Dokument: Trust in Voice-Based Assistants: Developing and Testing a Model of Hybrid Trust Across Technical, Social, and Media Dimensions
| Titel: | Trust in Voice-Based Assistants: Developing and Testing a Model of Hybrid Trust Across Technical, Social, and Media Dimensions | |||||||
| URL für Lesezeichen: | https://docserv.uni-duesseldorf.de/servlets/DocumentServlet?id=74806 | |||||||
| URN (NBN): | urn:nbn:de:hbz:061-20261029-143913-5 | |||||||
| Kollektion: | Dissertationen | |||||||
| Sprache: | Englisch | |||||||
| Dokumententyp: | Wissenschaftliche Abschlussarbeiten » Dissertation | |||||||
| Medientyp: | Text | |||||||
| Autor: | Bergschneider-Frehmann, Katharina [Autor] | |||||||
| Dateien: |
| |||||||
| Beitragende: | Prof. Dr. Ziegele, Marc [Gutachter] Prof. Dr. Vowe, Gerhard [Gutachter] | |||||||
| Dokumententyp (erweitert): | Dissertation | |||||||
| Dewey Dezimal-Klassifikation: | 300 Sozialwissenschaften, Soziologie | |||||||
| Beschreibung: | Voice-based assistants (VBAs) are a popular form of conversational artificial in-telligence (AI) that interact verbally with users and exhibit anthropomorphic traits. Users rely on them for diverse service tasks as well as for information retrieval. Studying trust in VBAs is essential to understanding why users engage with them despite limited con-trol and notable risks related to data privacy, misinterpretations, and unverifiable infor-mation. However, existing research lacks consensus on how to define and operationalize trust in VBAs. Given their hybrid character as both technical tools and social interactors, studies have so far applied either technical or interpersonal trust frameworks. A third relevant role, reflecting their role as information intermediaries, has not yet been ad-dressed. This dissertation therefore investigates the central research question:
What is trust in voice-based assistants, and how does it shape user interactions? Following a literature-based analysis of the technical and social roles of VBAs, the third role of information intermediary is introduced, and a first quantitative content analysis (Paper 1) evaluates how VBAs performed in this regard during the Covid-19 pandemic. Based on this, a theoretical model of hybrid trust is developed by integrating technical, interpersonal, and media trust frameworks (Paper 3). After outlining method-ological considerations for research on Human-Machine Communication (Paper 2), the model is empirically tested and refined through three studies: a qualitative content anal-ysis of user comments (Paper 3), a qualitative interview study with observed interac-tions (Paper 4), and an experimental online survey using a simulated VBA interaction (Paper 5). The dissertation contributes to theory and practice by providing an empirical-ly enriched model of hybrid trust in VBAs, which can inform future research on conver-sational AI. It also offers implications for theoretical development, empirical measure-ment, the design of VBAs, and policymaking regarding AI regulation and education. | |||||||
| Lizenz: |  Dieses Werk ist lizenziert unter einer Creative Commons Namensnennung 4.0 International Lizenz | |||||||
| Fachbereich / Einrichtung: | Philosophische Fakultät » Sozialwissenschaftliches Institut | |||||||
| Dokument erstellt am: | 29.10.2026 | |||||||
| Dateien geändert am: | 29.10.2026 | |||||||
| Promotionsantrag am: | 01.07.2025 | |||||||
| Datum der Promotion: | 05.12.2025 |
