The importance of making assumptions : why confirmation is not necessarily a bias.

The idea of a `confirmation bias - that people reason in ways that lead them to irrationally confirm whatever it is they already believe - is one of the most widely accepted psychological findings. In this thesis, I argue that the evidence for confirmation bias is much weaker than is often supposed, and that this raises some challenging questions about what it means for beliefs to influence reasoning in an irrational way. I suggest that the literature on confirmation bias faces three challenges. First, the term `confirmation bias has been used to refer to multiple different things by different people, creating a literature of disparate findings that are not well unified. Second, many of the tendencies commonly referred to as confirmation bias are either not robust, or do not lead to confirmation under all circumstances. Third, most findings of `confirmation bias do not do enough to demonstrate a genuine bias or irrationality, and do not adequately address the complex associated normative issues. I discuss the link between confirmation bias and the broader concept of `open-mindedness', suggesting that existing research on both these topics fails to recognise the necessity and benefits of making assumptions as we navigate our lives. Instead of making claims like\people fall prey to a confirmation bias" or \people should be more open-minded", I suggest that research should focus on understanding how people navigate trade-offs - between the benefits of having firm beliefs and of making assumptions, and the benefits of being `detached' from prior beliefs and able to change one's mind