A systematic review of risky-choice framing effects.

Classic decision theory requires that rational agents show description invariance: which description is chosen should not matter for judgments, preferences, or choices given the descriptions are co-extensive. Framing research has amply demonstrated a failure of description invariance by showing that the choice of the description has a systematic effect on judgments, preferences, and choices. Specifically, framing research has shown that linguistically different descriptions of seemingly equivalent options frequently lead to preference reversals. I summarize the research on framing in situations entailing risk. This includes the characterization of different research designs used, the size and robustness of the framing effects reported for those designs, and the theoretical accounts put forward to explain framing effects. The theoretical accounts are evaluated with respect to their merits, empirically and theoretically. I end by providing the implications of framing research. My central point is that the existence of framing effects points to the adaptiveness of the processes underlying human judgment and choice rather than simply showing human irrationality.