Selective access to decision support as a function of event uncertainty

The effective design of decision support systems requires an understanding of how system outputs affect users’ preferences and behavior. In an online experiment with 171 participants, we investigated how uncertainty influences users’ tendency to access additional information when using decision support (DS) with varying levels of reliability (high or low) that exhibit the same data in different ways. We compared preferences across three DS designs: binary independent output, likelihood independent output, and integrated likelihood output. We also examined the impact of DS displays and reliability on users’ performance and subjective trust. Users were more likely to seek additional information when presented with events incorporated with uncertain information. Moreover, the presentation of the information had a significant effect on performance and subjective trust.