How can teachers promote skills transfer?

According to the dictionary, “transfer” means “to apply skills or knowledge learned within one domain to a novel, oftentimes completely different, domain”. It should also say, “that thing nobody does well”.

Wherever teachers sit on the ideological spectrum of teaching, the vast majority would argue that “transfer” is a (some would even argue the) primary goal of education. Although it’s all well and good for students to perform strongly on assignments and exams, most educators are interested in ensuring that students can apply their learning beyond the classroom in order to flourish within broader familial, social and economic realms.

To this end, many curricular adjustments have been touted in order to address transfer. For instance, the modern focus on generic 21st-century competencies (1) - critical thinking, creativity and collaboration - has almost exclusively been justified by referencing the need for students to develop a set of all-purpose skills readily transferable to the rapidly changing and largely unpredictable world.

But, unfortunately, by any metric, transfer remains frustratingly out of reach. As most teachers can attest, students continue to struggle mightily with transferring even seemingly simple skills between nearly identical domains (for instance, moving arithmetic procedures (2) from numerical to word problems).

Similarly, industry is complaining ever louder about the underprepared nature of recent graduates (data suggests that more than 50 per cent of new hires don’t pan out (3), with successful employees requiring an average of 16 months (4) to begin effectively demonstrating those generic 21st-century competencies promised by modern curricula).

So, why is it that a skill set clearly exhibited in one domain often goes missing in action when transitioning into a new domain?

This phenomenon isn’t all that surprising. In fact, it is quite understandable.

It turns out that there are three main impediments to successful transfer: knowledge, contextualisation and adaptation. Let’s take a look at each in turn.

1. Knowledge

Chances are, within your particular field, you demonstrate strong critical-thinking skills. If you’re a science teacher, you can deconstruct and challenge fringe hypotheses. If you’re a school leader, you can clarify and integrate multiple pedagogical frameworks. If you’re a school counsellor, you can differentiate and assess subjective symptoms versus objective signs. Recognising that you already possess critical-thinking skills (and likely apply them every day), I want you to take a crack at the following problem:

“Recently, research was published suggesting that neutrinos within the medial temporal lobe, rotating in a clockwise fashion, were able to generate a radio signal across the frequency bands detectable by BOLD recordings. Thinking critically, what methodological error likely led to this result - or, if this result is accurate, what does this suggest about the historical basis of fMRI analysis?”

Uh-oh! Unless you’re a neuroscientist or an engineer, I’m guessing you weren’t able to make head or tail of this scenario. It turns out that all the critical-thinking skills in the world become obsolete when you try to apply them to information you don’t understand.

Unfortunately, within human beings, there is no such thing as a “standalone competency” that, once mastered, can be applied to any and every situation. Instead, our skills are intimately tied to the facts we must apply those skills to. Put simply, knowledge precedes skills (5).

As a simple example, students classified as either “strong” or “weak” readers (according to standardised tests) were asked to read a passage about baseball (6). Importantly, some of these students knew a lot about that sport while others knew very little.

When quizzed about what they understood from the reading, here’s what happened: those with good baseball knowledge performed best on the quiz whether or not they were strong or weak readers (though the strong readers did slightly better); those with poor baseball knowledge performed worse on the test whether or not they were strong or weak readers (though, again, the strong readers did slightly better). Basically, reading skill had limited bearing on performance: rather, having knowledge of baseball facts determined whether or not students were able to apply their reading skill.

Similarly, the reason you couldn’t make sense of the BOLD/fMRI scenario above is not because you lack the skill of critical thinking - it’s because you lack deep knowledge of the relevant facts.

Interestingly, it’s not enough to simply have access to facts (7). Right now, you could google the definitions of “BOLD”, “fMRI”, “neutrino” and “radio signal”, but chances are that it would get you no closer to applying your critical-thinking skills to this scenario. Rather, in order to apply your skills, you must deeply comprehend and internalise the relevant information.

For this reason, regardless of strength, depth or prior demonstration of skill, individuals will be unable to enact skills until they obtain sufficient factual knowledge relevant to the domain they need to apply those skills within.

