Blended Learning (BL) has become an increasingly prominent model in higher education, integrating digital tools with face-to-face instruction to enhance flexibility and engagement. However, implementing formative assessment in BL environments remains a complex and evolving area of research. This study contributes to the field by presenting a scoping review of formative assessment strategies within BL, mapping key approaches, challenges and emerging trends.

The research methodology followed a scoping review approach, employing traditional search techniques and Generative AI-powered tools (Scopus AI and Elicit) to identify relevant literature on formative assessment in BL. The study's predefined inclusion criteria focused on higher education settings, studies published between 2014 and 2024, and peer-reviewed articles written in English. Both empirical and conceptual studies discussing formative assessment tools, frameworks, and strategies were considered. While AI-powered tools helped with the identification of relevant literature in the Scopus and Web of Science databases, the selection process was carried out by human researchers following the PRISMA protocol. After screening 225 initial records, the final dataset included 32 studies.

By leveraging AI-powered platforms such as Scopus AI and Elicit, this study employed GenAI to refine search strategies, identify relevant literature, and extract conceptual patterns across studies. This approach has the potential to increase efficiency in navigating the expansive body of research while maintaining methodological rigour. However, this study also reports on critical discussions about the implications of AI-driven tools in literature reviews.

Our findings consider both the process and product of the scoping review. The review categorized BL approaches into transmissive (student-content interaction), active (student-teacher interaction), and collaborative (student-student interaction) models, mapping their use in formative assessment practices. Findings from this review highlight the diverse formative assessment strategies used in BL, including self-assessment tools, and collaborative learning through technologies are becoming more and more widespread. Formative assessment in BL literature was found to be increasingly supported by digital tools that enable personalized feedback, student engagement tracking, and adaptive learning pathways. However, challenges persist in aligning these (technology-enhanced) strategies with pedagogical goals, ensuring accessibility, and maintaining academic quality and equity. Moreover, new insights emerged into the intersection of technology, assessment, and pedagogy.

This study provides valuable insights into the evolving landscape of formative assessment in BL and its implications for higher education. By synthesizing existing research and exploring AI's role in the research process, it offers valuable insights for educators and researchers to refine assessment practices in BL environments, while emphasizing the need for continued investigation into the intersection of technology, pedagogy, and assessment strategies.

Keywords: Blended Learning, Formative Assessment, Generative AI, Scoping Review, Educational Research Methods.