Date
AI Overview
Why it matters
Information access systems optimize retrieval rather than helping users build effective search strategies or critical awareness. In complex or high-stakes contexts, skills such as evaluating sources and refining queries are crucial but rarely supported during interactions.
Study at a glance
- Question: How can information access systems be designed and evaluated as instructional environments that teach and scaffold user learning?
- Setting and data: Not stated in the paper
- Method: Conceptual analysis applying seven educational and behavioural frameworks across five interface case studies and two illustrative search tasks
- AI tools: ReAct
Key findings
- Traditional search features like query auto-completion and source labels influence behavior subtly through nudging but offer minimal support for deeper reflection, conceptual understanding, or strategy transfer.
- Conversational and agentic generative AI systems align more strongly with scaffolding, cognitive apprenticeship, and self-regulated learning by clarifying intent, preserving conversational memory, and exposing expert reasoning paths.
- Current generative and agentic systems still fail to fully realize instructional potential because they rarely diagnose user knowledge, adapt support to learning progress, or fade assistance over time.
Summary drafted with AI from the full paper and may contain errors. See the full paper for details.
Information access systems such as search engines and generative AI are central to how people seek, evaluate, and interpret information. Yet most systems are designed to optimise retrieval rather than to help users develop better search strategies or critical awareness. This paper introduces a pedagogical perspective on information access, conceptualising search and conversational systems as instructive interfaces that can teach, guide, and scaffold users' learning. We draw on seven didactic frameworks from education and behavioural science to analyse how existing and emerging system features, including query suggestions, source labels, and conversational or agentic AI, support or limit user learning. Using two illustrative search tasks, we demonstrate how different design choices promote skills such as critical evaluation, metacognitive reflection, and strategy transfer. The paper contributes a conceptual lens for evaluating the instructional value of information access systems and outlines design implications for technologies that foster more effective, reflective, and resilient information seekers.
Research Categories
What is the application?
Who is the user?
Who age?
Why use AI?
Study design
Methods
AI tools
Publication Source
Publisher
arXiv
