Repository logo
 

Theorising Interactions and Outcomes of Digital Voice Assistants: Experiences of Individuals with Visual Impairments

Files

Thesis(1.32 MB)

Date

2025

Authors

Supervisor

Waizenegger, Lena
Techatassanasoontorn, Angsana

Item type

Dissertation

Degree name

Doctor of Philosophy

Journal Title

Journal ISSN

Volume Title

Publisher

Auckland University of Technology

Abstract

General purpose conversational agents like digital voice assistants can offer significant benefits for people with visual impairments by helping them perform everyday tasks and overcome challenges in the physical and digital world. Their natural language capabilities, user-friendly interface, and universal availability make them a more integrative, friendly, and less stigmatising alternative to specialised technologies. Much of the previous research has focused on the use of digital voice assistants by sighted individuals, which may use these devices differently or for different purposes than people with visual impairments. Furthermore, existing studies tend to examine human-technology interactions from an Information Systems (IS) use perspective, often overlooking the agentic capabilities of digital voice assistants. Hence, this study aims to explore how individuals with visual impairments interact with voice assistants, the role of the technology’s agentic features in these interactions, and the specific outcomes that are generated. The study utilised a qualitative, interpretive approach, with data collected using semi-structured interviews involving participants with full blindness or low vision from Australia and New Zealand. The sample consisted of 21 interviews, which formed the basis of analysis using techniques of grounded theory coding, guided by an inductive-abductive approach. The inductive coding process revealed the various tasks that visually impaired individuals rely on voice assistants for, as well as the different outcomes generated from the interactions. These codes were then mapped against the theoretical frameworks of the sociotechnical perspective and Information Systems (IS) delegation to further analyse and theorise the interactions and their resulting outcomes. The findings reveal four delegation types that explain the dynamic interactions between visually impaired individuals and their digital voice assistants. Additionally, the findings also show the different instrumental and humanistic outcomes produced by these interactions. The main theoretical contributions of this study are the extension of the IS delegation framework, and that it integrates the IS delegation framework within the sociotechnical perspective that enables a deeper understanding of the micro-level dynamics of the interactions and outcomes between social (visually impaired individuals) and technical (digital voice assistants) systems. From a practical standpoint, the study offers valuable insights for designers and developers, suggesting ways to create more adaptable and context-sensitive voice assistants that better meet the needs of vulnerable populations like visually impaired individuals.

Description

Keywords

Source

DOI

Publisher's version

Rights statement

Permanent link

http://hdl.handle.net/10292/19407

Collections

Doctoral Theses