At the end of 2025, I had the pleasure of discussing HyFlex with Professor Kenan Dikilitaş (University of Bergen; HyComm project lead). My interest in HyFlex comes from a long-standing preoccupation with learning that holds up beyond the course in which it is produced. Over the past decade, much of my work has focused on designing learning architectures in which application, feedback, and evidence of capability are deliberately aligned so that what learners produce can travel into new contexts rather than expire at the end of a semester.
My interpretation of HyFlex is grounded in prior work designing competency-based learning architectures, most notably at Minerva University. There, learning was organized around explicitly defined capabilities and their enactment across contexts. The central design problem went beyond how content would be delivered, to how students would demonstrate capability across domains, situations, and formats.
This architecture deliberately extended beyond the academic container. Students were expected to enact competencies in real-world projects with external partners: companies, NGOs, and public institutions so that evidence of learning emerged under authentic conditions and could persist beyond a single course.
Working within this system allowed me to see the benefits of thinking architecturally about learning: separating capability from container and assessment from any single instructional mode. From this vantage point, HyFlex appears as a rigorous foundation for designing learning that can travel.
The problem: education that doesn’t travel
Traditional higher education remains organized around knowledge transmission, and it consistently struggles to produce learning that students would use and adapt in new contexts, not merely reproduce in familiar ones.
This gap becomes even more pronounced in the current GenAI context where institutions increasingly question assessment authenticity and validity of evidence.
HyFlex as a Design Discipline
HyFlex is framed as a configuration that offers participation pathways across face-to-face, synchronous online, and asynchronous modalities. What matters more for my purposes is what this requirement forces in design. Focusing on two of Brian Beatty’s principles – Equivalency and Reusability – I show how HyFlex can enable learning to move beyond a single course and into authentic practice settings.
Equivalency: from content coverage to capability-and-evidence
In HyFlex, Equivalency requires that learning activities in each participation mode lead to equivalent learning outcomes, even when engagement pathways differ. Implemented rigorously, equivalency forces at least three design moves:
First, this requirement compels educators to specify what students should be able to do independently of the delivery container, thereby separating substance from format. When “the lecture” is no longer the invariant, the guiding design question necessarily shifts: from content coverage (“What will I teach?”) to enacted capability (“What will students do?”).
Second, it requires that each engagement mode provide comparable practice, feedback, and reflection experiences. In other words, equivalency pushes design beyond distributing information and toward engineering experiences that can function across modalities.
Third, it requires that evidence criteria (the standards by which performance is judged) remain stable across modes. Designing for multiple modalities forces educators to define assessment criteria that remain valid regardless of how students participate.
What begins as a channel constraint becomes a qualitative shift.
Reusability: from course artefacts to boundary objects
Once Equivalency has made capability and evidence criteria explicit, the Reusability requirement becomes an integration gateway. In HyFlex, Reusability involves treating artefacts generated in each participation mode as learning objects available to all students. Reusability puts the spotlight on the artifact layer of learning design: what students produce, how it is curated, and how it is re-encountered as part of subsequent learning.
This reminds me of Star and Griesemer’s boundary objects. They describe boundary objects as entities that are adaptable to different viewpoints while remaining robust enough to maintain a common identity across them enabling coordination across different “social worlds.” In an educational context, a case study analysis, for example, can be a course final paper while at the same time mirroring what consultants actually produce and a code repository used in class becomes part of a professional portfolio. These aren’t just “authentic assessments.” They are objects designed to have lives beyond the course.
A concrete example comes from my work at Zayed University, where I collaborated with faculty to redesign learning around artefacts intended to operate across contexts rather than terminate within a single course. In one instance, a final research paper was structured as a shared object with several legitimate lives.
Academically, it provided evidence of rigorous disciplinary application. At the same time, its research question was defined with an external partner around a live organizational challenge, enabling feedback on practical relevance alongside scholarly quality.
That same artifact was then re-expressed as a presentation in a communication course and used as the focal point for collaborative work, peer feedback, and reflection on teamwork. These were coordinated encounters with the same object, each governed by distinct criteria. The design was scaffolded throughout the semester and sustained by coordination among faculty and partners to ensure academic standards were met while placing students in conditions that supported transfer beyond the course. The resulting artefacts were assessed within their immediate contexts but also curated for re-encounter. reused as case material, incorporated into portfolios, and mobilized in career development conversations.
In retrospect, this design anticipated what HyFlex formalizes: stable capability expectations and evidence criteria, combined with flexibility in how, where, and through which activities those capabilities are enacted.
From course boundaries to permeability
Equivalency makes capability explicit; reusability creates the connection layer through which evidence of that capability can travel across contexts.This is the pivot from “course design” to “learning architecture”. Under these conditions, the boundaries between courses become more permeable: one course’s outputs can become another’s inputs, and student work can be read against professional contexts rather than only academic ones.
In this light, I see HyComm’s emphasis on AI-supported HyFlex practice as a pragmatic response to the operational demands that Equivalency and Reusability create. AI can plausibly support the “connective work” that otherwise limits scale: curating and indexing reusable artifacts, flexibly translating them across modalities, and increasing feedback capacity.
While flexibility is a strong point of HyFlex, I believe permeability may be even more powerful as it creates the pre-conditions for learning to “travel”. Furthermore, as institutions grapple with generative AI, the reflex has been toward control: proctoring, detection, restriction. A more scalable and durable response may be architectural: building courses around artifacts that must travel, criteria that hold across contexts, and evidence that accumulates rather than expires.
___
Lucian Cosinschi
Learning Systems Design, Freelance