Judgments under uncertainty: evaluations of univocal, ambiguous and conflicting probability information

People often have to make decisions about problems that not only involve a probability of a risk, but also an uncertainty around this probability. In this paper, uncertainty that arises through a lack of knowledge (i.e. ambiguity) is distinguished from uncertainty that results from conflicting knowledge on the hazard’s probability of occurring. Previous studies have shown that people are ambiguity averse when comparing a univocal risky option with an ambiguous option. Moreover, they appeared to be conflict averse when comparing an ambiguous option to a conflicting option. Relatively little is known about how people evaluate univocal, ambiguous and conflicting information about the riskiness or effectiveness of a problem. Two studies were therefore conducted to investigate people’s choices and their perception of univocal, ambiguous and conflicting probability information among various scenarios. Respondents mainly appeared conflict averse and were only ambiguity averse when the information included both numerical and verbal descriptions of the uncertainty. Moreover, they did not distinguish between ambiguous and conflicting information in their judgements. I suggest that it is easier for people to distinguish conflicting information from univocal information than from ambiguous information, because conflicting and univocal information differ on two dimensions, certainty and credibility, whereas ambiguous and univocal information only differ on the certainty dimension. Future research and communicators of conflicting uncertainty information should therefore focus on the cause of the conflict and on the credibility of the information sources.