Abstract:

Formative assessment should be integrated into STEAM (Science, Technology, Engineering, Arts, and Mathematics) education to promote deep learning, creativity, and problem-solving abilities. The purpose of this review is to comprehensively extract recent (2015-2025) Chinese and English literature on the impact of teachers and AI-powered automated scoring tools on student learning motivation in formative STEAM education evaluation. To locate relevant sources, the current study relied mostly on academic databases such as Google Scholar, ResearchGate, Taylor & Francis, ScienceDirect, Scopus, and Web of Science. The search engine used English keywords including "formative assessment," "artificial intelligence," "AI," "teacher," "feedback," "student motivation," "STEAM education," "self-regulated learning (SRL)" and "self-determination theory." The preliminary search found 85 articles deemed related to the topic. Following a rigorous screening process that included title and abstract, as well as full-text review, the deduplication stage revealed 15 studies that matched the inclusion criteria.

Previous research comparing teacher grading to AI-powered automated scoring methods has frequently been extremely restricted in its focus on feedback quality and efficiency, reducing the topic to a binary decision. This paper contends that such a framework is restricted. Its goal is to propose a paradigm shift: instead of comparing feedback sources, it will investigate how AI can fundamentally reshape the formative assessment ecosystem in STEAM education, with the explicit goal of increasing student motivation by developing their self-regulated learning (SRL) skills. The ultimate goal is to develop a formative assessment environment that allows students to study freely, hence increasing their intrinsic drive in STEAM topic learning.

Keywords:

Formative assessment, STEAM education, AI tools vs Teacher, Student motivation, SRL.