Exploring the mechanism of technology-enhanced formative assessment through students' behavioral and cognitive task engagement

Journal article › Research › Peer reviewed

Publication data


ByHannah Pünjer, Ronja Schiller, Lars Höft, Jennifer Meyer, Thorben Jansen
Original languageEnglish
Published inLearning and Individual Differences, 131, Article 102980
Pages11
Editor (Publisher)Elsevier
ISSN1041-6080, 1873-3425
DOI/Linkhttps://doi.org/10.1016/j.lindif.2026.102980 (Open Access)
Publication statusPublished – 10.2026
KeywordsArgumentation, Computer-assisted learning, formative assessment, Behavioral engagement

Due to advances in educational technology, students receive automated feedback on their writing, requiring understanding of its mechanism of effectiveness. Although it is assumed that behavioral and cognitive task engagement play a crucial role in the effect of feedback on writing outcomes, their impact remains partially understood. We conducted an experimental study, comparing two groups of secondary school students who completed a writing task and revised their drafts with or without feedback (automated feedback [n = 348] vs. no feedback [n = 227]). We assessed students' behavioral task engagement using a linguistic similarity measure (i.e., edit distance) and students' cognitive task engagement using self-reports. Our results indicate that behavioral task engagement mediated the effect of automated feedback on writing performance. We found no mediating effect of cognitive task engagement. The findings imply that in writing, performance gains due to automated feedback emerge when students show proactive behavior in text revision.

Educational Relevance

Understanding how automated feedback supports students' writing skills allows educators to optimize learning environments. We provide evidence that automated feedback supports writing performance by engaging students in revisions of their texts. For educators, this underscores the importance of feedback that invites action and supports students in making revisions.