A framing effect of intertemporal and spatial choice.

Abstract

A given spatial distance can be measured using either a length or a time unit. A spatial-choice problem with given distances represented in a space frame and in a time frame is logically equivalent. Different representations of the same choice problem should yield the same preferences according to the invariance principle. To test invariance in the context of spatial choice, we used a constant velocity to construct six spatial (intertemporal) choice problems with single-placed (single-dated) outcomes in a space (time) frame and investigated whether invariance-violating behaviour could be detected under the two frames (Studies 1-3). If such behaviour existed, we then aimed to determine whether two models-the attribute-comparison model, which predicts a framing effect occurs if framing shifts people's judgement of the inequality relationship between the differences of two options in time/space dimension and that in outcome dimension, and the utility comparison model, which predicts that a framing effect occurs if framing shifts people's judgement of the inequality relationship between two options' overall utility-could account for it, and which of these two models is better supported by the data (Studies 2-3). The results indicate that a time-space framing effect existed, as people's preference orders were significantly changed by the different descriptions of spatial-choice problems, and this new time-space framing effect could be satisfactorily explained by the attribute-comparison rather than the utility comparison model. Our findings could support the creation of new forms of choice architecture that improve decisions about health, wealth, and happiness.