Exploring the UDL 3.0 version, inspired by an episode on the Tea for Teaching podcast

It took me a while to appreciate the UDL* framework since I first started looking into it 3 years ago. Without context, it seemed overwhelming in the amount of details that we are supposed to consider when designing teaching. Initially, I tried making sense of all the different facets by finding examples of what they could mean in practice. One book in particular, “reach everyone, teach everyone” by Tobin & Behling (2018), helped me see the bigger picture, and eventually I translated it into a framework that works for me, especially in combination with the office analogy. Now, on another great episode on one of my favourite podcasts, Tea for Teaching, there is a super helpful conversation about UDL 3.0 (with 3.0 marking the new version that was released in July 2024) with Tom Tobin (of the book I just mentioned!), Lillian Nave, and Jennifer Pusateri. I highly recommend listening to the episode!

*(UDL stands for “Universal Design for Learning”, and “universal” is not the claim that they have found the one and only solution that works, but rather the ambition to design in such a way that no adaptations are needed later for teaching to be accessible for individuals. Of course that can only ever be an ambition that is impossible to reach in practice, but it seems like a good goal to aspire to.)

Since there are no fundamental laws of how teaching and learning works that we can fall back on (or in any case, those have yet to be discovered), we have to rely on insights from a huge body of literature across many different fields, and there have been several attempts to compile them into manageable formats. For example, the Hattie study from 2009, where they compiled over 800 studies on teaching and learning on a total of more than 1/4 million students from kindergarten to university to come up with 138 effects of different influences on student achievement. Or the meta study by Schneider & Preckel (2017), where they compiled insights from review articles on learning in higher education specifically, and from those come up with a list of 10 cornerstone findings. Or, more recently, I found the book by Barbeau & Happel (2023), where they describe six critical teaching behaviours. And then there are of course lots of great frameworks like, for example, Transparency in Learning and Teaching “TILT“, that also try to condense the most important ideas into something that teachers can use to improve their teaching without having to dig through all the literature themselves. Overall, there seems to be a convergence of the big topics that stand out in all these frameworks, and there is a large overlap with popular teaching books, like Lang (2021)’s “small teaching”, as well as what we teach in our courses because there is just ecological validity in what the research finds and what we experience in our practice.

In the case of the UDL framework, they choose to display their main topics in a 3×3 matrix: the three columns (engagement — i.e. the “why”; representation — i.e. the “what”; action and expression — i.e. the “how”) which each have the three levels of access, support, and executive function. As already stated in the headline of the graphical organiser, and alluding to those three columns, “[t]he goal of UDL is learner agency that is purposeful & reflective, resourceful & authentic, strategic & action-oriented“. The strong focus on learner agency is new in this version of UDL, and makes the framework more relevant to higher education settings. In the episode, Tom Tobin says “we should be making space to help learners understand their own learning process and then build up into agency“, and there is a marked shift towards co-creation in the framework, a “move beyond the simple neurobiology argument of UDL toward a more social constructivist mode of practice“. There is also now a much bigger focus on the intersectional identities of teachers and students, as well as power imbalances and privileges, not just between teachers and students, but also within each of the groups. Lillian Nave says in the episode that “we really don’t want the same students to be uncomfortable all of the time, but we want all of the students to be uncomfortable some of the time, and so if we can spread that out, we get all of the students into that learning zone where they risk just a little, but they’re not always in that risky fight or flight mode, then we’re better serving our students“. The old UDL slogan that “[l]earning is as unique to individuals as their fingerprints or DNA” (see also Heelan et al., 2021) makes a lot of sense here — not only is there no average learner in terms of abilities, learning and motivation and interest are also super individual, situational, changing literally from one minute to the next. What’s interesting to me right now might not be interesting to me tomorrow in the same way, but maybe if I look at it from a different angle, or in conversation with a new person.

One part of the conversation in the episode was on how UDL and AI work together. Here, they referred to an earlier Tea for Teaching episode (which I then had to go follow up on), where they discuss the potential of GenAI to help with differentiated instruction (differentiated instruction meaning the reaction to pattern that we observe in the classroom, vs UDL which is what we do before). Students can use it to customise study methods, e.g. make AI give them quiz questions about the materials, or when stuck ask GenAI “don’t give me the answer, tell me the steps to solve this problem”. They can also use it to get feedback when teachers aren’t available (see also my summary of a really interesting study on how and why students use GenAI feedback). Students can also use it to outsource some executive functions by for example using GenAI to support scheduling or to gamify tasks.

Teachers, on the other hand, can use GenAI to get help on how to approach a problem like a novice; to find thresholds or bottlenecks, and to break problems down into small steps. Which yes, is maybe a good idea before meeting students, but once the students are there, that should surely be done with them?

