Pursue theory-inscribed customization that embeds pedagogical principles into AI system architecture¶

Strategy · All strategies
Evidence · 1 claim (1 against) · 1 study (1 review), q2 · 0 of 1 report an effect size · 1 claim rests on one study

Description¶

The review recommends reconceptualizing customization as "the theoretically constrained specification of system behavior, with mediation, interaction, and development positioned as first-order design concerns" rather than interface-level control or performance tuning. It points to hybrid and integrated systems that embed pedagogical rules, domain representations, and adaptive control mechanisms, and urges researchers to track longitudinal writing cycles, analyze interactional data such as dialogue logs and revision histories, and foreground learner agency and transparency.

Design Implications¶

Context¶

Requirements¶

• Learning theories must inform system architecture, adaptation logic, and mediation policies rather than only task framing

Constraints¶

• The review notes substantial training data requirements constrain the feasibility of customizing AI systems to function as developmental co-participants

Target Learners¶

• Higher education L2 academic writing learners

Target Learning Goals¶

• Writing development through mediated, dialogic, and developmentally sequenced AI support

Claims¶

• Technically oriented theoretical rationales are operationalized more often than learning-centered theories in AI writing customization [-W]

Related Strategies¶

• Use the integrated technical–pedagogical limitation framework to inform teacher education, instructional design, and institutional AI policies in undergraduate EFL programs

Examples¶

-

Key Sources¶

• Luo, Y. (2026). Customizing AI for writing pedagogy: a systematic review of pedagogical goals, theoretical principles, and technical design. Education Tech Research Dev. https://doi.org/10.1007/s11423-026-10697-w