The study, led by Ariel Torres of Universidad Tecnológica Indoamérica in Quito, together with colleagues at Universidad de Las Américas, is deliberately framed as a critical analytical reflection rather than a conventional systematic review. Instead of exhaustively cataloguing every primary study on university teaching, the team purposively selected the highest tier of evidence available: systematic reviews, meta-analyses, umbrella reviews, and scoping reviews. Searches of Scopus, Web of Science, PubMed, and Google Scholar were refined iteratively, with studies admitted only if they addressed higher education and touched at least one of five analytical domains the researchers had defined in advance. Opinion pieces, editorials, and narrative reviews without methodological grounding were excluded.

The analytical machinery behind the synthesis was equally deliberate. The researchers built an analytical matrix to organize each study’s characteristics, methodological design, key findings, and conceptual contributions, allowing systematic cross-study comparison. Iterative coding then grouped recurrent findings into thematic categories, which were refined until the team reached what they call conceptual stability: the point at which further analysis no longer produced substantive changes in a category’s content or boundaries. Interpretative validity was strengthened through conceptual triangulation across study designs and educational contexts, filtering out isolated or idiosyncratic findings in favor of patterns that held across multiple bodies of evidence.

What emerged from this process was striking convergence around five interrelated dimensions. The first is teacher–student interaction, which appeared throughout the evidence not as a standalone construct but as the connective tissue of pedagogical practice, embedded in simulation, team-based learning, clinical case discussion, collaborative work, and concept mapping. Meta-analytic evidence cited in the study indicates that teaching-related factors, including the classroom learning environment and the quality of teacher–student interaction, are significantly associated with student learning outcomes, giving this relational core an empirical anchor.

The second dimension encompasses higher-order cognitive processes: critical thinking, clinical reasoning, and decision-making. These were reported across simulation-based instruction, error-based learning, concept mapping, and self-regulated learning contexts, where students plan, monitor, and evaluate their own learning. In medical education specifically, clinical reasoning surfaced in diagnostic and decision-making tasks, while scoping reviews of artificial intelligence documented AI applications that engage learners in reasoning and judgment. The third dimension, pedagogical strategies, is notable for what it does not claim: no single method dominates. Rather, the evidence describes a constellation of active and adaptive approaches, from problem-based and game-based learning to experiential designs, whose effectiveness depends on how well they organize interaction and cognitive engagement rather than on any inherent superiority of one technique.

Assessment forms the fourth dimension, and the synthesis repositions it fundamentally. Formative and authentic assessment, particularly when built around feedback processes and real-world tasks, emerged not as a terminal stage of instruction but as an embedded component of learning itself. A systematic review of 94 studies on authentic assessment documented the use of professional-context tasks in which student performance is evaluated through work that mirrors practice. A separate review of 73 studies on online peer feedback found that cognitive, task-focused comments were the most frequent form of feedback, typically structured through rubrics and student preparation. Notably, the evidence also revealed gaps: relatively few studies combined peer, teacher, and automated feedback, and student-related variables within feedback processes were rarely examined.

Digital technologies constitute the fifth dimension, and here the study takes a deliberately contrarian stance against technological determinism. Across the evidence, technologies, from conventional online and blended environments to learning analytics and artificial intelligence, function as mediating tools embedded within pedagogy and assessment, not as independent drivers of quality. A scoping review synthesizing 278 publications documented AI applications across admissions, teaching, assessment, and clinical reasoning in medical education, while a systematic review of 83 studies identified instructional designs in AI-supported environments featuring automatic feedback, intelligent tutoring, and personalization. The message is that the impact of any tool depends entirely on how it is integrated into instructional design.

From this convergence, the researchers propose their central contribution: the Dynamic Pyramidal Model of Teaching Excellence. The pyramidal architecture, the authors stress, is not an empirically derived hierarchy or causal sequence but a theoretical structure for visualizing how the components relate. Social and contextual conditions, including family, institutional environments, and socio-economic factors, form the base within which educational processes operate. Teacher–student interaction sits at the relational core, interconnected with higher-order cognitive processes, pedagogical strategies, formative assessment, and technological mediation. Motivation and vocation, which did not emerge as primary categories from the analytical corpus but were incorporated from the broader theoretical and policy literature, appear as individual influences. Teaching excellence itself occupies the apex as an integrative conceptual construct rather than a proven endpoint, with bidirectional relationships representing continuous interaction and feedback among all components.

The model also surfaces an uncomfortable geographical truth about the evidence base itself. The reviewed studies reveal a pronounced concentration of high-level research in North America and Europe, particularly in areas related to artificial intelligence and innovative pedagogical practices. The authors argue that this imbalance highlights the need for substantially more high-level evidence from underrepresented regions, including South America, before any framework of teaching excellence can claim global applicability. The study is candid about its own limitations as well: the purposive, non-PRISMA search strategy may have introduced selection bias, and the interpretative nature of the analysis means that category identification remains influenced by the researchers’ analytical perspective.

Perhaps the most consequential implication is practical. If teaching excellence is a dynamic, systemic construct, then efforts to improve it cannot succeed through isolated interventions, such as a one-off training workshop on active learning or a standalone rollout of new educational software. The authors argue that quality enhancement requires the coherent integration of multiple dimensions within educational systems, aligning objectives, activities, assessment, and technology in a continuous feedback loop. The Dynamic Pyramidal Model has not yet undergone empirical validation, and the researchers explicitly call for future work to operationalize and test its proposed relationships across diverse disciplinary, institutional, and cultural settings. But as universities worldwide grapple with AI disruption, post-pandemic learning recovery, and intensifying scrutiny of teaching quality, the study offers a compelling reframing: excellence in higher education is not something a teacher possesses, but something a system produces.

Subject of Research: A conceptual model of teaching excellence in higher education integrating five interrelated dimensions derived from a synthesis of 35 evidence reviews

Article Title: A dynamic pyramidal model of teaching excellence in higher education: a critical analytical reflection

Article References: A dynamic pyramidal model of teaching excellence in higher education: a critical analytical reflection. (n.d.). https://doi.org/10.3389/feduc.2026.1960685

Image Credits: AI Generated

DOI: 10.3389/feduc.2026.1960685

Keywords: teaching excellence, higher education, dynamic pyramidal model, teacher-student interaction, formative assessment, digital technologies, critical thinking, pedagogical strategies, artificial intelligence in education, systematic review, educational research, constructive alignment