Skip to content

Four pedagogical principles for AI companion design derived from the HELF

Design · All designs
Evidence · 1 claim (1 for) · 1 study (1 theoretical), q2 · 0 of 1 report an effect size · 1 claim rests on one study

Description

Building on the Higher Education Learning Framework, the paper derives four principles for AI companion design: "deep and interactive learning, which prompts learners to actively process and connect new knowledge through varied cognitive tasks; guided scaffolding, which delivers adaptive guidance calibrated to the learner's Zone of Proximal Development"; learning to learn and higher-order learning, fostering metacognition and autonomy; and contextual learning, situating learning in authentic contexts for transfer and disciplinary identity. Each principle carries concrete companion strategies such as prompting generation and retrieval rather than provision.

Design Implications

Context

Requirements

  • Companions must make the student do the thinking rather than present information passively

Constraints

  • The paper notes current prompting approaches casting the model as a Socratic tutor often remain limited in scope and provide only surface-level support

Target Learners

  • higher education students

Learning Goals

  • deep durable understanding
  • metacognitive skills
  • transferable knowledge

Claims

Examples

-

Key Sources

  • Khosravi, H., Gašević, D., Sadiq, S., Yan, L., Lodge, J. M., Tangen, J. M., Denny, P., DiCerbo, K., Buckingham Shum, S., & Baker, R. S. (2026). Building AI Companions that Prioritise Learning over Performance. https://arxiv.org/abs/2605.04816. doi:10.48550/arxiv.2605.04816
Page metadata
Field Value
type design
id four-pedagogical-principles-ai-companion-design
title Four pedagogical principles for AI companion design derived from the HELF
description Building on the Higher Education Learning Framework, the paper derives four principles for AI companion design: "deep and interactive learning, which prompts learners to actively process and connect new knowledge thro...
status review
generated by process:settle-candidates on 2026-10-12
trust unverified — no one has confirmed this page's content
sources 1 — khosravi-2026

Frontmatter is the page's machine-readable record — see the Schema & Guide.