top of page

The word 'learning' is so vague as to be meaningless. Let's get multidimensional!

An abridged version of this article was published as a guest post for the Learning Scientists.

Definitions are important. Everybody knows this. If we can’t be clear about the meanings of the words we use, we should stop using them and find better ones instead. And yet, defining even the simplest of things can be horrible. If you're not sure what I mean, try defining ‘chair’ to a Devil's Advocate. It's likely to go something like this:

DA: What's a chair?.

You: It's a thing you sit on.

DA: I’m sitting on my bum, is my bum a chair?

You: A thing you sit on that isn't part of your body.

DA: Oh like the floor!

You: No, it's an object you sit on.

DA: Is a settee a chair?

You: A seat for one person.

DA: How about a tree stump?

You: No it has to have 4 legs.

DA: Like a cat?

You: It has to have 4 legs and not be alive.

DA: A dead cat.

You: Now you’re just being facetious.

DA: A table then?

You: 4 legs and a bit you can learn back against.

DA: Like this?

You: No, that's a stool.

DA: How about this, is this a chair?

You: Yes.

DA: I thought you said it has to have 4 legs?

You: Um...

DA: How about this, is this a chair?

If defining an everyday object like a chair can prove tricky, when it comes to pinning down something abstract like learning - well, you can be sure we're in for a rough ride. *

6 statements about learning

Here are six definitions of (or bold statements about) learning that I have encountered in recent years:

  1. "I define learning as the long-term retention of knowledge and skills and the ability to transfer between contexts. The retention bit is fairly straightforward and uncontroversial: if you can’t remember something tomorrow, can you really be said to have learned it?" (source)

  2. "If nothing has changed in long-term memory, nothing has been learned." (source)

  3. "Learning is acquiring knowledge and skills and having them readily available from memory so you can make sense of future problems and opportunities." (source)

  4. "Learning is what happens when you think hard." (source)

  5. "Learning is impossible without extended practice." (source)

  6. "Learning and memory are closely related concepts. Learning is the acquisition of skill or knowledge, while memory is the expression of what you’ve acquired. Another difference is the speed with which the two things happen. If you acquire the new skill or knowledge slowly and laboriously, that’s learning. If acquisition occurs instantly, that’s making a memory." (source)

To summarise: In the first 3 statements, learning is defined as something very closely related to memory. In statements 4 and 5, we find learning defined in terms of effort: learning is something you have to earn. And in statement 6, we find both of these ideas combined.

Here’s our Devil’s Advocate again, to give these statements a run for their money:

Confident blogger: "Learning is the long-term retention…"

DA: Learning is retention? Really? Is learning not more to do with acquisition – something that might lead to retention? Is it a combination of the two perhaps? Or are you defining learning as a noun here – that which is possessed by a learned person – rather than the process by which it is acquired?

Also – if learning is defined as retention, where does that leave forgetting? As Robert Bjork points out here, "forgetting enables learning and focuses remembering”. So if learning is retention, and forgetting (i.e. non-retention) enables learning – what exactly is going on? If you forget something, does that mean you never learnt it? Are you not confusing learning with 'the ability to remember'?

Also – what exactly do you mean by long-term? Psychologists use short-term memory (STM) to refer to the capacity for storing, but not manipulating, information which, in the absence of rehearsal, typically lasts for around 20-30 seconds before fading. Long-Term Memory (LTM) is considered to span from a few minutes through to a lifetime. This is quite a range! By defining learning as ‘long-term retention’, does this mean anything upwards of 30 seconds? A week? A month? A year? Also – how accurately does retention have to be maintained over the allotted time period in order to qualify as learning? Does 50% retention count as learning, forgetting or both?

CB: "…of knowledge and skills…”

DA: How are you defining knowledge here? Is all knowledge the same? Is knowledge of self the same as knowledge of history? Similarly, how do you define skills? Are skills something distinct to knowledge, or is the ability to juggle, say, merely a form of “procedural knowledge”? Is the learning process the same for memorising the names of 30 students, learning to solve a simultaneous equation, learning sign language, learning to get really good at playing bridge, or learning to play a musical instrument? Can we really lump "knowledge and skills" together in one definition?

