University of Twente Student Theses
Artificial Intelligence agents as our teammates? How accurate mental models can facilitate trust in technology
Weise, L.S. (2022) Artificial Intelligence agents as our teammates? How accurate mental models can facilitate trust in technology.
PDF
976kB |
Abstract: | In recent years, Artificial Intelligence (AI) has evolved to be a nearly ubiquitous concept of the modern world, as tool and even as team partner. For effective collaboration in a Human-Agent Team (HAT), means to facilitate and maintain trust in the robotic colleague must be found. This paper investigated the concept of mental models, cognitive representations of systems such as AI agents, as a predictor of trust in AI as well as effects of AI familiarity and anthropomorphic agent embodiment. It was expected that mental model congruence with the assigned AI agent impacts initial trust, with further influence of AI familiarity. In addition, anthropomorphic embodiment was expected to yield higher initial trust. A 4 (anthropomorphic, zoomorphic, mechanical, virtual) x 2 (initial trust and after a violation) online experiment including a written scenario and mental model sketch was used. Majority of the 80 participants were students at the University of Twente. No support for the mentioned effects were found. However, this study makes a promising advance into the assessment of mental models. |
Item Type: | Essay (Bachelor) |
Faculty: | BMS: Behavioural, Management and Social Sciences |
Subject: | 77 psychology |
Programme: | Psychology BSc (56604) |
Link to this item: | https://purl.utwente.nl/essays/91584 |
Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Repository Staff Only: item control page