It seems that the new model that is being proposed is more aligned with the assumptions of ecological psychology view of cognition rather than the view of classical cognitive psychology. I wonder do you think the debates between different paradigm also help shape our understanding of concepts, including working memory?
Absolutely. Thanks for the link!! The diagram is great, it helps to show the same point I was trying to make - that while the component approach is simpler to explain, the interaction approach is more sensible ecologically. I definitely think the debate challenges cognitive scientists to test new hypotheses, but from what I see it is still quite “classical” cognitive in nature. But perhaps we should also look elsewhere…
Your reframing of the way we talk about WM really resonates with me. In my own practice I always end up worrying about its limitations rather than maybe playing to its strengths. Focusing on what WM can do and how it connects new bits of information to the existing network is a good perspective shift.
As I was reading, I couldn’t help but being reminded of the way that apps like Obsidian use graph theory to visualise knowledge networks (https://obsidian.md/). In that sense, I find it helpful to imagine your model as a 3D graph where new nodes are connecting to existing points across the network and accessing nodes repeatedly and in new contexts strengthens existing and builds new connections. Or am I equating human learning too strongly to computers?
Thanks, I’m really glad to hear, this was certainly the goal…
I definitely see what you mean!
I often use another model, network based, to make similar points, and when I talk about the pyramids I urge people to imagine it as a 3D space with intricate structures and cross connections. So while I had drawn this picture in mind many times, I think that In both cases (both models), when the illustrations become too complex they lose their explanatory power… so I usually choose to leave it for people’s imagination :) happy that it worked for you!
A fascinating way to visualise working memory. The pyramid analogy captures something teachers recognise every day: new learning sticks when it connects to what students already know.
It’s a helpful reminder that the challenge isn’t just managing cognitive load, but deliberately activating prior knowledge and building strong conceptual “chunks” over time. When those structures exist, working memory suddenly has much more to work with.
This is sublime! I have been wrestling with the same problem regarding visualising or forming an analogy for Cowan’s model. I believe this to be the best I’ve seen :)
The language of an “item” has always been too vague for my liking.
I’d love to suggest a consideration alongside this model if I may.
Could we see the larger compound triangles as being compressed? This was my idea for further representing the seemingly endless nature of long term memory in a finite mass brain.
I like your breakdown and illustration of the Embedded Process Model. The fact that it shows the working memory as actually working will help people connect with it. I always imagined WM as more active then usually envisioned, constantly working by shifting data, swapping it in and out, chunking bits that appeared to connect, discarding. It's called working memory so I saw it as active, not passive. I think EPM would have applications to PD and adult training, where the basic concept could be explained to adult students to help their understanding of they learn.
The more I think about it, I definitely think it should b. In many aspects it is a simpler model. it just requires a shift in thinking. we’ll see how it goes…
I’m a visual learner, so your diagrams for each case were helpful.
One question I had while looking at Case 1: you illustrate four “hooks.” How would the model shift when a teacher introduces a larger set of new terms at once—for example, ten vocabulary words in a lesson?
In classrooms, I wonder if you think the real barrier is often the pacing guide? When teachers see a model that requires smaller chunks of new information and shows that additional exposure is needed to move information into LTM, the immediate reaction is often, “I don’t have time for that.” They may believe the science and even try it briefly, but the pressure to stay on pace tends to pull instruction back toward the familiar pattern of covering material quickly.
When the teacher's guide lists 10 vocabulary words, what do you do 🤷♀️
I think it's important to realize that what's in WM changes all the time. One lesson offers dozens of opportunities to use this limited function meaningfully. In addition we're using many types of memory aids like writing the words on the board and in the notebooks. So what this means is that using several new words in one sentence/ paragraph is overloading. But there are many good ways to do meaningful work with 10 new words in one lesson. The big question is in what context (e.g familiar sentence structures and context details), which way? is it just listening or also writing, reading and speaking. How it is sequenced and importantly - rehearsed. So again - it less about the exact number and more about the prior knowledge and the structure of knowledge
It seems that the new model that is being proposed is more aligned with the assumptions of ecological psychology view of cognition rather than the view of classical cognitive psychology. I wonder do you think the debates between different paradigm also help shape our understanding of concepts, including working memory?
https://complexity-methods.github.io/book/component--vs.-interaction-dominant-dynamics.html
Absolutely. Thanks for the link!! The diagram is great, it helps to show the same point I was trying to make - that while the component approach is simpler to explain, the interaction approach is more sensible ecologically. I definitely think the debate challenges cognitive scientists to test new hypotheses, but from what I see it is still quite “classical” cognitive in nature. But perhaps we should also look elsewhere…
Thank you for the reply. Would love to see more technical debates in cognitive science that has practical implications in education. Subscribed!