CB: “…that is transferred.”

DA: I’ll come on to transfer in a minute.

CB: “The retention bit is fairly straightforward and uncontroversial: if you can’t remember something tomorrow, can you really be said to have learned it?”

DA: I assume you are using the word “tomorrow” figuratively, because it’s obviously a pretty low bar for retention. But the question remains: what is your cut-off? Also – what is this “something” of which you speak? Does the appropriate amount of retention not depend on the nature of the material to be learned?

If I’m in a restaurant and I can learn and retain what’s on the specials board for the few minutes it takes me to return to my table and place my order (note, longer than the standard 30-second STM decay), what’s the problem if I can’t remember next week whether the ribs came with wedges or Cajun fries? Does this mean I never learned the information in the first place?

The American Psychological Association might say “that wasn’t learning, it was just making a memory”, as per statement 6 above. But since they define memory as an expression of learning, they would be on thin ice.

An alternative view would be that I learned this information for as long as I needed it, and then I forgot it so it wouldn’t clutter up my mind. (Thank goodness - who wants to be thinking about Cajun fries 20 years after you ate them?) Why can we not recognise the functional, short-term retention and forgetting of information as a form of temporary learning?

To use a longer-term example: compare learning something simple - like the chemical symbol for Gold - with learning something more complex, like learning the words to The Element Song. If tomorrow I can’t remember that the symbol for Gold is Au, then we can agree that I haven’t learned that thing. But what if on the day of study I can remember all the words to The Element Song, and then a week later I can only remember half the words? Have I learn half the song, or forgotten half? And if I misremember some of the words - how do we define that?

Obviously, retention has something to do with learning. Perhaps it can be seen as a kind of product of learning, or a performance enabled by - but distinct from - the process of learning. But one thing is clear: defining learning as retention is hopelessly inadequate.

CB: “Acquiring knowledge and skills and having them readily available from memory so you can make sense of future problems and opportunities.”

DA: As a child, I learned to play the piano to about grade 4 or 5. I stopped playing when I was 14. When I was 24, I joined a band and started playing keyboard. I was hopeless, and essentially had to learn again from scratch. However, to my amazement I could still play the Maple Leaf Rag, note perfect. This phenomenon is sometimes referred to as "muscle memory" - a kind of implicit, procedural memory.

My ability to play the Maple Leaf Rag was not “transferrable” in any meaningful way. I could not play anything else to nearly the same standard, and it did not help me learn how to play the kind of keyboard I needed to play in a band, age 24. Nor have I yet encountered a situation where I have been able to transfer my ability to play the Maple Leaf Rag to solve a problem or make sense of an opportunity, although I remain eternally vigilant.

If 'acquired knowledge and skills' needs to be transferrable in order to satisfy your definition of learning, does the fact that my ability to play the Maple Leaf Rag is non-transferrable mean that I didn't learn it? I could give you a thousand more examples of things I have learned which are not transferrable. I'm not saying transfer is not desirable, or something to work towards in certain circumstances - but should it really feature in the very definition of learning? And what exactly do you mean by transfer? Is answering a quiz question transfer?

CB1: “Learning is what happens when you think hard” CB2: “Learning is impossible without extended practice” CB3: “If nothing has changed in long-term memory, nothing has been learned”

DA: One thing I have learned recently is that on the Monday of half-term, my son went trampolining at 4 o clock. When I woke up that morning, I did not know this information. I do now, because that day I happened to overhear a conversation about what time he was going trampolining at. You might say I acquired this knowledge through experience. It was of no importance – I didn’t have to take him there or pick him up. I did not process, manipulate or rehearse this information. It’s just a thing I noticed, which I can now recall some 19 days later - far beyond the 30-second STM window. If this is not learning – and yet it is an example of information that has been retained in the LTM – what shall we call it?

To use an example with a longer time-span, I can recall that when I was about 5, I went to a bonfire and ate a toffee apple. I have not reinforced this information with extended practice, or even really thought about it since; it’s just a thing that happened more than 30 years ago.

