EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

AI-generated feedback on student solutions to graphical problems in kinematics and dynamics

John Tipaz, Kevin Ortega-Chávez, Johnny Hidalgo, Juan López-Ruano, Darwin Casaliglla

Frontiers in Education · 12.06.2026 · 11

DOI: 10.3389/feduc.2026.1769356 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Frontiers in Education, online erschienen am 12.06.2026. Verfasst von John Tipaz und 4 weiteren Autor:innen. Die Arbeit nennt 28 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The integration of artificial intelligence (AI) in physics education has shown remarkable potential, particularly in learning analytics and personalized feedback for interpreting graphical problems in kinematics and dynamics. Although models such as ChatGPT-4 show promising capabilities in problem-solving, they still face significant challenges in accurately interpreting complex graphical representations. While these tools may support learning and provide immediate feedback, their limitations become evident in contexts requiring visual inference or more sophisticated contextual interpretation. Inconsistencies observed in some AI-generated responses reinforce the need for human oversight, as the instructor's role remains essential for verifying information, correcting misunderstandings, and providing pedagogically sound feedback. This study underscores AI's value as a complementary educational tool, emphasizing the instructor's irreplaceable role in ensuring a deep and rigorous understanding of physical concepts. Through a learning analytics lens, this exploratory mixed-methods study examined student perceptions and expert-rated quality of AI-generated feedback in a convenience sample of 48 students, using survey responses and an expert rubric; results showed favorable perceptions of clarity and usefulness, greater variability in trust and perceived technical accuracy, and substantial agreement between expert raters. Given the convenience sample of 48 students and the absence of a control group, the findings should be interpreted as exploratory and context-dependent rather than causal or statistically generalizable.

Zitieren

APA

Tipaz, J., Ortega-Chávez, K., Hidalgo, J., López-Ruano, J., & Casaliglla, D. (2026). AI-generated feedback on student solutions to graphical problems in kinematics and dynamics. Frontiers in Education, 11. https://doi.org/10.3389/feduc.2026.1769356

Vancouver

Tipaz J, Ortega-Chávez K, Hidalgo J, López-Ruano J, Casaliglla D. AI-generated feedback on student solutions to graphical problems in kinematics and dynamics. Frontiers in Education. 2026;11. doi:10.3389/feduc.2026.1769356

BibTeX
@article{Tipaz2026,
  author = {Tipaz, John and Ortega-Chávez, Kevin and Hidalgo, Johnny and López-Ruano, Juan and Casaliglla, Darwin},
  title = {AI-generated feedback on student solutions to graphical problems in kinematics and dynamics},
  journal = {Frontiers in Education},
  year = {2026},
  volume = {11},
  doi = {10.3389/feduc.2026.1769356}
}
Kostenlos starten →