Exploring Trust in Human-Agent Collaboration

Abstract

Collaborative ensembles will increasingly involve agents like robots in the near future. A key part of collaboration is trust. While trust has been mainly studied between humans in CSCW, trust in human-agent research has been mostly studied in dyadic formations divorced from the broader context. This exploratory paper critically discusses previous work on trust across CSCW and HCI-related areas, taking into account recent practice approaches in CSCW and what they can contribute for understanding trust in human-agent collaboration. To make better sense of how trust emerges in collaborative ensembles with agents, we suggest that concepts that have been proposed in the field of human-agent interaction need to be further explored in real-life settings, while concepts embraced in CSCW can lead to a more thorough understanding of the situatedness and dynamics of trust going beyond the attributes of the agent itself.