Cognitive Load Theory: Emerging Trends and Innovations - some notes

A new special edition reflects new directions in CLT research and practice. It includes 15+ empirical and theoretical papers published between August 2024 – April 2025.

Since its inception in the 1980s, Cognitive Load Theory has informed our understanding of how learning happens and instructional design by systematically explaining how the architecture of human cognition constrains learning. For many educators it’s been a very useful theoretical lens through which to think about designing lessons and helping students learn new concepts but it has also been disputed in some quarters, specifically how we measure cognitive load.

In Cognitive Architecture and Instructional Design: 20 Years Later, Sweller, van Merriënboer, & Paas reflected on how the theory had moved from simple lab tasks to robust, classroom-tested principles that influence everything from textbook design to teacher training. But that was six years ago and since then, CLT has continued to evolve. The field is now grappling with newer questions: How do motivation, interest, and self-regulation interact with cognitive load? Can virtual and augmented reality support deep learning or do they just add distraction?

This new special issue on CLT reflects that forward momentum and explores some really interesting areas. I made some notes on the papers I thought would be interesting to educators.

1. Best of Both Worlds? Combining Physical and Mental Self-Management Strategies to Support Learning from Split-Attention Examples

Summary: In Best of Both Worlds?, Combining Physical and Mental Self-Management Strategies to Support Learning from Split-Attention Examples Björn B. de Koning explored whether combining physical strategies like annotation with mental strategies such as imagining connections between text and images helps reduce cognitive load. Learners using both strategies reported lower mental effort and greater instructional efficiency, though learning outcomes didn’t significantly improve. Still, doing the same work with less strain is important, especially for complex materials.

What do teachers need to know?
This study is useful for teachers who often deal with materials that are less than ideal where diagrams and explanations are separated. While the results aren’t revolutionary, they reinforce the idea that students can learn to manage cognitive load themselves when materials aren’t perfectly designed. If you teach in resource-constrained settings, this is empowering: a few minutes spent modelling how to annotate and mentally “connect the dots” could make a real difference in how students handle complex material. The study doesn’t claim big gains either but reducing mental strain for the same result is a win, especially for struggling learners.

Interestingly, the authors suggest that whether students are working on paper or a screen can actually change how well they learn from split-attention materials, which is something we often overlook.

This question seems to fascinate educators (at least on Twitter) but I’m not sure the differences are as great as the fascination warrants tbh.

2. The Relationship between Interleaving and Variability Effects: A Cognitive Load Theory Perspective

Summary: This study is quite useful for teachers and explores the relationship between two instructional strategies synonymous with cognitive load theory: interleaving (mixing different types of content) and the variability effect (using examples with diverse surface features but similar underlying structures). Through two experiments, the researchers demonstrate that interleaving is beneficial specifically when students need to learn to discriminate between seemingly different problems that actually require similar solutions. The first experiment showed no benefit when interleaving obviously different subjects (math and language), while the second experiment found significant benefits when interleaving different-looking math problems that required the same solution approach. The researchers conclude that the interleaving and variability effects are closely related cognitive phenomena, both work by helping students recognise essential similarities across superficially different problems, increasing intrinsic cognitive load in a productive way.

What do teachers need to know? this one confirms something we’ve known for a while: interleaving works best when students need to recognise similarities beneath surface differences in problems, not when content is already easily distinguishable. In other words, mixing up various math problems with identical solution strategies improves learning, while interleaving obviously different subjects like math and language provides no advantage. The key insight is that interleaving works by increasing productive cognitive load through discrimination learning, helping students develop the crucial skill of correctly categorising problems despite superficial differences.

One thing I find very interesting here is an idea that seemed quite obvious to me - that the interleaving effect is linked to spacing but the authors here challenge that. See here:

When they interleaved math and language topics (which created spacing between similar content), they found no learning advantage over blocked practice. This suggests that the benefits of interleaving come specifically from discrimination learning, ie helping students learn to distinguish between similar problem types rather than from the spacing effect. *This distinction is important for teachers because it means we shouldn't just space out practice for the sake of spacing; instead, we should strategically interleave problems that require students to identify subtle differences in solution approaches.

*I think this will become a lethal mutation unless we understand this difference.