2. Contextualisation

It would be wonderful if the definition and application of skills were identical across every situation. Unfortunately, it takes only a moment of thought to recognise that the same skill can have a significantly different definition when applied to different scenarios (8). Take collaboration, for example:

• When performing surgery, collaboration requires individuals to apply their respective expertise in order to achieve a specific outcome. This often requires assertive communication with little regard for open debate.
• During academic research, collaboration requires individuals to question their respective expertise in order to assess a hypothesised outcome. This often requires open debate with little regard for concession.
• When designing a new product, collaboration requires individuals to develop a shared expertise with no clearly defined outcome. This often requires concession with little regard for assertive communication.

The fact that definitions and applications of skills vary across contexts means there can be no all-purpose competency directly applicable across every domain. Rather, skills must be characterised according to the demands of divergent fields. Put simply, skills must be contextualised.

For this reason, after embodying the facts relevant to a particular domain, individuals must synthesise this knowledge in order to effectively characterise the skill set relevant to those facts. Without contextualisation, no (or incorrect) skills will emerge.

3. Adaptation

In a seminal experiment from the 1980s, William Chase and Anders Ericsson asked a pair of students to spend three to five hours per week simply listening to, memorising and repeating back random strings of digits (9). These researchers wanted to determine if otherwise normal human beings could develop savant-like memorisation skills.

After two years (yes, two years!) of continual practice, here’s what happened: whereas both students started as expected (able to memorise and repeat back about seven digits), over time they were able to improve dramatically. In fact, by the end of the experiment, each could easily take in and verbally recite more than 70 digits.

From this, we learn a very important fact: with deliberate effort and practice, any human being can come to master any skill (within physical and/or cognitive constraints, of course).

But, importantly, this was not the end of the study. After the participants grew exhausted and decided to quit, each was asked to play one final game. As before, they would have to listen to, memorise and repeat back a string of information - except, this time, instead of using random digits, the researchers would use random letters (10).

Given their new, powerful memorisation skills, this simple switch from numbers to consonants should have been no problem at all, right?

In fact, both individuals dropped right back to normal and were able to memorise and repeat back about seven letters.

So, what happened? Clearly these individuals possessed a skill (memorisation); clearly they had the relevant knowledge (deep comprehension of letters); and clearly they understood the context (take in and verbally recite letters). Why did their performance decline so dramatically?

It turns out that human beings possess a unique, dual-system learning apparatus that swings between active engagement and passive automaticity (11).

Whenever we first learn a new skill (be it cognitive or behavioural), we must consciously and effortfully engage with it. For instance, think back to when you first learned how to drive a car; chances are it was overwhelming and exhausting.

However, the more we practise a particular skill, the more we can subconsciously and effortlessly perform it. Think back to your last commute; you probably remember almost nothing about it. Since you’ve mastered driving, your brain can run this skill on autopilot with little to no input from you.

Here’s the important bit: once we have automated a skill within a specific context, it becomes incredibly difficult to consciously access and adjust. Returning to driving, I learned how to drive in the US, where the gearstick sits on the right. A decade ago, I moved to Australia, where the gearstick sits on the left. This should have been an incredibly simple skill adjustment but, if you’ve ever driven a car in a foreign country, you already know how difficult this proved to be. Put simply, automaticity kills adaptation (12).

The inability to access, adjust and adapt deeply automated skills within new contexts is why most athletes are typically only great at a single sport, why most musicians are typically only truly great at a single instrument, and why many individuals who have honed their skills within a classroom may find it difficult to demonstrate deeply rooted competencies at home or in the workplace.

The only freely transferable skill

In summary, the prerequisites to effectively transfer a skill to a novel scenario are knowledge, contextualisation and adaptation. You must first embody relevant facts, synthesise these facts to characterise relevant skill parameters and then adjust your skills to match these parameters.

If this process sounds familiar, it’s because it has a much more common name: learning.

Unlike other skills that change and morph across contexts, the process of learning appears to be largely acontextual (13). This makes sense, seeing as how learning is the process that allows us to comprehend varied contexts and acclimatise to novel circumstances. It doesn’t matter whether you desire to transfer skills into law, medicine or business - the same three-step process (knowledge, contextualisation, adaptation) will always prove mandatory. This means that learning is the only freely transferable skill we know (14).