Another suggestion is for teachers to use GenAI to help be more mindful of student identity: By sharing the syllabus or materials plus some background info about the teacher, their identity, etc, the idea is to ask GenAI to suggest changes to make sure the syllabus or materials or whatever inclusive of other identities. Which is maybe a good idea for when you feel stuck, but I would be very much aware that GenAI is also biased in many ways, so definitely take whatever comes out with a large pinch of salt! And bias comes up a lot throughout the rest of the conversation, too, for example when thinking about ethical considerations. The three other main things highlighted for consideration are “have vs have not” (see also Marie Leijons recent comments on that), water and electricity usage, and that, as Tom Tobin says, “we preach respect for intellectual property, while the most common LLMs and generative AI models have been trained on oceans of copyrighted content without consent“…

Last idea from that episode that I thought was interesting are that there are now “UDL review tools” that can review teaching materials and suggest ways to make them more accessible. I am curious about those and would like to try just for fun, maybe in combination with that student advisory panel!

But back to the original episode on UDL 3.0, where they talk about the “lopsided house” diagram to describe how, on the path from beginner to proficient user to practitioner, the proportion of “designed” UDL decreases, while the proportion of both the tool use and agentic UDL increase. This goes back to the goal of UDL: “learner agency that is purposeful & reflective, resourceful & authentic, strategic & action-oriented”. This can, in part, be scaffolded through design, but there needs to be more and more room for learners to step up and take their own learning into their own hands. Going back to the 3×3 matrix, this corresponds to the rows going down from more scaffolding to more learner agency. In the episode itself, they of course could not go into all the details of the version 3.0. But then I am curious, so I did dig a bit after listening to the podcast. And I strongly recommend that you do, too! Here are just some details that I found particuarly interesting:

For example, in the “access” row of the “engagement” column, the considerations, under the headline of “welcoming interests and identities” are optimize choice and autonomy; optimize relevance, value, and authenticity; nurture joy and play; and address biases, threats, and distractions. All of those are so aligned with what I value and prioritise in my own teaching — designing for feelings of autonomy, competence, and relatedness; joy and play with activities like active lunch breaks, kitchen oceanography, bingos; and then of course sensitivity to identities and threats.

In the “representation” & “support” field, under “language and symbols”, there is a consideration about “cultivat[ing] understanding and respect across languages and dialects“, and the link leads to more details about what they mean by that, among others “[s]upport translanguaging, the ability to move fluidly between languages, as an approach that encourages learners to use their full linguistic capital“. As for all consideration, they give a bunch of literature that this consideration is based on, and I find this one in particular intriguing, since a while ago a colleague of mine, in a brainstorming phase in a course, explicitly encouraged participants to take notes in their first language, not the course language, because it just activates and engages different associations and processes. This stood out to me as super interesting, as in 95% of the time, I write notes to myself in English, and the remaining 5% are sometimes even in Swedish, not German, depending on context. So this feels very interesting and relevant to explore further, I wonder what I would actually be thinking if I did do it in German?

Looking into the documentation of the UDL version 3.0, I really appreciated that both a comparison of the previous version 2.2 with 3.0, as well as a rationale for updates, are available upon request. In the latter, there is a description of the process (including, for example, an advisory board, more than 40 focus groups with 181 education professionals, and a systematic literature review), and even example quotes of the concerns that went into changed language, for example from “provide” to “design”, or “checkpoint” to “consideration”. This is so interesting to browse and shows all the thought that has gone into this new version, and a great example of transparent development!

To wrap up this post, I think it is important to remember that while there is a wealth of literature and consideration that has gone into UDL 3.0 and it would probably take days to read it all and even longer to digest it, UDL is ment to be a mindset rather than a checklist, and if you want to change things in your instruction, the “plus-one approach” of doing one baby step at a time is strongly recommended rather than a complete overhaul (Heelan et al., 2021). And if you come across ideas or literature that you would like considered in the next version, they welcome input!

With that — happy exploring of UDL 3.0, and start taking baby steps! :)

Featured image: When I am thinking about inclusive teaching, I am now also always thinking about the nonclusion paper (“The problem with inclusion? It is done by somebody to somebody“), and their photo study of how signs can signal which bodies are welcome in which facilities, or just what functionality to expect behind a door. From that sign on the featured image, I am expecting basically a hole in the ground, and also probably no running water. In any case, when I saw this sign, I knew I had to take a picture because there would be a time when I would need it on my blog :-D

CAST (2024). Universal Design for Learning Guidelines version 3.0. Retrieved from https://udlguidelines.cast.org

Heelan, A., Tobin, C., & Ryder, D. (2021). UDL for FET Practitioners’ Guidance for Implementing Universal Design for Learning in Irish Further Education and Training. https://www.solas.ie/f/70398/x/81044b80ce/fet_practitioners-main.pdf