Exploring the Effects of Interruptions in Different Phases of Complex Decision-Making Tasks

Abstract

Objective The study aims to examine the effects of interruptions in major phases (i.e., problem-identification, alternative-development, and evaluation-and-selection) of complex decision-making tasks. Background The ability to make complex decisions is of increasing importance in workplaces. Complex decision-making involves a multistage process and is likely to be interrupted, given the ubiquitous prevalence of interruptions in workplaces today. Method Sixty participants were recruited for the experiment to complete a procurement task, which required them to define goals, search for alternatives, and consider multiple attributes of alternatives to make decisions. Participants in the three experimental conditions were interrupted to respond to messages during one of these three phases, whereas participants in the control condition were not interrupted. The impacts of interruptions on performance, mental workload, and emotional states were measured through a combination of behavioral, physiological, and subjective evaluations. Results Only participants who were interrupted in the evaluation-and-selection phase exhibited poorer task performance, despite their positive feelings toward interruptions and confidence. Participants who were interrupted in the problem-identification phase reported higher mental workload and more negative perceptions toward interruptions. Interruptions in the alternative-development phase led to more temporal changes in arousal and valence than interruptions in other phases. Conclusion Interruptions during the evaluation-and-selection phase undermine overall performance, and there is a discrepancy between behavioral outcomes and subjective perceptions of interruption effects. Application Interruptions should be avoided in the evaluation-and-selection phase in complex decision-making. This phase information can be either provided by users or inferred from coarse-grained interaction activities with decision-making information systems.