Influence of Device Performance and Agent Advice on User Trust and Behaviour in a Care-taking Scenario

Abstract

We present insights obtained from two user studies performed in the context of a web-based game set in a care-taking scenario — a retirement village, where elderly residents live in smart homes equipped with monitoring systems. These systems should raise alerts when adverse events happen, but they do not function perfectly (they may issue false alerts or miss true events). Players, who “work” in the village, perform a primary task whereby they must ensure the welfare of the residents by attending to adverse events in a timely manner, and a secondary routine task that demands their attention. In the first user study, we investigate the relationship between the performance of different monitoring systems, in terms of error type, and user behaviour and trust in these systems. In the second study, we examine the effect of the advice offered by an advisor agent on users’ behaviour. Our contributions are (1) the game itself, which supports experimentation with various trust-related factors, and a version of the game augmented with an advisor agent; (2) a methodology for calibrating the parameters of the game; (3) insights regarding the relationship between device accuracy and user behaviour and trust in automation; (4) findings about the training effect of an advisor agent; and (5) insights from predictive models about factors that influence trust and behaviour.