3. The Worked-Example Effect and a Mastery Approach Goal Orientation

Summary: This study investigated how a student's learning mindset, specifically, a mastery approach goal orientation or MAGO (lol!) affects their learning when using worked examples in mathematics. The researchers had 98 ninth-grade boys (average age 13.9 years) learn algebra through either worked examples (where they studied step-by-step solutions before solving similar problems) or through independent problem-solving. They found that worked examples generally helped students retain information better and reduced their cognitive load (mental effort) compared to problem-solving alone.

What do teachers need to know? The worked example effect is one of the most well evidenced areas of CLT but it doesn’t always work, this study really surprised me for example. Probably the most striking finding was that worked examples were primarily beneficial for students with a high mastery orientation. In other words, those who genuinely wanted to learn and master the material rather than just complete the task. Prior knowledge was the only factor that predicted transfer performance. Also, simply providing worked examples wasn't enough to help students transfer knowledge to new problems, suggesting the need for additional instructional strategies to promote transfer.

4. The More, the Better? Exploring the Effects of Modal and Codal Redundancy on Learning and Cognitive Load: An Experimental Study

Summary: This study looked at how different types of redundancy such as adding extra written or visual information, impact learning. While visual redundancy (image + text) supported learning, verbal redundancy (spoken + written text) actually increased cognitive load and hurt performance. The best outcomes came when visual support was used without overloading the same modality.

What do teachers need to know? This paper is particularly relevant in an era of over-produced slides and multimedia lessons. It shows that adding more modes of input like reading slides out loud while students are also reading them, can backfire. The key takeaway? More isn’t always better. What’s interesting here is that even when content is technically the same, how it’s delivered really matters. If you’re designing presentations or digital lessons, consider pairing spoken explanations with visuals not with more text. This confirms previously established findings: redundant talk-and-text can feel overwhelming, not helpful.

One aspect of this was kind of surprising: the study found no advantage for narration over written text when paired with images, contradicting Mayer and Moreno’s well-known modality effect (learning is enhanced when verbal information is presented in an auditory format such as spoken narration rather than as visual text when it accompanies related visual materials such as animations, diagrams, or pictures.) This suggests that in some contexts like tasks with lower complexity, both formats may engage cognitive processing similarly, blunting the expected CLT effects. The authors write:

5. How Scientific Is Cognitive Load Theory?

Summary: This review compared CLT research to other areas of educational psychology and found it to be more methodologically rigorous. Most CLT studies use experimental designs and avoid making practical claims not backed by evidence—unlike many correlational studies in the field.

What do teachers need to know? This isn’t a classroom-facing study, but it’s reassuring for anyone using CLT-informed strategies. The paper argues that CLT, unlike some of educational psychology, is actually based on experimental evidence not vague correlations

I’ve mentioned this paper before. This thread includes some comment from Dylan Wiliam and some criticism from others.

6. Self-Regulated Learning and Cognitive Load (Gorbunova et al., 2024)

Summary: This study examined how students' ability to self-regulate their learning (things like planning, setting goals, and reflecting) interacted with their prior knowledge to influence learning in an online legal task. The students with better self-regulation and more background knowledge performed better and experienced more productive cognitive load.

What do teachers need to know?
This study is probably one for teachers using online or blended classrooms, where students often have more freedom but also more responsibility. It reinforces what many teachers see daily: some students flounder not because they can’t understand the material, but because they don’t manage their own learning well.

7. Interest and Mental Effort

Summary: This one investigated how student interest in a topic influenced how much mental effort they were willing to invest, especially across tasks of varying difficulty. The researchers found that interest led to more effort for easier tasks, but as tasks got harder, interest didn’t make much difference, the effort depended more on the perceived difficulty. Perhaps most notably, the study challenges simplistic views of cognitive load by showing that identical tasks can impose different cognitive burdens depending on student interest. This suggests that traditional instructional design principles focused solely on task complexity might be insufficient without considering motivational factors.

What do teachers need to know?
This is a helpful corrective to the idea that “engagement” solves everything. Yes, students who are interested in a topic are more likely to put in the effort, but only when the material is within reach. When things get tough, interest alone isn’t enough. The implication? Basically, for difficult content, clear explanations, strategic scaffolding, and carefully sequenced instruction are more important than making content "exciting."

