I recently came back into the world of teaching after several decades in practice/industry. The world has changed a bit since I last taught, and I am trying to find my way back in, to find ways to be pedagogically effective. I have found your posts to be helpful in figuring out what works and what does not.
Your comments in this post made me wonder if I am more focused in the classroom on the recall component than I should be. It made me question whether, in the finance and business statistics undergraduate foundational classes I teach, I am really helping students get good at recalling concepts and formulas, or whether I am in fact getting them to a place where they can apply them. I suspect the practice problems and homework assignments help, but I am not convinced I am creating enough circumstances where students have to work through problems not directly addressed in lectures or discussions.
I am hoping to use the summer to sort through what this means for my classroom in the Fall — how to implement these ideas without overburdening my students or scaring them away!
Reminds me of the anecdote of the toddler singing the Mickey Mouse clubhouse song. They belt out M-I-C-K-E-Y M-O-U-S-E at the top of their voice but when they are asked to spell Mickey Mouse they haven’t a breeze!!
In his method certification course, Davis recommends not teaching kids the alphabet song because recall of oral syllables is managed by a different part of the brain than the translation of semantic meaning into textual symbols (spelling). Although Davis’s techniques were originally developed to overcome dyslexia and have subsequently been applied in teaching neurodivergent people, they may also be very helpful for ‘normal’ learners.
The Ophius-Cox study is quite clever! Do we take from it that the children learning the rules didn't have enough practice to become yet fluent, OR that examples, practice, and feedback are actually how students become fluent? In oral and written language especially, much of what we know is subconscious pattern recognition. Not rule memorization and application.
Great article. As an ELA teacher this would drive me nuts. I drilled parts of speech and conventions into the students but there was no measurable impact on their writing. It took me years of pulling my hair out until I understood the difference between lower level recall and high level application and creation.
Excellent post (as usual). Thank you for what you do for our field.
This post got me ruminating on inquiry and concept based learning. Perhaps these approaches to teaching and learning are where we can start constructing application practice for learners. I’m unsure. A safe answer is to wield the typical formative assessment, but what if the web you mentioned in your post (the history y7 ex) was intertwined with many aspects of inquiry and concept based learning experiences, as well as retrieval practise and the other activities as well. That may work to give kids multiple application experiences prior to summation of learning.
I have revisited this article, continuing with rumination on your Y7 example. Now I wonder how we can make this "web of learning" visual (or perhaps even visual) to learners?
Think anchor chart, but perhaps anchor web(s)?
I may give it a try in the future and update here further with my narrative account.
Thanks for all you do (and for kindling my direction of thought).
So interesting, so informative--so intriguing! It confirms what we already knew about ourselves: we teachers are so busy changing the tire while driving the car that we neglect the necessity of situating what we do within the big picture. Phonics instruction is necessary--but not sufficient to achieve understanding. Likewise, vocabulary definitions that are not situated within a knowledge web can abandon us when the words are embedded in a different context.
You say: "Where we have failed to do the groundwork of building a shared store of background knowledge and vocabulary, retrieval practice ends up doing double duty as both memory booster and makeshift semantic scaffolding." I would add that the semantic scaffolding we need also pertains to 'procedural knowledge'--what are the 'strategies' (NOT skill of the week!) we use to attack new text once we have sufficient knowledge to understand the words and content. That, too, is part of the broader teaching context that we often neglect.
Really? We teach them how to cook and clean up after themselves at school because most of the migrant parents work ridiculously long hours to give their kids opportunity and Cabramatta in NSW was ground zero for Viet and Cambo migrants who lost parental authority in the move to Australia due to language barriers and long work hours. You want to die on this hill why exactly?
The other practical point is what I’d call the ‘anchoring’ effect well chosen retrieval can have in subjects like history. The point you make about retrieval being less significant when the crucial knowledge is part of repeated webs of meaning is great to have articulated and related to learning science research. It’s the kind of thing say, Counsell, has been making in recent years and relating it to learning science is a great step.
My experience is that in practice more is learned in history when teachers make very careful choices about retrieval so it provides a skeleton on which more interconnected meaning can hang. There’s a limit time-wise to how much meaningful repetition lessons can actually deliver and you get a more optimal outcome, in limited time, if there are *some things* totally secure in memory that the structure of meaning can hang on to. The second point here is that the reason retrieval does not lead as successfully to meaningful application in subjects like history is often because the retrieval choices are too often unmeaningful factoids, as if the only thing that can be memorised, ie learned, are these. You can do retrieval of arguments - in history of cause/effect relationships etc. etc. These may often provide a much more powerful ‘anchor’ or ‘skeleton’ towards meaningful application of knowledge in the ways desired.
I am a bit confused by your examples of "applying"
If a student is asked, "Is a whale a mammal or a fish? Justify your answer."
One kid writes:
"Whales are mammals because they are big and smart."
Another kids writes:
"Whales are classified as mammals because they breathe air using lungs, give birth to live young, and feed their offspring milk. These are defining characteristics of mammals rather than fish."
