On the Definition of Learning....

6/26/2017

There was a brief, lively thread on Twitter over the weekend concerning the definition of learning. To tip my hand here at the outset, I think this debate—on Twitter and elsewhere--is a good example of the injunction that scientists ought not to worry overmuch about definitions. That might seem backwards—how can you study learning if you aren’t even clear about what learning is? But from another perspective don’t we expect that a good definition of learning might be the result of research, rather than a prerequisite?

The Twitter thread began when Old Andrew asked whether John Hattie’s definition (shown below) was not “really terrible."

I'll first consider this definition (and one or two others) as our instincts would dictate they be considered. Then I'll suggest that's a bad way to think about definitions, and offer an alternative.

Hattie's definition has two undesirable features. First, it entails a goal (transfer) and therefore implies that anything that doesn’t entail the goal is not learning. This would be….weird. As Dylan Wiliam pointed out, it seems to imply that memorizing one’s social security number is not an example of learning.

The second concern with Hattie’s definition is that it entails a particular theoretical viewpoint; learning is first shallow, and then later deep. It seems odd to include a theoretical perspective in a definition. Learning is the thing to be accounted for, and ought to be independent of any particular theory. If I’m trying to account for frog physiology, I’m trying to account for the frog and it's properties, which have a reality independent of my theory.

The same issue applies to Kirschner, Sweller and Clark's definition, "Learning is a change in long-term memory." The definition is fine in the context of particular theories that specify what long term memory is, and how it changes. Absent that, it invites those questions: “what is long term memory? What prompts it to change?” My definition of learning seems to have no reality independent of the theory, and my description of the thing to be explained changes with the theory.

It’s also worth noting that Kirscher et al’s definition does not specify that the change in long term memory must be long-lasting…so does that mean that a change lasting a few hours (as observed in repetition or semantic priming) qualifies? Nor does their definition specify that the change must lead to positive consequences…does a change in long term memory that results from Alzheimer’s disease qualify as learning? How about a temporary change that’s a consequence of transcranial magnetic stimulation?

I think interest in defining learning has always been low, and always for the same reason: it’s a circular game. You offer a definition of learning, then I come up with a counter-example that fits your definition, but doesn’t sit well with most people’s intuitions about what “learning” means, you revise your definition, then I pick on it again and so on. That's what I've done in the last few paragraphs, and It’s not obvious what’s gained.

The fading of positivism in the 1950s reduced the perceived urgency (and for most, the perceived possibility) of precise definitions. The last well-regarded definition of learning was probably Greg Kimble's in his revision of Hilgard & Marquis’s Conditioning and Learning, written in 1961: “Learning is a relatively permanent change in a behavioral potentiality that occurs as a result of reinforced practice,” a formulation with its own problems.

Any residual interest in defining learning really faded in the 1980s when the scope of learning phenomena in humans was understood to be larger than anticipated, and even the project of delineating categories of learning turned out to be much more complicated than researchers had hoped. (My own take (with Kelly Goedert) on that categorization problem is here, published in 2001, about five years after people lost interest in the issue.)

I think the current status of “learning” is that it’s defined (usually narrowly) in the context of specific theories or in the context of specific goals or projects. I think the Kirschner et al were offering a definition in the context of their theory. I think Hattie was offering a definition of learning for his vision of the purpose of schooling. I can't speak for these authors, but I suspect neither hoped to devise a definition that would serve a broader purpose, i.e., a definition that claims reality independent of any particular theory, set of goals, or assumptions. (Edit, 6-28-17. I heard from John Hattie and he affirmed that he was not proposing a definition of learning for all theories/all contexts, but rather was talking about a useful way to think about learning in a school context.)

This is as it should be, and neither definition should be confused with an attempt at a all-purpose, all-perspectives, this-is-the-frog definition.

