Teaching Spell Checkers to Teach: Pedagogical Program Synthesis for Interactive Learning

TL;DR

SPIRE introduces Spelling Inquiry Engine, a system that brings Structured Word Inquiry-based pedagogy into everyday writing by synthesizing short, context-aware instructional programs from misspellings. It encodes SLP instructional moves in a domain-specific language, uses a multi-module pipeline to diagnose errors, synthesize adaptive intervention traces, and render them inline during writing. Expert and preliminary learner evaluations show strong pedagogical validity and promising usability, indicating potential for integration into classroom and therapy workflows. The work demonstrates how Pedagogical Program Synthesis can scale expert, inquiry-driven spelling instruction beyond therapist offices, with implications for personalized literacy support and future AI-assisted pedagogy.

Abstract

Spelling taught through memorization often fails many learners, particularly children with language-based learning disorders who struggle with the phonological skills necessary to spell words accurately. Educators such as speech-language pathologists (SLPs) address this instructional gap by using an inquiry-based approach to teach spelling that targets the phonology, morphology, meaning, and etymology of words. Yet, these strategies rarely appear in everyday writing tools, which simply detect and autocorrect errors. We introduce SPIRE (Spelling Inquiry Engine), a spell check system that brings this inquiry-based pedagogy into the act of composition. SPIRE implements Pedagogical Program Synthesis, a novel approach for operationalizing the inherently dynamic pedagogy of spelling instruction. SPIRE represents SLP instructional moves in a domain-specific language, synthesizes tailored programs in real-time from learner errors, and renders them as interactive interfaces for inquiry-based interventions. With SPIRE, spelling errors become opportunities to explore word meanings, word structures, morphological families, word origins, and grapheme-phoneme correspondences, supporting metalinguistic reasoning alongside correction. Evaluation with SLPs and learners shows alignment with professional practice and potential for integration into writing workflows.

Figures (5)

• Figure 1: This figure demonstrates the usage scenario. We showcase the learner going through three interventions for the following inquiries: (a) "is the sense/meaning of the word known?", (b) "is the base of the word known?" and (c) "are the morphological relatives known?"
• Figure 2: Overview of the System Architecture of SPIRE. The number labels on the edges mark the timestep at which each module is activated.
• Figure 3: This figure showcases an example for the pipeline, providing each step's output.
• Figure 4: This figure showcases some example hypothesis testing in the interface (right) and the program that guides that interface (left).
• Figure 5: System Usability Survey responses from 7 children. On the left column, higher is better, while on the right column, lower is better.