If it is true that “if nothing has changed in the long-term memory, nothing has been learned”, does this also work in reverse? If something is retained in the long-term memory, does that mean that that thing has been learned?

If so, does this undermine the idea that learning is something that is impossible without extended practice, or thinking hard? You might argue that this is simply a memory of an experience, rather than an example of learning. But why then do I remember that particular instance, rather than the bonfire at which I ate a candy floss or a nice slice of Parkin?

A definition of learning as "memory accruing from effort" should be able to explain why some things are retained in the LTM in the absence of hard thinking and extended practice, while others are not. Clearly, there are other factors at play which do not feature in your definition of learning.

CB: “Learning and memory are closely related concepts. Learning is the acquisition of skill or knowledge, while memory is the expression of what you’ve acquired. Another difference is the speed with which the two things happen. If you acquire the new skill or knowledge slowly and laboriously, that’s learning. If acquisition occurs instantly, that’s making a memory.”

DA: Memory is expression? Really? What about memories that are not being expressed - or which may never be expressed - but which are nevertheless stored? Also – one minute you say memory is an expression of learning; the next, merely “making a memory” does not constitute learning. That seems kind of inconsistent. How and where do you draw the line between learning and memory? Also – how exactly do you define “slowly and laboriously?” Is there a cut-off point? What if you’re a so-called “fast learner”? Is this statement equally true of novices and experts?

One of the problems here is the sense of certainty with which these 6 statements were made. With the exception of Professor Rob Coe, who follows his "hard thinking" definition with the caveat that “obviously, this is over-simplistic, vague and not original”, the statements above all seem to assume that learning is *a thing*, and all we need to do is accurately describe that thing.

We see such declarative language all the time: “Learning is tripartite”, or “Learning is liminal”. Or how about this apparently serious attempt to refine the definition, from the Journal of Psychology: “Learning is the process by which a relatively stable modification in stimulus–response relations is developed as a consequence of functional environmental interaction via the senses”.

This is even worse than I thought.

My question is this: What if learning is more than just *a thing* which can be defined in a single sentence? What if learning is… multidimensional?

Pay attention

For example, one thing that is missing from any of the six statements above is any mention of attention. Clearly, while our Devil’s Advocate did not rehearse or think hard about the fact that their son went trampolining at 4pm, on some level they attended to it while overhearing it. So perhaps attention is required for learning.

At face value, this might seem so obvious as to not even be worth mentioning. But in this this study, researchers found that learning can happen in the absence of attention. So even the requirement to "pay attention" is not as straightforward as it might seem!

In recent years, the advent of social media has precipitated a period of accelerated "learning about learning" among teachers. The profession is better informed than ever before. But I think it's also clear that if we're struggling to define learning, we still have some distance left to run. I would argue that if we are truly to navigate our way through the 'swampy lowland' of institutionalised learning, it's time to stop thinking of learning as a simple construct and embrace it in all of its dynamic, multidimensional glory.

To get the ball rolling, here are ten dimensions of learning that we might wish to consider (please note, this is not intended as an exhaustive list).

Ten dimensions of learning

Dimension 1: Learning as memory

To the extent that learning is synonymous with memory, the spectrum almost writes itself. At one end, we have stuff that is immediately forgotten. A 6-digit security code that is texted to you so you can type it into a website, say. It enters your working memory for the few seconds that you need it, perhaps in 2 or 3 digit chunks, and then fades almost immediately.

At the other end, we have things that are learned to the point of automaticity: your name, where you live, the fact that socks go on your feet. Save some brain injury or disease, you’re unlikely to forget these things in a hurry.

But even within the realm of memory, there are more things to consider than mere retention. Here are 4 other dimensions of learning related to aspects of memory:

Dimension 2: Implicit vs explicit memory

Implicit memory is sometimes referred to as automatic or unconscious memory, which may be expressed in the absence of deliberate recollection. Examples include being ride a bike, button a shirt, play the piano etc. These are examples of procedural memory (see below). Implicit memory can also include behaviours, such as locking the door as you leave the house, or associations, such those related to smells.