20 Comments

Dylan Wiliam

6/26/2017 11:45:45 am

I think the problem with the word "learning" is that it is used to describe both the process of acquiring new capabilities (verb) and the resulting new capabilities (noun). Finding a definition that adequately describes both the process, and the resulting status, is in my view going to be extremely difficult. Paul Black and I tried to finesse the issue about specific goals, the long/short-term issue, and the learning/performance distinction, by defining learning (in the sense of the resulting change, rather than the processes producing it) as "an increase, brought about by experience, in the capacities of an organism to react in valued ways in response to stimuli" (Black & Wiliam, 2009 p. 10).

Steve Myran

6/26/2017 03:27:59 pm

Alexander, Schallert & Reynolds (2009) offer an effective way of addressing many of the challenges discussed above.

While this is taken out of the larger context of the article they define learning as follows:

"Learning is a multidimensional process that results in a relatively enduring change in a person or persons, and consequently how that person or persons will perceive the world and reciprocally respond to its affordances physically, psychologically, and socially. The process of learning has as its foundation the systemic, dynamic, and interactive relation between the nature of the learner and the object of the learning as ecologically situated in a given time and place as well as over time" (p.186).

http://www.tandfonline.com/doi/pdf/10.1080/00461520903029006?needAccess=true

Christopher Parsons link

6/26/2017 05:22:19 pm

My first instinct is to ridicule the cumbersome nature of that description - at first glance it doesn't at all appear to be cutting to the heart of anything. However, I'm also sceptical of idea that all truth must be simple and elegant, and I think actually that the value of Alexander et al's description is it's precision, rather than its elegance.

Brian

6/28/2017 12:42:08 pm

I really like this definition. The first sentence defines learning but I believe restricting the definition to "person or persons" is incorrect for me. Clearly human beings are not the only objects in the universe with the capability to learn, are they?
Surely we can learn a lot from learning in animals, plants and all forms of artificial intelligence. Any definition I would have thought needs to encompass every existent.

Second sentence is for me an elegant elaboration of the learning transaction (Dunkin and Biddle) which I always find persuasive.

Gregg Collins

6/26/2017 04:59:00 pm

If I don't get too balled up in questions like "what does 'reciprocally' mean in this context" and "does 'affordance' mean anything if you describe all of human activity as 'responding to affordances'?" this seems to cover everything I could possibly ever imagine describing as "learning." Which to me makes the author's point--what's the use of a definition that doesn't rule anything out?

Anneke Smits

6/28/2017 03:50:21 am

Thank you for this clarification of the dynamics of the discussion between scientists who dare to define learning. Lately I find myself trying to clarify what I think 'learning' is (on the basis of Alexander, Didau, Nuthall and Zull) and falling in some of the same pitfalls, especially with research colleagues. Nevertheless my attempts to talk about 'what is learning' have brought about a positive change in the way my students see their own teaching and design their own lesson sequences. What worried me is that before I made this attempt my students (who are experienced teachers) seemed to think a lot about teaching and testing and not so much about learning. I would not dare to share the definition that I use here, and I know it is as imperfect as any other definition. But I would like to argue that thinking about 'what is learning' with education students is of utmost importance. Especially in a time where learning in actual school practice seems to be more and more equivalent to teaching students to produce favourable test results.

Paul Cochrane

6/28/2017 06:35:16 am

Is learning not just:

-getting a mark? ;-)
-knowing something you didn't know before?
-being able to do something you couldn't do before?

Brian

6/28/2017 12:46:21 pm

How about if you could do it before, but now you can do it better (even if you don't know why)

Rick Nelson link

6/28/2017 08:21:45 pm

First, I don’t believe the characterization of the position of Sweller is accurate. He writes, “learning, which in the Behaviorist era was defined as a change in behavior, should now have been defined as a positive change in long-term memory.” (Tobias and Duffy, page 131).
And, “The function of learning is to store large amounts of information in long-term memory so that it can be brought effortlessly into working memory enabling us to function in a large variety of complex environments.” (T&D page 137)
Second, to what extent, Dr. W., is the storage of sensory elements of knowledge in the neurons of the brain’s long-term memory merely a theory, as opposed to being an observable and measurable scientific fact?

rick nelson link

6/28/2017 10:43:02 pm

In addition, from Clark, Sweller, Kirschner (Am. Educator 2012): “if nothing has been added to long-term memory, nothing has been learned.” I think that addresses the issue of Altzheimers. And is not a temporary change by definition not a change in long-term memory?
And from van Merrienboer and Sweller (Ed. Psych. Rev. 2005, page 155): “The aim of instruction should be to accumulate rapidly systematized, coherent knowledge in long-term memory.” Instruction (hopefully) is learning with expert guidance.

