Moderate evidence
More
406 students reported less autonomy writing with optional AI, a study found
Study title: Motivational organization in AI-supported writing: A self-determination perspective
- Evidence type
- Observational study
- Sample
- n = 406
- Sources
- 1 sourcePeer-reviewed
- Published
- 2 min read
- Truvace record
- TRV-2026-2892cite or check it
Study title: Motivational organization in AI-supported writing: A self-determination perspective
A study of 406 students in Grades 5–8 compared how they were motivated during comparable creative writing tasks, first without generative AI and later with optional AI access. Average levels of motivation stayed largely stable between the two occasions, but students' satisfaction of their need for autonomy declined when AI was available.
Read the study abstract
This study examined how students’ motivational regulation during creative writing was organized when generative artificial intelligence (AI) was unavailable or optionally available. Drawing on Self-Determination Theory (SDT), motivation was conceptualized as configurations of intrinsic motivation, identified regulation, introjected regulation, and external regulation. The sample comprised 406 students in Grades 5–8 who completed comparable creative writing tasks without AI (Time 1) and with optional AI access (T…
The problem
In a study of 406 students in Grades 5–8, autonomy satisfaction declined when optional generative AI was available for a creative writing task, compared with an earlier task done without AI.
In brief
The researchers used Self-Determination Theory and grouped students into motivational profiles built from intrinsic motivation, identified regulation, introjected regulation and external regulation. These profiles showed differences in how common they were across the two tasks that averages did not reveal.
How much students actually used AI did not significantly predict which profile they fell into. Students with higher satisfaction of their needs, especially autonomy and competence, were more likely to show more self-determined profiles, and end-of-term grades also distinguished profile membership on both occasions.
Main points
- The sample was 406 students in Grades 5–8 who completed comparable creative writing tasks without AI (Time 1) and with optional AI access (Time 2).
- Autonomy satisfaction declined at Time 2, when AI was optionally available.
- Mean levels of motivational regulation remained largely stable, while profile analyses showed differences in how common each motivational configuration was.
- Higher need satisfaction, particularly autonomy and competence, was associated with more self-determined motivational profiles.
- The degree of AI use did not significantly predict profile membership, which separates AI availability from how intensively it was used.
The full story
The study used a micro-longitudinal, person-centered design. Latent profile analyses identified motivational configurations at each occasion and tested psychological need satisfaction and academic achievement as predictors of which profile a student belonged to. The degree of AI use was examined at Time 2.
Cross-context analyses showed some correspondence between profiles alongside redistribution of students across them. End-of-term grades differentiated profile membership at both occasions.
The authors suggest AI-supported writing should be understood in relation to students' psychological needs, academic functioning and the demands around support, delegation, authorship and agency.
Sources
1 on file- Computers and Education Opendoi:10.1016/j.caeo.2026.100430
Talk it over
Average motivation stayed largely stable across the two writing tasks, yet autonomy satisfaction declined when AI was optional. Which of those two results would matter more to you if you were one of the 406 students?
406 students reported less autonomy writing with optional AI, a study foundhttps://truvace.com/item/motivational-organization-in-ai-supported-writing-a-self-determination-perspecti
Average motivation stayed largely stable across the two writing tasks, yet autonomy satisfaction declined when AI was optional. Which of those two results would matter more to you if you were one of the 406 students?
Answer in the threadThe researchers compared a first task without AI and a later task with optional AI in the same students. What could change between those two occasions besides AI being available, and how would that affect the result?
Answer in the threadHow much students actually used AI did not significantly predict their motivational profile. What would you want to learn about how the Grades 5-8 students used the tool to make sense of that?
Answer in the thread
If you have more time
Ask a teacher or student you know how it feels to be allowed, but not required, to use AI for a creative task. Which of the researchers' ideas about support, delegation, authorship and agency does their answer touch?
Answer in the threadEnd-of-term grades differed between motivational profiles on both occasions. What further evidence would show whether grades shape motivation, motivation shapes grades, or something else drives both?
Answer in the thread
ace

What readers say