The Effects of Transparency and Task Type on Trust, Stress, Quality of Work, and Co-Worker Preference During Human-Autonomous System Collaborative Work

摘要

In this paper, we examined how transparency (i.e., cues about autonomous system work process) and task type influence trustfulness, perceived quality of work, stress level, and co-worker preference during human-autonomous system interaction. In addition we examined how technology readiness index (TRI) and age moderate the effects. Participants were led to believe (even though the stimuli was fixed by the researchers) that they were working with a new fully autonomous system ('a roboworker') performing the same tasks (two routine tasks and two planning tasks) over the internet. User were either given no information about the work of the collaborative autonomous companion or cues about the work of roboworker was displayed in textual format on a web page before and in the middle of each task. Self-report trustfulness (self-trust and trust on roboworker), quality of work, stress level, and co-worker preference were collected after each task. The preliminary results showed that (when age or TRI were controlled for) feedback about the work of roboworker increased perceived quality of work and self-trust (for trust on roboworker there was no difference). In connection with the type of task, people preferred roboworker, and were less stressful, during routine as compared to planning tasks. In addition, the results were moderated by age and TRI. The results are useful and important in connection with designing human-robot interaction.