Making sense of surprise: An investigation of the factors influencing surprise judgments.

Surprise is often defined in terms of disconfirmed expectations, whereby the surprisingness of an event is thought to be dependent on the degree to which it contrasts with a more likely, or expected, outcome. The authors investigated the alternative hypothesis that surprise is more accurately modeled as a manifestation of an ongoing sense-making process. In a series of experiments, participants were given a number of scenarios and rated surprise and probability for various hypothetical outcomes that either confirmed or disconfirmed an expectation. Experiment 1 demonstrated that representational specificity influences the relationship that holds between surprise and probability ratings. Experiment 2 demonstrated that the inclusion of an enabling event lowers surprise ratings for disconfirming outcomes. Experiment 3 explored the reason for this effect, revealing that enabling events lower surprise by reducing uncertainty, thus enhancing ease of integration. Experiment 4 evaluated the contrast hypothesis directly, showing that differences in contrast are not correlated with differences in surprise. These results provide converging support for the view that the level of surprise experienced for an event is related to the difficulty of integrating that event with an existing representation. (PsycINFO Database Record (c) 2010 APA, all rights reserved).