Perceived differences in tradeoff difficulty

Many theories of human judgment and decision making assume that people make tradeoffs. These theories (e.g., MAUT) are generally good descriptive models of subjects\u27 behavior. However, they fail to make certain psychological distinctions. In particular, they do not explain why subjects perceive some attributes as more difficult to trade off than others. We propose that subjects might experience such difficulty if they are: (1) unable to make a tradeoff; (2) unable to make a tradeoff reliably; (3) unwilling to make a tradeoff. We present 7 experiments to investigate which of these situations exist. Experiments 1-6 investigate one possible source of unreliability by asking whether attribute weights are influenced the range of the stimuli presented. Subjects made judgments when the range (i) should not affect the weight, (ii) may affect the weight, and (iii) should affect the weight. In Experiments 1-5 subjects\u27 responses adhered closely to the normative models. Experiment 6 required subjects to assign weights that were inversely proportional to the range of each attribute. Subjects showed some sensitivity to the range, but did not attend to it sufficiently. In general, however, weights are not influenced non-normatively by stimulus range, and that therefore this cannot account for perceived variations in tradeoff difficulty. Experiment 7 investigated directly which features of tradeoffs may cause perceptions of difficulty. Subjects rated tradeoffs for perceived difficulty, and on 12 other dimensions. Six features were found to be particularly useful in modeling variation in difficulty. Experiment 7 also revealed two different psychological states (indifference and indecision) that both produce the same behavior of indifference, as defined by the indifference relation. Failure to distinguish between these states may conceal the different psychological processes involved. We believe that these experiments will be of practical (as well as theoretical) import, by characterizing situations in which decision analytic tools utilizing tradeoffs may be inappropriate, and in alerting decision analysts to tradeoffs that may cause their clients difficulty