TITLE:
The Peer in the Machine: Evaluating the Impact of a “Peer-Like” AI Persona on Writing Motivation in Low-Proficiency L2 Learners
AUTHORS:
Xiaoping Mai
KEYWORDS:
L2 Writing Motivation, AI Persona, Peer-Like Feedback, Low-Proficiency Learners
JOURNAL NAME:
Open Journal of Applied Sciences,
Vol.15 No.11,
November
20,
2025
ABSTRACT: Artificial intelligence (AI) writing tools are often framed as authoritative tutors, prioritizing grammatical accuracy. For low-proficiency L2 writers, who frequently experience high anxiety and demotivation, this approach may be counterproductive. Grounded in theories of motivation and sociocultural learning, this study investigates whether designing an AI with a “peer-like” persona can more effectively foster positive motivational outcomes. This study was conducted with 280 low-proficiency L2 English learners, randomly assigned to interact with either a “Peer-Like” AI, which provided encouraging and suggestive feedback, or a “Tutor-Like” AI, which delivered direct corrective feedback, over a six-week intervention. Quantitative results from pre/post questionnaires indicated that the Peer-Like AI group demonstrated a statistically significant greater increase in intrinsic motivation and a significantly greater reduction in writing anxiety compared to the Tutor-Like AI group. Qualitative analysis of stimulated recall interviews revealed that learners perceived the Peer-Like AI as less intimidating and more supportive, which encouraged risk-taking and sustained engagement with writing tasks. The findings conclude that the strategic design of an AI’s social persona is a critical factor in enhancing L2 writing motivation, thereby highlighting the need for pedagogical tools that are designed to support learners’ affective states.