Task engagement through the lens of generative artificial intelligence: A thematic reanalysis of Lambert, Philp and Nakamura (2017)

This brief report presents initial results from ongoing research on using generative artificial intelligence (GenAI) in task engagement research. We report a thematic reanalysis of the dataset from Lambert, Philp and Nakamura (2017) using MAXQDA Tailwind and Guided AI Thematic Analysis (GAITA) (Nguyen-Trung, 2025). Lambert et al. (2017) operationalised task engagement in terms of frequencies of language forms serving specific functions in learners’ discourse. This discourse analytic framework was termed Engagement in Language Use (ELU) (Lambert & Aubrey, 2023). ELU was provided to the GenAI as an initial analytic template to guide but not limit the analysis. The GAITA procedure identified five themes in the dataset - disclosure, affirmation, critical thinking, self-assessment, and self-presentation - each with sub-clusters and codes. Results provide an initial dataset-specific coding template and analytic heuristic, generated through human-GenAI interaction, provisionally termed Self-Expression in Language Use (SELU). The results raise questions for task engagement theory and demonstrate affordances of GAITA for future task engagement research.