Effects of Prompting Style and Editing on Psychological Ownership in Human–AI Co-creative Writing
Abstract
Human-AI co-creation has a long history, with creative domains ranging from videogame design to poetry writing. Recent advances in AI, namely the evolution of large language models (LLMs), have drastically increased the quality of output and accessibility of AI tools for co-creation of text to a large demographic. Our focus here is creative writing as a particular popular domain of studying human-AI co-creation on writing. In such co-creative settings, humans retain varying degrees of influence over the final text depending on how the AI is used and how the interaction is designed and unfolds. Psychological ownership refers to the subjective feeling that something is one’s own and has been theorized to emerge in part through perceived control, personal investment, and intimate knowledge of the object of ownership (Pierce et al., 2001). Recent work has shown that writing with AI can lower psychological ownership compared to writing without AI, but that increasing human input, e.g. through longer prompts, can partially restore this sense of ownership (Joshi & Vogel, 2025). Related findings further suggest that users’ perceived influence over generated text plays a central role in experienced ownership in AI-assisted writing (Draxler et al., 2024). The present study builds on this work by examining whether, and if so, how psychological ownership over AI-assisted stories varies as a function of prompting style and the ability to edit AI-generated text. By studying (1) single-prompt and iterative prompting approaches as well as (2) the (im-)possibility to edit the created text in a factorial design and comparing the resulting conditions to a no-AI baseline, this study investigates how different forms and degrees of human influence and control in human–AI co-creation shape experienced psychological ownership over creative outcomes.
Fields of Study & Subjects
Innovative Human-Technology Interaction
Digital Media and Visual Art
Social Robot Interaction and HRI
Publication details
- Publication date
- 1 January 2026
- Document type
- Preprint