Innovation in Language Learning
Who Did the Learning? Perceived Cognitive Responsibility in Generative AI-Assisted Language Learning
Sümeyra Akkaya, İnönü University (Turkey)
Emel Kulaksız, National Defence University, Air NCO Vocational School (Turkey)
Abstract
Generative artificial intelligence (GenAI) is increasingly integrated into language learning, yet the production of high-quality language outputs does not necessarily indicate that meaningful learning has occurred. This study investigates how university students perceive learning, cognitive responsibility, and cognitive engagement across different forms of GenAI use in language learning. A scenario-based quantitative research design will be employed with approximately 300 university students who have experience using generative AI for language-related tasks. Participants will evaluate 18 short scenarios representing three forms of GenAI use: AI-assisted learning, in which learners retain primary responsibility for thinking and production; AI-augmented learning, in which GenAI extends learners’ ideas and supports decision-making; and AI-substituted learning, in which substantial portions of the cognitive task are delegated to the AI system. Following each scenario, participants will rate perceived learning value, cognitive responsibility, cognitive engagement, appropriateness of AI use, and the need for output verification on a seven-point Likert scale. Data will be analyzed using descriptive statistics, reliability analysis, repeated-measures analysis of variance, correlation, and regression analysis. Particular attention will be given to the relationship between perceived learning and perceived cognitive responsibility in AI-substituted scenarios. The study aims to identify whether learners distinguish between producing a better language product and actually engaging in learning. By shifting attention from the amount of AI use to the distribution of cognitive responsibility between learner and AI, the study seeks to provide a conceptually focused perspective for designing responsible and learning-oriented GenAI integration in language education.
Keywords: generative AI; language learning; cognitive responsibility; cognitive engagement; learner autonomy