Importantly, those who know how to learn effectively and efficiently will be the first to determine relevant knowledge, define relevant skill sets and adapt relevant standards within emerging fields.

In other words, the best learners will set the bar for competencies (including critical thinking, creativity and collaboration) across novel circumstances in the future.

Can pedagogies promote transfer?

Decades of research into the transfer problem have led to very few steadfast suggestions. Although various pedagogies (problem-based learning (15), small-group learning (16), dialogic discussion (17), etc) have demonstrated some impact, these effects are sporadic and highly dependent upon the context that the particular pedagogy is applied within (for instance, problem-based learning only appears to impact transfer when implemented after content learning (18) and when using design- rather than dilemma-based problems (15)).

So, what advice is there? Pool the research, and there are only two consistent messages: be explicit; and be practised.

As noted above, as we hone skills within a particular context, biological mechanisms serve to automate these skills and remove them from conscious reflection. This process frees up cognitive resources, thereby allowing us to learn new and different skills. However, in order to effectively transfer a skill, it must be brought into conscious awareness so that the process of adaptation can take place.

Herein lies the first message: be explicit. If we want students to demonstrate transfer, we must make the process of transfer explicit and continually reference this process (19) when tackling novel domains. Employing questions such as “Where have you seen this material before?”, “How is this similar/different to what you learned last week?”, “What previously learned skills do you think might be relevant here?” and “How do you need to tweak your skills to make them effective here?” will help to keep transfer in mind and facilitate the accessing and adaptation of skills.

Unfortunately, even with explicit focus, automated skills have an annoying tendency to reconstitute and override attempted adaptations. For instance, during a particularly stressful period earlier this year, I kept accidentally getting into the wrong side of the car. Despite having lived in Australia for nearly 10 years, my deeply rooted US driving skills continue to hold tremendous sway over my performance.

Herein lies the second message: be practised. Oftentimes, an individual will demonstrate effective transfer one day, only to revert back to confusion and uncertainty the next. Just like everything else, skill adaptation must be repeated multiple times (20) within specific contexts in order to begin constructing deep, reliable, useable memories for this process. It’s an unfortunate fact of human learning that transfer does not happen automatically. Read that again because it’s incredibly important: transfer does not happen automatically.

And the trick is to recognise that the answer to transfer does not exist in any particular curriculum, pedagogy or programme.

Rather, the answer lies in understanding that transfer must be made explicit and that it (like any other process) requires repeated exposure and practice across each domain. This means, as teachers, that it’s not enough to simply wish for a Trojan Horse-like transfer into our work - rather, we must clearly embed transfer within our course progression and unambiguously alert students to its presence and purpose.

Once we recognise the process of transfer and come to see it as integral to our teaching, then issues of horizontal alignment within year levels begin to make a lot of sense. More specifically, when transfer is made explicit, teachers across maths, English, sociology and all other year-level subjects have reason to join forces and build lessons under the same conceptual themes. This will ensure there is plenty of opportunity to develop, practise and adapt similar skill sets across varied domains and contexts.

Recognition of the transfer process further paves the way for vertical alignment between year levels. Again, when transfer is made explicit, different teachers working across year levels to evolve compound skills from simple to complex (such as persuasive essay writing) now have reason to communicate and pool resources. Again, this will ensure there is accurate reference back to previously learned skills in order to ensure effective adaptation and practice.

In the end, generic skills are important and will be integral to success in our rapidly changing world, but they remain highly contextual and cannot be counted on to appear in novel domains until targeted and specific learning has occurred.

As such, making the learning process explicit and concretely helping students to recognise the process of transfer is arguably the only way to help them have agency over their own progression and move their skills beyond the classroom.

Jared Cooney Horvath is a neuroscientist, educator and author, and is director of the Science of Learning Group. He is an honorary research fellow at St Vincent’s Hospital (an arm of the University of Melbourne’s Medical School) and the Melbourne Graduate School of Education. He is the author of Stop Talking, Start Influencing: 12 insights from brain science to make your message stick. Read the Tes review at bit.ly/Tes_Horvath

This article originally appeared in the 13 September 2019 issue under the headline “Failed transfer”