It would seem the latter kid has dome more retrieval practice on the definition if a mammal and the traits of whales, right?
I'm certainly persuaded by this post. But, a) Just getting retrieval practice embedded and sequenced 3 times in a bad "skills based" curriculum is already a Herculean tasks, and b) it seems the kind of reform you're talking about is beyond the scope of an average classroom teacher, and ideally at the level of curriculum designers or textbook authors.
Surely this is because rules aren’t helpful when it comes to the most difficult words? Hardest words are idosyncratic and unique whereas rules are, well, rules. Perhaps this just produced overwhelm on what was to be recalled.
He answered this halfway down the article. Surprise surprise, it's cognitive load!
Less that the rules don't match the words needed (the study wasn't English words and their chaotic origins if you recall). But that recalling the rules took all of working memory, leaving nothing left for apprehension of the task.
Quite! Which seems to suggest it’s not that learning syntax of generic “rules” isn’t helpful, but that the kinds of task are expecting kids to do determines what knowledge is most helpful.
Very elequently explained. My paraphrase of the conclusion: that retrieval practice that is applied within a bigger meaningful whole is where knowledge is acquired in a much more integrated and conceptual way.
I am wondering if you can clarify what you mean in this part of your conclusion: " ...is a strategy of the first kind in two senses: it strengthens memory rather than induction, and within memory it is especially good at shoring up the lonely, fragile facts that would otherwise fall away."
I am stuck on how to you are referring to induction.
This article raises some really thought-provoking points about retrieval practice and its limitations in the classroom. It challenges the common assumption that more memory equals better application and makes an important distinction between “knowing that” and “knowing how.” It’s interesting to see how retrieval practice might not always translate into better performance, especially when the knowledge structure is already well-established. Definitely an article worth reflecting on!
This is what we call in sports coaching “knowledge of” vs. “knowledge about”. Knowledge of is being able to do something, knowledge about is being able to explain how to do something. I aim to coach the “knowledge of” as my athletes need to be able to do things, not just talk about them!
The problem is particularly acute in mathematics education, where the concept of rich knowledge is not sufficiently theorized. It is sometimes reduced to an amorphous notion of "big ideas" which don't capture the fine structure of your history example. Ideas of meaningful learning end up in a false dichotomy with retrieval practice.
Great post.
I recently came back into the world of teaching after several decades in practice/industry. The world has changed a bit since I last taught, and I am trying to find my way back in, to find ways to be pedagogically effective. I have found your posts to be helpful in figuring out what works and what does not.
Your comments in this post made me wonder if I am more focused in the classroom on the recall component than I should be. It made me question whether, in the finance and business statistics undergraduate foundational classes I teach, I am really helping students get good at recalling concepts and formulas, or whether I am in fact getting them to a place where they can apply them. I suspect the practice problems and homework assignments help, but I am not convinced I am creating enough circumstances where students have to work through problems not directly addressed in lectures or discussions.
I am hoping to use the summer to sort through what this means for my classroom in the Fall — how to implement these ideas without overburdening my students or scaring them away!
Thank you for the post and the substack!
Reminds me of the anecdote of the toddler singing the Mickey Mouse clubhouse song. They belt out M-I-C-K-E-Y M-O-U-S-E at the top of their voice but when they are asked to spell Mickey Mouse they haven’t a breeze!!
For more on this phenomenon, check out davismethod.com
In his method certification course, Davis recommends not teaching kids the alphabet song because recall of oral syllables is managed by a different part of the brain than the translation of semantic meaning into textual symbols (spelling). Although Davis’s techniques were originally developed to overcome dyslexia and have subsequently been applied in teaching neurodivergent people, they may also be very helpful for ‘normal’ learners.
The Ophius-Cox study is quite clever! Do we take from it that the children learning the rules didn't have enough practice to become yet fluent, OR that examples, practice, and feedback are actually how students become fluent? In oral and written language especially, much of what we know is subconscious pattern recognition. Not rule memorization and application.
Retrieval practice isn’t failing.
It’s being misused as a proxy for something it doesn’t produce.
The system rewards recall because it’s measurable—
then mistakes that measurement for capability.
Great article. As an ELA teacher this would drive me nuts. I drilled parts of speech and conventions into the students but there was no measurable impact on their writing. It took me years of pulling my hair out until I understood the difference between lower level recall and high level application and creation.
Excellent post (as usual). Thank you for what you do for our field.
This post got me ruminating on inquiry and concept based learning. Perhaps these approaches to teaching and learning are where we can start constructing application practice for learners. I’m unsure. A safe answer is to wield the typical formative assessment, but what if the web you mentioned in your post (the history y7 ex) was intertwined with many aspects of inquiry and concept based learning experiences, as well as retrieval practise and the other activities as well. That may work to give kids multiple application experiences prior to summation of learning.
I have revisited this article, continuing with rumination on your Y7 example. Now I wonder how we can make this "web of learning" visual (or perhaps even visual) to learners?
Think anchor chart, but perhaps anchor web(s)?
