Investigating Status Quo Effects in Visualization-Aided Choice
Authors
Başak Oral (Utrecht University), Wouter Ubbink (Utrecht University), Evanthia Dimara (Utrecht University), Angelos Chatzimparmpas (Utrecht University)
Presentation
- Session
- Can We Trust This Chart? (Asking for a Friend)
- Time
- Wednesday, Nov 11, 08:09 – 08:18 (US/Eastern) · session 08:00 – 09:30
- Location
- Hall America south
Keywords
Visualization, decision making, decision support systems, cognitive biases, status quo, defaults, empirical studies.
Abstract
Status quo effects refer to an increased likelihood of selecting a default option over available alternatives. Although widely documented in behavioral research, their role in visualization-supported decision making remains unclear. We examine default adherence in probabilistic medical treatment decisions while systematically varying visualization format (quantile dotplots, density plots, probability bars, hypothetical outcome plots), default quality (good, average, or poor expected value), time pressure, and gain-loss framing. In a mixed-factorial experiment (N = 123), visualization was manipulated between participants and the remaining factors within participants. Default adherence was analyzed using mixed-effects logistic regression; decision accuracy (normalized expected value of the selected option) and response time were analyzed as complementary outcomes. Default adherence increased with default quality. Visualization format showed a marginal main effect on default adherence, though the observed differences appeared to be driven by specific visualization-default quality interactions under time pressure. Gain-loss framing showed no reliable association with default selection. Overall, our results suggest that choice-architecture factors such as defaults and time pressure can shape decision behavior in ways distinct from probabilistic decision accuracy. Supplementary material: https://osf.io/349jh/overview?view_only=89adaaaee52b4b4eb567dc5f05ffe058
For Practitioners
This paper will be of interest to practitioners who design, analyze, or rely on data-driven decision-support systems, particularly in settings where visualizations, defaults, recommendations, or time constraints shape decisions. Relevant audiences include visualization practitioners, data scientists, data analysts, UX/HCI practitioners, decision-support system designers, and professionals working with dashboards or structured decision interfaces. The findings may also be relevant to domain practitioners such as emergency managers, healthcare decision-support professionals, and other decision-makers who must interpret uncertain information and act under time pressure. More broadly, the paper is useful for practitioners interested in how cognitive biases, especially status quo effects, can influence data interpretation and decision-making beyond simple probabilistic accuracy.