Explicit memory involves the conscious, intentional remembering of information. Remembering your sort code and account number is an example of explicit memory.

Dimension 3: Procedural vs declarative

Procedural memory is a subset of implicit memory, and involves being able to recall how to do things like driving a car, juggling or brushing your teeth to the point of automaticity. Declarative memory involves the factual recall of dates, facts, events and concepts. Remembering the capitals of countries, the properties of transition metals or what happened on the way to school are examples of declarative memory.

Dimension 4: Semantic vs episodic

Declarative memory is divided into two types. Semantic memory is the recall of general facts, while episodic memory is recall of personal facts. Remembering the capitals of countries is an example of semantic memory, whereas being able to recall what happened on the way to school is an example of episodic memory.

This distinction is often exploited by people who memorise competitively, using episodic memory (e.g. walking through a familiar house or location) as a framework for memorising declarative objects for by associating them with visual cues rooted in episodic memory).

Dimension 5: Recall vs recognition

Psychologists distinguish between two types of memory retrieval. Recognition refers to our ability to, er, recognise an event or piece of information as being familiar. This is the kind of memory that students might use in a multiple choice test.

Recall, on the other hand, requires all the information to be retrieved from the Long-Term Memory. This is the kind of retrieval a student might need to write an essay from memory, or to answer a long-answer question in an exam, although this might also require procedural memory as well. Recall is generally considered to require a greater depth of information processing, storage and retrieval than recognition.

If you will pardon the pun, it is also important to remember that memory is not the only game in town. Here are 5 more dimensions of learning:

Dimension 6: Naturally occurring vs. elicited data

Some things we learn can be considered naturally occurring data: your sibling’s names say, or the names for common foods. Other things we learn are elicited: you either look it up, or you’re taught it, or it’s some combination of the two.

As well as simply describing data, this can be seen as a kind of continuum which overlaps to some extent with what schools often refer to as ‘attitude to learning’. At the ‘elicited’ end of the spectrum, we find the autodidact. At the opposite end, the incurious drifter who takes life as it comes. In the middle, there are things you encounter naturally – a book that catches your eye, say – which you may then choose to pursue in more detail. There are also some grey areas here. Material you are taught on a compulsory course – is this naturally occurring or elicited?

Dimension 7: Intrinsic vs extrinsic motivation

Compare these two statements:

“I am learning about memory because I find it is fascinating”


“I am learning about memory because I want to pass the test and please my parents”.


“I am learning to solve a Rubik’s cube because I enjoy the challenge”


“I am learning to solve a Rubik’s cube so I can show off to my friends”.

Each of these motivations may be strong or weak. We also may hold several such positions simultaneously.

Dimension 8: Classical vs operant conditioning

Put simply, classical conditioning is “a learning process that occurs when two stimuli are repeatedly paired: a response which is at first elicited by the second stimulus is eventually elicited by the first stimulus alone” (source). This was made famous by Pavlov’s (rather cruel) experiments with dogs, who were repeatedly exposed to the twin stimuli of food and a bell; eventually, the bell alone was sufficient to prompt them to drool.

Classical conditioning forms the basis of much animal training, where treats are used to elicit desirable behaviours. However classical conditioning may also feature in certain types of school-based learning, such as training students to tidy up by playing a particular piece of music.

Operant conditioning is “a type of learning in which the strength of a behaviour is modified by the behaviour's consequences, such as reward or punishment” (source). Obviously this idea underpins many schools' behaviour management systems.

Dimension 9: Inductive vs deductive

Inductive reasoning is essentially a “bottom up” approach to learning, where the focus moves from the specific to the general. For example, students may be presented with several examples of a phenomenon (e.g. photographs with examples of specific animal adaptations) and they are required to identify general patterns or “rules” (e.g. camouflage, body size, ear shape).

Deductive reasoning is a “top-down” approach to learning, where the focus moves from the general to the specific. In the example above, students may be taught types of adaptations types of adaptations first, and then these “rules” are tested with particular examples (e.g. the photos of animal adaptations).

