EUVIMEDEUVIMED entdecken
Systematische ÜbersichtMetaanalyseOpen Access · CC BY-SA 4.0

Artificial intelligence-supported motor skill performance in physical education and sport: A systematic review and meta-analysis informed by motor learning theory

Yulingga Nanda Hanief, Vera Septi Sistiasih, Ferdinando Cereda, Spyridon Plakias

Journal of Artificial Intelligence in Education & Learning Innovation · 2026

DOI: 10.56003/jaieli.v2i1.753 ↗

Kurz gefasst

Systematische Übersicht in der Zeitschrift Journal of Artificial Intelligence in Education & Learning Innovation, 2026. Verfasst von Yulingga Nanda Hanief und 3 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Artificial Intelligence (AI) has increasingly been integrated into physical education (PE) to support motor learning through technologies such as computer vision, motion tracking, intelligent tutoring systems, virtual reality, and generative AI. However, evidence regarding its effectiveness remains fragmented across different intervention types and learning contexts.

Objectives

This study aimed to evaluate the effects of AI-supported interventions on motor skill performance in physical education and sport settings and to synthesize complementary learning-related outcomes.

Methods

A systematic review and meta-analysis were conducted in accordance with PRISMA 2020. Scopus, PubMed, and ProQuest were searched through 22 June 2026. Eligible studies evaluated AI-supported, adaptive, or intelligent interventions in physical education, sport, or motor skill-learning contexts and reported learner-level motor performance; complementary learning-related outcomes were synthesized narratively. Risk of bias was assessed using RoB 2 and ROBINS-I, and standardized mean differences (SMDs) with 95% confidence intervals (CIs) were synthesized using a random-effects model.

Results

Fifteen studies were included in the qualitative synthesis; six contributed to the meta-analysis and nine were synthesized narratively. The pooled estimate favored the designated AI-supported experimental conditions over their comparators (SMD = 3.15, 95% CI 1.86–4.44; p < .001; I² = 98%). Given the small evidence base, very high heterogeneity, and variable risk of bias, the magnitude of this pooled effect is uncertain. Qualitative findings concerned short-term motor skill performance and complementary outcomes such as engagement, motivation, learning interest, and self-directed learning.

Conclusions

AI-supported interventions may improve short-term motor skill performance in some physical education and sport contexts and may support complementary learning-related outcomes. However, the evidence is preliminary and highly heterogeneous, and immediate post-intervention performance should not be interpreted as definitive evidence of durable motor learning, retention, or transfer. AI should be considered a complementary pedagogical tool rather than a replacement for teacher or coach expertise.

Zitieren

APA

Hanief, Y. N., Sistiasih, V. S., Cereda, F., & Plakias, S. (2026). Artificial intelligence-supported motor skill performance in physical education and sport: A systematic review and meta-analysis informed by motor learning theory. Journal of Artificial Intelligence in Education & Learning Innovation. https://doi.org/10.56003/jaieli.v2i1.753

Vancouver

Hanief YN, Sistiasih VS, Cereda F, Plakias S. Artificial intelligence-supported motor skill performance in physical education and sport: A systematic review and meta-analysis informed by motor learning theory. Journal of Artificial Intelligence in Education & Learning Innovation. 2026. doi:10.56003/jaieli.v2i1.753

BibTeX
@article{Hanief2026,
  author = {Hanief, Yulingga Nanda and Sistiasih, Vera Septi and Cereda, Ferdinando and Plakias, Spyridon},
  title = {Artificial intelligence-supported motor skill performance in physical education and sport: A systematic review and meta-analysis informed by motor learning theory},
  journal = {Journal of Artificial Intelligence in Education & Learning Innovation},
  year = {2026},
  doi = {10.56003/jaieli.v2i1.753}
}
Kostenlos starten →