As machines' autonomy increases, their capacity to learn and adapt to humans in collaborative scenarios increases too. In particular, machines can use artificial trust (AT) to make decisions, such as task and role allocation/selection. However, the outcome of such decisions and the way these are communicated can affect the human's trust, which in turn affects how the human collaborates too. With the goal of maintaining mutual appropriate trust between the human and the machine in mind, we reflect on the requirements for having an AT-based decision-making model on an artificial teammate. Furthermore, we propose a user study to investigate the role of task-based willingness (e.g. human preferences on tasks) and its communication in AT-based decision-making.@en

The introduction of artificial teammates in the form of autonomous social robots, with fewer social abilities compared to humans, presents new challenges for human–robot team dynamics. A key characteristic of high performing human-only teams is their ability to establish, develop, and calibrate trust over long periods of time, making the establishment of longitudinal human–robot team trust calibration a crucial part of these challenges. This paper presents a novel integrative model that takes a longitudinal perspective on trust development and calibration in human–robot teams. A key new proposed factor in this model is the introduction of the concept relationship equity. Relationship equity is an emotional resource that predicts the degree of goodwill between two actors. Relationship equity can help predict the future health of a long-term relationship. Our model is descriptive of current trust dynamics, predictive of the impact on trust of interactions within a human–robot team, and prescriptive with respect to the types of interventions and transparency methods promoting trust calibration. We describe the interplay between team trust dynamics and the establishment of work agreements that guide and improve human–robot collaboration. Furthermore, we introduce methods for dampening (reducing overtrust) and repairing (reducing undertrust) mis-calibrated trust between team members as well as methods for transparency and explanation. We conclude with a description of the implications of our model and a research agenda to jump-start a new comprehensive research program in this area.

@en

This paper is a call to the field of human-robot interaction to focus research efforts on the development and repair of trust within human-robot teams. To guide this effort, this paper describes the initial development of a framework for trust development in human-robot interaction research with a specific focus on trust repair. This framework identifies several unique trust-dyads within an example domain of Urban Search and Rescue Operations (USAR) that are suitable relationships for study. We conclude with several areas of research that should be addressed under a trust repair framework including trust measurement, model development and validation, mutual dynamic trust calibration, and long term trust development.@en