 1.

This is an Example of a Table with an APA Style Caption

The benefit of the doubt
One domain which seems to immediately hint that CB appears when it should not, is in instances of ambiguity: if one claimant describes factors they identify with, and one’s that their competitors do, then if the factors are hard to identify or measure, such as bias, scientific support and quality, a person will give the claimant the benefit of the doubt – and accept the given information as accurate. People seem to believe a hypothesis if the evidence is vague, rather than ignore or disbelieve it (Busemeyer, Medin & Hastie, 1995).

A diagnosing strategy
Some research has attempted to decipher where in our logical processes we have picked up this generalised bias for confirmation, and one particular experiment is thought to have identified one of the basic facets CB grows out from: the ‘Wason four-card selection task’.

When people could choose between two of four cards to prove whether a rule between all four of them is true, perhaps that every card with an odd number has a vowel, or inversely every even number has a consonant, people will choose turn a card that will confirm the given statement, rather than prove that no instance disconfirms it (Nickerson, 1988).

The ideal logical strategy, statistically or mathematically speaking, is to disconfirming alternatives based on Bayesian-like logic, where you can be sure a statement is true because you know that what would make it untrue has not been found to exist – in the scientific method, this is called falsifiability. Based on discussion in the literature about Wason’s selection task and his other experiments, humans have been showed to follow a diagnosing strategy instead – not to look for exceptions, but show that at least what is expected is true, and no more than that (Busemeyer, Medin & Hastie, 1995)....

Hypothesis preservation
By extension of granting the benefit of the doubt to vague or contentious situations, it has been found that people give confirmatory information more weight than disconfirming data: in experiments where participants received mixed evidence for capital punishment, those that were for, became more supportive than when they began, and those that were against, became more critical than when they began, seemingly discounting the contradictory information, and using the congruent information to bolster their opinions. When told that some of the presented information is subject to error, participants will select the information that is incongruent to be the supposedly erroneous data (Busemeyer, Medin & Hastie, 1995).

Is CB completely accidental?
...

Are there practices which minimise CB?
...

Test yourself on biases
Here are some example quiz questions - choose the correct answers and click "Submit": {Approximately how many neurons are in the human brain? - 1,000,000 (1 million) - 10,000,000 (10 million) - 100,000,000 (100 million) - 1,000,000,000 (1 billion) + 10,000,000,000 (10 billion)
 * type=""}

{A typical neuron fires ________ per second. - 1 to 4 + 5 to 49 - 50 to 99 - 100 to 199 - 200 to 499
 * type=""}

For more information, see Help:Quiz.

Conclusion

 * What are the take-home messages?