Dimension 10: Significant vs non-significant, or less-significant learning

Carl Rogers once wrote: “It seems to me that anything that can be taught to another is relatively inconsequential and has little or no significant influence on behaviour. I realise increasingly that I am only interested in learnings which significantly influence behaviour.”

It is worth noting that Rogers was a psychotherapist, and not a teacher. And it should also be borne in mind that significance is rather a subjective notion. Nevertheless, since almost all students ask the question “Why do I need to learn this?” at some point or other in their school career – understandably, some might say – the question of significance remains worthy of reflection.

So learning is multidimensional. So what?

These ten dimensions are just a few examples of the many lenses through which learning can be viewed. We could easily add more, but this blog post is an impolite length as it is and I feel the point has been made.

My question is this: how can we hope to pull together so many dimensions of learning into a single definition? One is almost tempted to come up with some kind of a visual metaphor to capture the complexity. With tongue firmly in cheek, I humbly propose Learning Kerplunk:

In Learning Kerplunk, each straw is a dimension of learning. Collectively, they hold your marbles in place. Disregard one, and the others may compensate. Disregard more than a few, and you may lose your marbles entirely.

I have no desire to add to the pantheon of ludicrous learning metaphors. However I hope I have succeeded in making the point that we need to find new ways of thinking about learning; any bold, simplistic definitions are woefully inadequate at capturing the complexity.

To bring ourselves back to earth, and to return to the start of this post:

"Definitions are important. Everybody knows this. If we can’t be clear about the meanings of the words we use, we should stop using them and find better ones instead."

Before we go any further, let's do what you should always do when we need a dose of sanity, and reach for the dictionary definition:

"Learning: The acquisition of knowledge or skills through study, experience, or being taught." (source)

Now, hold on just a minute... this is actually pretty good! It talks about acquisition rather than retention. It mentions the possibility of learning through experience, rather than limiting it to hard thinking and extended practice. And it includes the possibility for learning to arise either through teaching or through independent study. However, this definition doesn't mention memory at all. That's not going to go down well in some quarters.

Let's see what Wikipedia has to say on the matter:

"Learning is the act of acquiring new, or modifying and reinforcing existing, knowledge, behaviours, skills, values or preferences and may involve synthesising different types of information." (source)

Hold on to your hats! This definition might be flagged up as "citation needed"– never a good sign – but come on: "acquiring new, or modifying and reinforcing existing, knowledge, behaviours, skills, values or preferences"... that is kind of amazing.

And yet even this muscular sentence – this gymnastic performance of commas and caveats – makes no mention of memory, attention, rehearsal, motivation, retrieval, metacognition, chunking, agency, elicitation, intentionality, schema...

Reflecting upon all of the above, I can think of only one "learning is..." statement that I could possibly make with any confidence that it might actually be true. This I will present in the form of a motivational poster:

If this seems a rather downbeat note to end on, allow me to suggest that recognising and accepting this simple truth is the first step toward a brighter, better-defined future.

Perhaps meaningless is a bit strong. As far as the layperson is concerned, the fact that 'learning' is a bit nebulous doesn't really matter. We all sort of know what it means. But for educators, we should think of 'learning' as an umbrella term at best, taken to mean something like "the thing we want students to do". Beyond that, we really need to roll our sleeves up and start talking about the specific kinds and features of learning that we're interested in.

For example, I might say something like “I am doing a PhD in self-regulated learning. In particular, I am interested in the interplay of metacognition, self-regulation and oracy in helping people become more effective at acquiring and retaining useful knowledge and skills. To define these terms further..."

To help us navigate our way through this brighter, more sharply-defined future, we’re going to need something practical – not a new definition or metaphor, but something to help us understand an abstract umbrella term comprising many other concepts.

In my next post, I will propose a glossary of learning terms.

* The Oxford English Dictionary neatly sidesteps the entire issue by defining chair as ‘a seat for one personal that typically has 4 legs and a back’

Recent Posts

See All
bottom of page