Researchers debate the role of reasons and argument in the conduct of aﬀairs. For some, decisions are “often reached by focusing on reasons that justify the selection of one option over another” (Shaﬁr, Simonson, & Tversky, 1993, p. 34). For others, decisions, opinions and attitudes are products of other, perhaps implicit, factors. For instance, although opinions can be retrieved from memory (Eagly & Chaiken, 1993), they may also be constructed on the spot (Hastie & Park, 1986; Iyengar & Ottati, 1996) and so may be aﬀected by information that is temporarily accessible at the time of judgement (Wilson, Lindsey, & Schooler, 2000). In support of this claim, Wänke, Bless, and Biller (1996) asked individuals to write down three or seven arguments either in favour of public transport or against public transport before reporting their attitudes. Participants reported more favourable attitudes when they experienced an easy time in generating positive arguments or a diﬃcult time in generating negative arguments. Haddock (2002) extended this line of research. In his study, individuals wrote down either two or ﬁve positive or negative characteristics of Tony Blair (the British Prime Minister at the time). The group was split into those with an interest in British politics and those with no strong interest. For those interested in British politics there was no eﬀect on opinion of the number and type of attributes generated. In contrast, for those with no interest, more favourable opinions followed the generation of positive compared to negative attributes. In addition, attitudes were slightly more positive for those generating two positive attributes compared to ﬁve positive attributes and for those generating ﬁve negative attributes compared to two negative attributes. In other words, for those with little interest in an issue, it is not only the nature of the generated arguments but also the ease of their generation that is important. But do these data undermine the importance of reasons and argument in judgement and opinion? No. Ease of generation may be a good proxy for the strength of arguments. Expressed diﬀerently, diﬃculty in generation may be a good indicator of the absence of strong arguments and might count as a “fast and frugal heuristic” (Goldstein & Gigerenzer, 2002). But is there any need to
postulate a proxy for argument strength? No. Opinion might be directly sensitive to the strength of produced arguments. The requirement to generate ﬁve or seven arguments on an issue may elicit weak arguments that deplete the average strength of arguments for the position. If this suggestion proves correct, no proxy, nor any additional, implicit process is needed to explain the results. Opinion is sensitive to the strength of arguments or reasons. Alternatively, both factors may be important and tap diﬀerent aspects of the process of generating an opinion.