I may give it a try in the future and update here further with my narrative account.
Thanks for all you do (and for kindling my direction of thought).
So interesting, so informative--so intriguing! It confirms what we already knew about ourselves: we teachers are so busy changing the tire while driving the car that we neglect the necessity of situating what we do within the big picture. Phonics instruction is necessary--but not sufficient to achieve understanding. Likewise, vocabulary definitions that are not situated within a knowledge web can abandon us when the words are embedded in a different context.
You say: "Where we have failed to do the groundwork of building a shared store of background knowledge and vocabulary, retrieval practice ends up doing double duty as both memory booster and makeshift semantic scaffolding." I would add that the semantic scaffolding we need also pertains to 'procedural knowledge'--what are the 'strategies' (NOT skill of the week!) we use to attack new text once we have sufficient knowledge to understand the words and content. That, too, is part of the broader teaching context that we often neglect.
Would you say once phonemic awareness is high you can get a majority of the "vocabulary building" benefit from video and podcast etc?
You can listen to a podcast while doing the dishes.
Really? We teach them how to cook and clean up after themselves at school because most of the migrant parents work ridiculously long hours to give their kids opportunity and Cabramatta in NSW was ground zero for Viet and Cambo migrants who lost parental authority in the move to Australia due to language barriers and long work hours. You want to die on this hill why exactly?
The other practical point is what I’d call the ‘anchoring’ effect well chosen retrieval can have in subjects like history. The point you make about retrieval being less significant when the crucial knowledge is part of repeated webs of meaning is great to have articulated and related to learning science research. It’s the kind of thing say, Counsell, has been making in recent years and relating it to learning science is a great step.
My experience is that in practice more is learned in history when teachers make very careful choices about retrieval so it provides a skeleton on which more interconnected meaning can hang. There’s a limit time-wise to how much meaningful repetition lessons can actually deliver and you get a more optimal outcome, in limited time, if there are *some things* totally secure in memory that the structure of meaning can hang on to. The second point here is that the reason retrieval does not lead as successfully to meaningful application in subjects like history is often because the retrieval choices are too often unmeaningful factoids, as if the only thing that can be memorised, ie learned, are these. You can do retrieval of arguments - in history of cause/effect relationships etc. etc. These may often provide a much more powerful ‘anchor’ or ‘skeleton’ towards meaningful application of knowledge in the ways desired.
I am a bit confused by your examples of "applying"
If a student is asked, "Is a whale a mammal or a fish? Justify your answer."
One kid writes:
"Whales are mammals because they are big and smart."
Another kids writes:
"Whales are classified as mammals because they breathe air using lungs, give birth to live young, and feed their offspring milk. These are defining characteristics of mammals rather than fish."
It would seem the latter kid has dome more retrieval practice on the definition if a mammal and the traits of whales, right?
Isn't this application?
I'm certainly persuaded by this post. But, a) Just getting retrieval practice embedded and sequenced 3 times in a bad "skills based" curriculum is already a Herculean tasks, and b) it seems the kind of reform you're talking about is beyond the scope of an average classroom teacher, and ideally at the level of curriculum designers or textbook authors.
Surely this is because rules aren’t helpful when it comes to the most difficult words? Hardest words are idosyncratic and unique whereas rules are, well, rules. Perhaps this just produced overwhelm on what was to be recalled.
He answered this halfway down the article. Surprise surprise, it's cognitive load!
Less that the rules don't match the words needed (the study wasn't English words and their chaotic origins if you recall). But that recalling the rules took all of working memory, leaving nothing left for apprehension of the task.
Quite! Which seems to suggest it’s not that learning syntax of generic “rules” isn’t helpful, but that the kinds of task are expecting kids to do determines what knowledge is most helpful.
Very elequently explained. My paraphrase of the conclusion: that retrieval practice that is applied within a bigger meaningful whole is where knowledge is acquired in a much more integrated and conceptual way.
I am wondering if you can clarify what you mean in this part of your conclusion: " ...is a strategy of the first kind in two senses: it strengthens memory rather than induction, and within memory it is especially good at shoring up the lonely, fragile facts that would otherwise fall away."
I am stuck on how to you are referring to induction.
Excellent article. Thanks and keep them coming.
This article raises some really thought-provoking points about retrieval practice and its limitations in the classroom. It challenges the common assumption that more memory equals better application and makes an important distinction between “knowing that” and “knowing how.” It’s interesting to see how retrieval practice might not always translate into better performance, especially when the knowledge structure is already well-established. Definitely an article worth reflecting on!
This is what we call in sports coaching “knowledge of” vs. “knowledge about”. Knowledge of is being able to do something, knowledge about is being able to explain how to do something. I aim to coach the “knowledge of” as my athletes need to be able to do things, not just talk about them!
The problem is particularly acute in mathematics education, where the concept of rich knowledge is not sufficiently theorized. It is sometimes reduced to an amorphous notion of "big ideas" which don't capture the fine structure of your history example. Ideas of meaningful learning end up in a false dichotomy with retrieval practice.
This is pure gold!