Dylan Smith

6/30/2017 10:14:01 am

For five hundred years now, communities of like-minded academics have been defining constructs so that experimental work in their field of study can be understood and replicated. A typical script quickly emerged: Advancing knowledge will lead to increasing specialization and a working language, and, as a result, even greater homogeneity within the academic field.

The benefits of defining constructs are obvious. If I use Skinner boxes to investigate some aspect of, say, selective attention in rats, others interested in replicating/extending my work will be able to read my write-up and note that I used appetitive reinforcement with “hungry” male juvenile Long-Evans rats, and that the construct of “hungry” had been operationalized as, e.g., “maintained at 85% of their ad lib free-feeding weight.”

The primary interest of educators in the term “learning” stems from the likelihood that greater understanding will further the ability to reproduce its antecedent conditions. But educators also suffer from an inconvenient reality: discussion communities are not as homogenized as those typically seen in academia. There’s much to be said about this, but suffice to say that educators like John Hattie understand the importance of flashing a presentation slide with an operationalization, however broad or imperfect, in order to gather all employee/stakeholder/interest groups in the audience-of-the-day.

But here is why I think Daniel’s blog post and the comments above are so important. For all the benefits of social media, scientists -- the so-called “experts” -- need to consider and account for the perils of wide open, popularized discussion on social media. When engaging a heterogeneous lay population, greater care must be taken to define what is and isn’t being discussed. The ongoing “learning styles” debate is a fine example of an irresponsible and avoidable communications fiasco. Science has never been free of powerful rhetoric, but when we see otherwise well-reputed scholars “naming and shaming” or otherwise angrily mocking viewpoints from other expertise areas, I worry where 21st c. Science is headed.

Michael Pye

7/8/2017 04:12:15 pm

Would you mind offering more detail on the last few sentences
.
The ongoing “learning styles” debate is a fine example of an irresponsible and avoidable communications fiasco. Science has never been free of powerful rhetoric, but when we see otherwise well-reputed scholars “naming and shaming” or otherwise angrily mocking viewpoints from other expertise areas, I worry where 21st c. Science is headed

Dylan Smith

7/11/2017 08:59:04 am

Apologies for my delay in replying, Michael... I’ve been occupied with pressing personal matters. Here is a GoogleDoc blurb providing more detail on my claim, respectfully (you may have to cut-and-paste): https://docs.google.com/document/d/1XaBEnyOgAG7nDVkjK_cYf6TVmQ2VUgmaysdy34_xXoU/edit?usp=sharing

Mario

7/6/2017 04:52:06 pm

Dear Professor Willingham,

Where can I read about the role of emotions regarding learning? Most of what I've read it's quite mechanistic, if the student follows the right techniques (retrieval, varied practice, interleaving, etc.) learning takes place. I find it hard to believe that emotions are not part of the process (expectations, self-steem, autonomy, willpower, etc.).

Thanks

Michael Pye

7/11/2017 12:18:23 pm

Thanks for that Dylan. I understand your points but I am still slightly confused on the argument you are making.

Do you agree that learning styles don't work but criticise the way in which they have been challenged or do you believe they have a role to play?

Dylan Smith

7/11/2017 03:25:18 pm

I can understand your confusion. It is fair to say my response to your question was an argument, and I’ll try to clarify. I equate learning with processing skill. I believe young children develop skill and wherewithal for processing unimodal inputs. In an educational setting, teachers promote learning and wellness by providing for modality choices/pathways in learning activities. This includes allowing a child to exclude a particular modality that he/she feels is likely to be ill-suited for learning that task on that day.

