PhD project examining how students develop ideas and experience authorship when, during prewriting, they use a chatbot that only asks questions and does not generate text.
This project examines how students develop and evaluate ideas when, in the prewriting phase, they use a chatbot that only asks questions and never produces text. The aim is to understand how this interaction shapes idea development and what students consider appropriate or “text-worthy” when they later write independently.
By analyzing chatlogs, student texts, and interviews, it explores how this question-only interaction shapes writing processes in lower secondary school.
Three studies
The project consists of three studies: “When AI Only Asks” shows how students and a question-only chatbot co-create forward, redirective and stalled trajectories in prewriting. Through analysis of sensory, memory, reflective and follow-up questions, the article explains how these questions shape the flow and depth of emerging ideas.
“Is There a Text in This Chat? Tracing Idea Movement and Evaluative Sense in AI-Supported School Writing” examines how ideas from the dialogue change, disappear, or reappear in students’ final texts, and what these movements reveal about their evaluative sense of school writing.
A third article uses student interviews to explore how they experience working with a question-only chatbot.
The project analyses AI dialogues, student texts and interviews, and is relevant for schools, teacher education, researchers and technology developers. The goal of the project is to identify how AI can support early idea development in ways that preserve students’ ownership, voice and learning.
More information
Supervisors
Main supervisor: Marte Blikstad-Balas, Professor, Department of Teacher Education and School Research, University of Oslo.