"(In this situation, false retrieval of words that were not actually presented, like ‘retrieval’, is also very likely)."
I literally made this exact error when deciding to test myself, and then I continued reading and found you'd predicted it.
Astonishing post.
Haha, you made my day!
That's actually a very well known scientific effect called DRM (Deese–Roediger–McDermott paradigm)
The assumptions of cognitive load theory are very problematic. It ultimately does a very poor job of explaining how we learn.
https://m.youtube.com/watch?v=InzgJsL4VEg&pp=ygUfRGFubnkgaGF0Y2hlciBvbiB3b3JraW5nIG1lbW9yeQ%3D%3D
I do not refer to them in this post, as you may have noticed. I do see value in the translational research.
Your reframing of the way we talk about WM really resonates with me. In my own practice I always end up worrying about its limitations rather than maybe playing to its strengths. Focusing on what WM can do and how it connects new bits of information to the existing network is a good perspective shift.
As I was reading, I couldn’t help but being reminded of the way that apps like Obsidian use graph theory to visualise knowledge networks (https://obsidian.md/). In that sense, I find it helpful to imagine your model as a 3D graph where new nodes are connecting to existing points across the network and accessing nodes repeatedly and in new contexts strengthens existing and builds new connections. Or am I equating human learning too strongly to computers?
Thanks, I’m really glad to hear, this was certainly the goal…
I definitely see what you mean!
I often use another model, network based, to make similar points, and when I talk about the pyramids I urge people to imagine it as a 3D space with intricate structures and cross connections. So while I had drawn this picture in mind many times, I think that In both cases (both models), when the illustrations become too complex they lose their explanatory power… so I usually choose to leave it for people’s imagination :) happy that it worked for you!
A fascinating way to visualise working memory. The pyramid analogy captures something teachers recognise every day: new learning sticks when it connects to what students already know.
It’s a helpful reminder that the challenge isn’t just managing cognitive load, but deliberately activating prior knowledge and building strong conceptual “chunks” over time. When those structures exist, working memory suddenly has much more to work with.
This is sublime! I have been wrestling with the same problem regarding visualising or forming an analogy for Cowan’s model. I believe this to be the best I’ve seen :)
The language of an “item” has always been too vague for my liking.
I’d love to suggest a consideration alongside this model if I may.
Could we see the larger compound triangles as being compressed? This was my idea for further representing the seemingly endless nature of long term memory in a finite mass brain.
Thank you!!
So happy to hear I’m not the only one bothered by this :)
I like the idea a lot - I can see it work especially well in animated form - where you can see them compress over time. Should be fun to try!
I like your breakdown and illustration of the Embedded Process Model. The fact that it shows the working memory as actually working will help people connect with it. I always imagined WM as more active then usually envisioned, constantly working by shifting data, swapping it in and out, chunking bits that appeared to connect, discarding. It's called working memory so I saw it as active, not passive. I think EPM would have applications to PD and adult training, where the basic concept could be explained to adult students to help their understanding of they learn.
Thank you, I’m glad to hear!
The more I think about it, I definitely think it should b. In many aspects it is a simpler model. it just requires a shift in thinking. we’ll see how it goes…
I’m a visual learner, so your diagrams for each case were helpful.
One question I had while looking at Case 1: you illustrate four “hooks.” How would the model shift when a teacher introduces a larger set of new terms at once—for example, ten vocabulary words in a lesson?
In classrooms, I wonder if you think the real barrier is often the pacing guide? When teachers see a model that requires smaller chunks of new information and shows that additional exposure is needed to move information into LTM, the immediate reaction is often, “I don’t have time for that.” They may believe the science and even try it briefly, but the pressure to stay on pace tends to pull instruction back toward the familiar pattern of covering material quickly.
When the teacher's guide lists 10 vocabulary words, what do you do 🤷♀️
We're all visual learners.... :)
I think it's important to realize that what's in WM changes all the time. One lesson offers dozens of opportunities to use this limited function meaningfully. In addition we're using many types of memory aids like writing the words on the board and in the notebooks. So what this means is that using several new words in one sentence/ paragraph is overloading. But there are many good ways to do meaningful work with 10 new words in one lesson. The big question is in what context (e.g familiar sentence structures and context details), which way? is it just listening or also writing, reading and speaking. How it is sequenced and importantly - rehearsed. So again - it less about the exact number and more about the prior knowledge and the structure of knowledge