Put another way: A young student at a particular point in development is often likely to have learning-related reasons (beyond what have been called “preferences”) for involving or excluding a particular sensory modality on a particular learning task. Later, as the child develops, he/she becomes more and more skilled at multisensory integration. (We continue to process unimodal inputs, mind you, and even in other cortical areas, i.e., non-heteromodal areas. Meanwhile, as multisensory integration improves, the cooperative efforts provide some big pay-offs, e.g., shorter reaction times. But it’s also true to say, I believe, that modalities often “cooperate” in order to work in clever opposition.)

You may correctly infer that I don’t “believe in learning styles” per se, but that I think scientists jumped the gun in dismissing the developmental importance of task-specific modality choices in a youngster's "learning."

Brian

7/12/2017 06:24:32 am

Dylan. Like others I much appreciate your replies. The last paragraph for me sums the whole issue up perfectly. Are there any sources you could recommend that I read around "developmental importance of task-specific modality choices in a youngster's "learning." .

There will I am sure be a wide range of topics/sources which illuminate the area but hopefully there will be some work on the specific issue.

I will use this last paragraph when discussing the issue if the is ok and attribute to you every time.

Thank you.

Brian

Dylan Smith

7/12/2017 01:11:52 pm

To Brian:

Brian, thank you so much for your reply, to which I seem unable to directly reply... Hope you see this.

I’m afraid I am unaware of readings that specifically address the topic. I’m not sure of your background, but will tell you that my own views tend to be triangulations drawn from quite diverse research readings.

Having said that, there have been numerous publications in the last decade on the nature of tasks -- much more narrowly defined -- and task boundaries, expectations, goals, instructions, top-down influences, etc. Again, depending on your own background and purposes, you might find something useful in two papers I’ve found helpful, one a study, the other a thorough position paper:

Freedberg, et al. (2014). Incidental learning and task boundaries.

Ten Oever, s., et al. (2016). The COGs (Context-Object-Goals) in multisensory processing.

I’m sorry I can’t help you more on that, and please feel free to quote anything I have written above.

Michael Pye

7/12/2017 05:45:48 am

That is much clearer. Thank you for taking your time to reply.

If I have understood your idea correctly two key bottlenecks to supporting and utilising this idea stood out. How do you define and measure these different modalities especially if they change frequently, which I believe you implied? Secondary how are these modalities to be reliably and accurately utilised to improve teaching? There is a similar problem in excercise were we know people respond differently but have little in the way of clear diagnostic tools to differentiate approach. In this wake opinion, guesswork and plain psudoscience prevail.

Dylan Smith

7/12/2017 12:26:29 pm

You ask very good questions, Michael. In my own view, most educators do not find much interest or utility in “learning” at this grain size. Understandably, we are more typically interested in schoolhouse learning and achievement that is measured off-line, and promoted with impactful strategies such as those that Mr. Hattie, armed with Cohen’s d, has helped the profession to sort and understand.

Neuroscientists, on the other hand, are interested in the fundamental nature of “learning” defined as a change in neural processing. For some time, these researchers have operationalized and measured learning online using fMRI and ANOVA. This work will need to continue for at least a decade before the unique roles of modalities in learning are more clearly distinguished and understood -- not with certainty, but confident consensus. A personal opinion: My gut says it will require years of very clever research designs/paradigms that can eventually provide greater clarity on the oscillation hierarchy.

In the classroom, to address your second question, many teachers already address issues related to modality choices with understanding, flexibility, and universal design for learning (UDL).

Understanding goes a long way when Jane, aged 6 years and 2 months, doesn’t seem to hear you and continues to struggle with her scissoring, jaw moving back and forth, even though you were standing immediately alongside when you issued the instruction that everyone put their scissoring away.

Flexibility goes a long way when Robbie, aged 10, approaches you to plead that he be allowed to exercise or demonstrate his learning in a unique way.

UDL goes a long way in providing all students with the necessary range and autonomy to make intuitive choices in the interest of their learning and development.

On the Definition of Learning....

Leave a Reply.

Purpose

Archives

Categories