We gratefully acknowledge the National Science Foundation for their support on our project 1801539, “SaTC: CORE: Medium: Situation-Aware Identification and Rectification of Regrettable Privacy Decisions”.
Project Summary
People today are faced with many privacy decisions in their daily interactions with mobile devices. In the past decade, researchers have studied the design of many tools and mechanisms, such as privacy nudges, that aim to help individuals make better privacy decisions. But just like decision support tools in other domains, these tools cannot make users perfect decision-makers. Users still make mistakes and regret their privacy decisions later. This project casts a fresh perspective on Privacy-by-Redesign by helping users revisit and rectify past privacy decisions that they may regret. In order to avoid annoying users through repetitive alerts, a focus of the project is to identify which past privacy decisions most likely trigger regrets, and to ask users to revisit only those decisions. This project has high societal importance, given that more than 75% of Americans own a smartphone today and need to make frequent privacy decisions. The broader impacts of the project also reach technology developers, policy makers, and consumers by connecting the social analysis of privacy behaviors with the technical design of privacy tools.
This project is rooted in integrating substantive bodies of multidisciplinary knowledge to address the acute challenges of mobile privacy. It develops a theory on how three types of factors, cognitive appraisal, affective states, and environmental cues undercut high-effort decision making and move people toward low-effort information processing, which ultimately leads to regrettable privacy decisions. For the social analysis of privacy behaviors, this project employs a novel combination of experience sampling method and factorial vignette studies to empirically validate the theoretical framework. For the technical design of privacy tools, the project develops an expert-augmented prediction model that infers from data collectible by a mobile operating system the influential factors of cognitive appraisal, affective states, and environmental cues, so as to predict the quality of a privacy decision. The long-term vision of this project is to enable technological designs that help bridge the discrepancies between users’ privacy decisions and their perceptions, especially in the context of a mobile system.