Designing Language Activities for Young Learners Based on Human and AI-generated Stories

The article presents the results of a study conducted among novice teachers of English language, trained at two major universities in Bulgaria. The objectives of the study were: 1) to evaluate the extent to which students perceived stories, authored by human and artificial intelligence, as applicable in educational context; 2) to gain an insight into their competence to design story-based language learning activities for preschool and primary school children; and 3) to seek possible relations between their teaching competence, language proficiency levels, skills to use AI-tools and their perceptions of the usefulness and practical classroom application of the two types of narratives. Results showed that students viewed both types of texts as useful in foreign language teaching, but the displayed degree of usefulness was stronger for the AI-generated one. A tentative explanation of this finding could be the capacity of ChatGPT as a large language model to mimic human linguistic production fairly convincingly. Overall, students exhibited a reasonably high degree of understanding of language learning as a collaborative experience and showed abilities to apply interactive approaches and techniques in the activities they proposed. The higher the teaching competence of the participants, the stronger their awareness of the various interdisciplinary links a teacher can make in their language classes to other fields of knowledge. This is a promising indication for the capacity of teachers to capitalize on the affordances of story-based language learning and interweave cross-curriculum references in their teaching practices which brings added value to the learning process.