Comparing two tiny giants or two huge dwarfs? Preference reversals owing to number size framing

Previous research has found that changes in irrelevant information, evaluation scales, or evaluation modes can lead to preference reversals. Drawing upon prospect theory, we introduced number size preference reversals, which occur under two conditions sharing identical information, evaluation mode, and evaluation scale. That is, for multiple options that have tradeoffs between attributes, an option is favored more when its superior attribute is framed with small numbers than when it is framed with large numbers. We tested the number size preference reversal in six experiments, demonstrating that it occurred in different background settings, remained stable when different evaluation scales were used, and was not contingent upon positive-negative attribute frames.