EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Too good to resist? Affective engagement and a competence–regulation gap predict self-reported over-reliance on generative AI among vocational foreign-language learners

Huimei Gao, Feng Yang, Qiang Liu, Zhuo Wang, Zishu Meng

Frontiers in Psychology · 20.11.2026 · 17

DOI: 10.3389/fpsyg.2026.1941348 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Frontiers in Psychology, online erschienen am 20.11.2026. Verfasst von Huimei Gao und 4 weiteren Autor:innen. Die Arbeit nennt 75 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Generative artificial intelligence (GenAI) has been adopted by students faster than nearly any earlier educational technology, yet over-reliance remains underexamined precisely among the learners inspiring that concern. This sequential explanatory mixed-methods study identified predictors of self-reported over-reliance among foreign-language majors ( n = 249) at a Chinese higher vocational college. Quantitative findings revealed two patterns. First, affective factors dominated: fear of missing out ( β = 0.50) and anthropomorphic attachment ( β = 0.27) were the strongest predictors, whereas neither self-regulated learning nor perceived AI literacy was related to over-reliance on its own ( r = −0.03 and 0.08, both ns). Because these two overlap substantially ( r = 0.67), the opposite-signed coefficients they take in the model cannot be interpreted in isolation. What mattered instead was their gap: the discrepancy between feeling AI-literate and reporting that one regulates one’s own learning predicted over-reliance ( β = 0.21), persisting even under controls for acquiescent responding. Second, person-centered and configurational analyses agreed: learners low in affective engagement reported the least over-reliance ( M = 3.19). A profile elevated on every indicator did not survive scrutiny for response bias, and fear of missing out combined with anthropomorphic attachment was consistently associated with over-reliance regardless of self-rated competence. Interviews with five learners suggested candidate explanations. Humanlike warmth lowered the psychological cost of delegation. Fear of falling behind, normalized by peers, gradually eroded personal standards through repeated choices. Perceived competence enabled over-reliance rather than restraining it, interrupted only by enacted self-regulation. The emotional relationship formed with the tool tracked over-reliance more closely than learners’ own competence assessment.

Zitieren

APA

Gao, H., Yang, F., Liu, Q., Wang, Z., & Meng, Z. (2026). Too good to resist? Affective engagement and a competence–regulation gap predict self-reported over-reliance on generative AI among vocational foreign-language learners. Frontiers in Psychology, 17. https://doi.org/10.3389/fpsyg.2026.1941348

Vancouver

Gao H, Yang F, Liu Q, Wang Z, Meng Z. Too good to resist? Affective engagement and a competence–regulation gap predict self-reported over-reliance on generative AI among vocational foreign-language learners. Frontiers in Psychology. 2026;17. doi:10.3389/fpsyg.2026.1941348

BibTeX
@article{Gao2026,
  author = {Gao, Huimei and Yang, Feng and Liu, Qiang and Wang, Zhuo and Meng, Zishu},
  title = {Too good to resist? Affective engagement and a competence–regulation gap predict self-reported over-reliance on generative AI among vocational foreign-language learners},
  journal = {Frontiers in Psychology},
  year = {2026},
  volume = {17},
  doi = {10.3389/fpsyg.2026.1941348}
}
Kostenlos starten →