Trust Impacts Driver Glance Strategy in Multitasking

Abstract

Growing evidence supports the idea that patterns of gaze are important to human-machine trust, as they are to human-to-human trust (LaFrance & Mayo, 1976; Kendon, 1967), and indeed potentially all primate social dynamics (Emery, 2000). A growing literature explores trust and gaze toward anthropomorphic robots (Mutlu et al., 2009; Stanton & Stevens, 2014; Van de Brule et al., 2014, Hancock et al., 2011). Less work has investigated far-more-common non-anthropomorphic systems, despite evidence suggesting that operators deploy the same trust patterns toward such interface that they might toward fellow humans (Nass, 1996; Fogg & Nass, 1997), and that they change patterns of visual allocation based upon that trust (Hergeth et al., 2016; Geitner at al., 2017). In critical operational settings, such as driving while multitasking, maximum safety and stability is associated with maximum visual attention devoted to the road (Hancock & Warm, 1989; Strayer, Drews & Johnston, 2003; Sawyer et al., 2014). Social gaze strategies deployed toward an interface suggest competition for these resources, and so applied consequences in terms of adopting appropriate information gathering strategies.