8. Cognitive Reappraisal: The Bridge between Cognitive Load and Emotion

The paper builds a case that emotions directly impact cognitive load by consuming working memory resources, affecting motivation, influencing cognitive processes, and sometimes constituting intrinsic cognitive load themselves. When negative emotions like anxiety arise during learning, they create extraneous cognitive load that competes with essential processing. Cognitive reappraisal emerges as an adaptive strategy that helps learners manage these emotional responses, potentially reducing cognitive overload and improving learning outcomes.

Table 2. Ways Emotion Influences Cognitive Load [5].

What do teachers need to know? Cognitive load has emotional dimensions: When students become overwhelmed, this isn't just a cognitive phenomenon—it's also an emotional one. Feelings of anxiety or frustration directly consume working memory resources and create extraneous cognitive load. With practice, cognitive reappraisal can become more automatic, reducing its own cognitive demands while maintaining its benefits for managing emotional load.

9. Mixed Reality and Procedural Learning

This paper is not really in my wheelhouse but I found it very interesting. Basically the researchers looked at whether applying the split-attention principle from cognitive load theory could improve learning outcomes when teaching surgical knot-tying skills using mixed reality (MR) technology.

Summary: Using Microsoft HoloLens, this study tested whether integrated visuals in mixed reality (MR) could improve procedural learning (e.g., knot-tying). Although performance wasn’t significantly better, learners reported lower intrinsic load and more efficient processing with integrated formats.

The researchers note that the split-attention effect may only become apparent when there are many connections between information sources. For teaching complex technical skills, ensure that instructional materials emphasize these connections.

What do teachers need to know?
This study is a helpful reminder that flashy technology doesn't replace sound instructional design. If you’re using VR or AR in the classroom, the key is to make sure the instructions are spatially aligned with the visuals, just like you would on a worksheet or PowerPoint slide. The tech here didn’t magically improve outcomes, but it did make the process feel less mentally taxing for learners. That’s still valuable I guess, especially for procedural skills or hands-on learning where overwhelm can be a barrier to entry.

What's particularly interesting is how they connect their procedural learning task (knot-tying) to the broader research that has primarily focused on declarative knowledge. The bit here highlights an important nuance - the split-attention effect might apply differently to different types of knowledge acquisition. It affects recall of factual information but not necessarily inference or skill-based learning.

10. Comparing Real and Imitative Practice with No Practice during Observational Learning of Hand Motor Skills from Animations

Again, this one is a little outside my field but I found it so fascinating and will be of great interest to many teachers teaching physical skills. I’m particularly interested in this myself as I’m encouraging my daughters to take up a musical instrument.

Summary: In Comparing Real and Imitative Practice with No Practice during Observational Learning of Hand Motor Skills from Animations, Mian et al. conducted two experiments comparing different practice types when learning motor skills through instructional animations. The first experiment examined piano playing, finding imitative practice (gesturing in air) improved learning efficiency compared to no practice. The second experiment focused on paper folding, where real practice led to significantly better performance than observation alone. Interestingly, gender was a moderating factor—females benefited significantly from both practice types in paper folding, while males showed no significant improvement with practice. Task complexity also influenced results, with practice becoming more beneficial as tasks grew more difficult.

What do teachers need to know? The key finding is that practice helps but the type of practice that works best depends on both the task and the learner. From a practical standpoint, the study recommends that when developing instructional animations, educators should consider "transience, pace, and levels of element interactivity" (p. 2) to avoid overwhelming students' working memory. Breaking animations into "shorter sections with pauses" (p. 2) can help reduce cognitive load and improve learning outcomes.

However there is a cost: "The transient nature of animations has been found to cause a burden on our limited working memory" (p. 2). Presumably, this is because the information presented in animations disappears quickly (is transient), requiring learners to hold information in memory while trying to process new incoming information.

---

ok that’s it. I guess the next frontier will inevitably be how AI informs the field, but this special issue shows that CLT has much to offer any area where human learning is considered in detail. What stands out across these papers is not just the consistency of core principles like split attention and redundancy, but how researchers are starting to test those ideas in more complex, realistic, and tech-rich learning environments.