Artificial Intelligence and the Future of Second Language Acquisition
No existing title bridges second language acquisition theory, AI and NLP technical literacy, and practitioner guidance in a single volume. Current books tend to be either edited collections with uneven quality, practitioner guides lacking theoretical depth, or computational linguistics texts inaccessible to language educators.
Intelligent Input fills this gap by grounding every AI application in established SLA frameworks -- Krashen's Monitor Model, Long's Interaction Hypothesis, Swain's Output Hypothesis, Sociocultural Theory, Usage-Based Approaches, and Complex Dynamic Systems Theory -- while providing the technical literacy language researchers need to critically evaluate and design studies around AI-mediated learning.
Written with the benefit of three years of empirical evidence since the generative AI inflection point, this book is neither techno-utopian nor Luddite, applying a critical lens informed by equity, access, linguistic diversity, and the limits of current AI systems.
Unlike edited volumes, every chapter builds on a unified argument. One analytical voice across fifteen chapters, with consistent theoretical framing and cross-references.
The author holds a doctorate in Educational Technology and active language learning credentials, combined with industry experience in AI-powered instructional design for federal and corporate clients.
Systematically covers input, interaction, output, feedback, pronunciation, vocabulary, grammar, assessment, identity, multilingualism, and research methodology.
Every chapter includes practical applications, evaluation rubrics, and "Try This Monday" boxes for classroom practitioners alongside rigorous theoretical analysis.
From foundational theory through skills-based applications to broader implications for identity, multilingualism, and the future of the field.
Why AI and SLA need each other. Traces the parallel histories of CALL and AI/NLP, establishing the book's central argument that AI fundamentally challenges and extends core SLA constructs.
A rigorous review of major SLA theories -- Krashen, Long, Swain, VanPatten, Schmidt, Vygotsky, DeKeyser, Ellis, Larsen-Freeman -- each examined for compatibility with AI-mediated learning environments.
NLP fundamentals for SLA researchers. Demystifies tokenization, embeddings, attention, and transformers at a conceptual level sufficient for critical evaluation and research design.
How AI transforms the provision of linguistic input. Evaluates adaptive readers, AI-generated graded texts, and text simplification against Krashen's i+1 and VanPatten's Input Processing.
Can AI chatbots trigger negotiation of meaning? Examines whether AI-human conversation fulfills the interactional conditions identified by Long, Pica, Varonis, and Gass.
AI tools and learner output through the lens of Swain's Output Hypothesis and Skill Acquisition Theory. Addresses the paradox of AI making output easier while SLA values productive struggle.
AI as the tireless interlocutor. Reviews the extensive CF literature and evaluates which feedback types -- recasts, prompts, explicit correction, metalinguistic explanation -- AI systems deliver effectively.
ASR-based pronunciation instruction through L2 phonology research. Addresses whose pronunciation norms AI systems encode and implications for World Englishes.
From flashcards to contextual intelligence. Traces the evolution from spaced repetition to AI-powered adaptive vocabulary engines, evaluated against depth of processing and the Involvement Load Hypothesis.
How AI mediates grammar across the implicit/explicit continuum. Investigates whether AI can deliver reactive, meaning-contingent focus on form or defaults to discrete-point teaching.
Construct validity, reliability, washback, and fairness in AI scoring. From TOEFL iBT automated scoring to emergent LLM-based assessment of complex language performances.
Social, cultural, and identity dimensions. Does conversing with AI constitute legitimate social interaction? What happens to learner identity when the social risk of error is removed?
Beyond the monolingual bias. Challenges English-centric AI orientations, examines translanguaging, and evaluates machine translation as pedagogy rather than threat.
Methodological guidance for investigating AI in SLA contexts. Addresses AI variability, measuring gains under personalization, ethics, and the reproducibility problem with commercial AI systems.
Agentic AI, embodied interaction in AR/VR, multimodal systems, and brain-computer interfaces -- examined through SLA theoretical lenses. Proposes a research agenda for the next decade.
Estimated publication: Fall 2028
Contact the Author