Khosravi, Buckingham Shum, Chen, Conati, Tsai, Kay, Knight, Martinez-Maldonado, Sadiq & Gaลกeviฤ (2022) โ Computers and Education: Artificial Intelligence, 3, 100074.
๐ DOI: 10.1016/j.caeai.2022.100074 | PDF
Core Contribution
This paper introduces XAI-ED, a framework for explainable AI that is purpose-built for education. It argues that while XAI in education shares common ground with broader XAI (accountability, fairness, privacy), education has distinctive needs: learning data is noisy at many levels, explanations can directly support metacognition and self-regulated learning, and different stakeholders need fundamentally different kinds of explanations.
The XAI-ED Framework โ Six Aspects
1. Stakeholders
XAI in education must serve multiple audiences โ learners, teachers, administrators, parents, researchers, developers, and policymakers โ each with different explainability requirements. A learner needs to understand why a hint was given; a teacher needs to understand which students are at risk and why; a developer needs to debug model behavior. This connects to teacher-role and student-experience research on multi-stakeholder design.2. Benefits
Explanations serve multiple purposes in educational contexts:- Trust โ building confidence in AI-driven recommendations and assessments
- Fairness โ detecting and mitigating algorithmic bias-mitigation
- Debugging โ helping developers and researchers identify model flaws
- Usability โ making AI interfaces more transparent and actionable
- Learning โ explanations as pedagogical tools that support metacognition
- Regulatory compliance โ GDPR right to explanation and similar mandates (privacy, regulation)
- Adoption โ reducing resistance to AI tools among educators
3. Approaches for Presenting Explanations
The paper catalogs multiple explanation modalities: visual (heatmaps, decision trees), textual (natural language), example-based (counterfactuals, nearest neighbors), feature importance rankings, rule extraction, and model simplification. The key insight is that the optimal approach depends on the stakeholder and the pedagogical context โ an insight directly relevant to scaffolding design and intelligent-tutoring interface research.4. Classes of AI Models
The framework maps explanation approaches to model types:- White-box (decision trees, linear models, rule-based systems) โ inherently interpretable
- Black-box (neural networks, ensemble methods) โ require post-hoc explanation methods
- Glass-box โ newer approaches that balance accuracy with transparency
This taxonomy connects to the adaptive-learning literature's ongoing tension between model complexity and interpretability.
5. Human-Centred Design
Explanations are not purely technical artifacts โ they are communication acts. The interface must be designed for the specific stakeholder's cognitive needs, not just optimized for technical accuracy. This aligns with ai-literacy research showing that technical transparency without pedagogical framing often fails to support actual understanding.6. Potential Pitfalls
The framework identifies risks specific to educational XAI:- Explanation overload โ too much information overwhelms users, undermining the benefit
- Misleading explanations โ post-hoc explanations may not reflect actual model reasoning
- Confirmation bias โ users may selectively attend to explanations that confirm existing beliefs
- Over-trust โ explanations can create false confidence in flawed systems, connecting to over-reliance
- Gaming the system โ students may exploit explanations to circumvent learning, a known risk in intelligent-tutoring
Four Case Studies
OnTask โ Instructor-Facing Learning Analytics
A learning analytics platform that provides instructors with interpretable student risk indicators. XAI-ED applied to surface the rules and features driving each risk flag, enabling instructors to make informed interventions. Demonstrates the teacher-role shift toward data-informed decision-making.OATutor โ Adaptive Tutoring System
An open-source ITS where XAI explains mastery predictions and hint selections to both students and instructors. Connects to intelligent-tutoring and adaptive-learning research on transparency in automated instruction.SRES โ Student-Facing Recommender
A learning resource recommender that explains why specific resources are suggested based on the learner's knowledge state and goals. Supports self-regulated-learning by making the recommendation logic visible and actionable.CLA (Connected Learning Analytics) โ Multimodal Collaboration
Analytics for collaborative learning environments that surface group dynamics and individual contributions. XAI helps students and instructors understand team interaction patterns โ connecting to collaborative-learning and learning-analytics research.Significance for AIED
The XAI-ED framework has become a foundational reference in AIED, cited across the ai-k12-evidence-base, ai-tutor-safety-harms, and ai-literacy literatures. It bridges the gap between technical XAI research (which often ignores pedagogical context) and educational practice (which often treats AI as a black box). The framework's emphasis on distinctive educational needs anticipates later work on pedagogical-safety and human-in-the-loop design.
Related Pages
Citation
APA: Khosravi, H., Buckingham Shum, S., Chen, G., Conati, C., Tsai, Y.-S., Kay, J., Knight, S., Martinez-Maldonado, R., Sadiq, S., & Gaลกeviฤ, D. (2022). Explainable Artificial Intelligence in education. Computers and Education: Artificial Intelligence, 3, 100074. https://doi.org/10.1016/j.caeai.2022.100074