Designed, not delegated: Human-centred prompt engineering as a promising approach to AI-supported learning design

Generative AI can improve students’ immediate performance while reducing the cognitive and metacognitive work required for durable learning. Although recent work on AI learning companions offers a promising response, these systems remain largely beyond the resources available to individual educators. This paper argues that educators can adopt a more modest form of human-centred AI design by encoding pedagogical content knowledge (PCK) into prompt design and sustaining student partnership throughout iterative development. Drawing on human-centred design, PCK, technological pedagogical content knowledge, and Students as Partners, the paper develops a conceptual account of how pedagogical, content, technical, and contextual knowledge can shape what a large language model (LLM) provides, withholds, and asks students to do. An illustrative case of an AI peer tutor for identifying and evaluating peer-reviewed sources demonstrates how this approach was enacted through five major prompt iterations and progressively deeper forms of student involvement. The case shows how educator knowledge of recurring student difficulties can be translated into pedagogical pathways, while student partnership can contribute contextual knowledge and complementary expertise that further develops the educator’s technological pedagogical content knowledge (TPACK). Evaluation indicated substantial improvement in peer-tutoring behaviour and favourable student perceptions, while highlighting the need for continued iteration and research examining effects on student learning.

Authors

Institutions

Publication Details

Journal
ASCILITE Publications
Published
2026-11-17
DOI
https://doi.org/10.65106/apubs.2026.3249
Primary Topic
Artificial Intelligence in Education
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Designed, not delegated: Human-centred prompt engineering as a promising approach to AI-supported learning design

Rae Mair
5 links
ASCILITE Publications
Artificial Intelligence in Education
article

Designed, not delegated: Human-centred prompt engineering as a promising approach to AI-supported learning design

Rae Mair
article en

Abstract

Generative AI can improve students’ immediate performance while reducing the cognitive and metacognitive work required for durable learning. Although recent work on AI learning companions offers a promising response, these systems remain largely beyond the resources available to individual educators. This paper argues that educators can adopt a more modest form of human-centred AI design by encoding pedagogical content knowledge (PCK) into prompt design and sustaining student partnership throughout iterative development. Drawing on human-centred design, PCK, technological pedagogical content knowledge, and Students as Partners, the paper develops a conceptual account of how pedagogical, content, technical, and contextual knowledge can shape what a large language model (LLM) provides, withholds, and asks students to do. An illustrative case of an AI peer tutor for identifying and evaluating peer-reviewed sources demonstrates how this approach was enacted through five major prompt iterations and progressively deeper forms of student involvement. The case shows how educator knowledge of recurring student difficulties can be translated into pedagogical pathways, while student partnership can contribute contextual knowledge and complementary expertise that further develops the educator’s technological pedagogical content knowledge (TPACK). Evaluation indicated substantial improvement in peer-tutoring behaviour and favourable student perceptions, while highlighting the need for continued iteration and research examining effects on student learning.

ASCILITE Publications
University of Central Florida (US)
Quality education
Artificial Intelligence in Education
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon