Learning That Travels

That's the quiet failure sitting at the centre of most studying. We check whether material comes back when we ask for it in the way we practised asking, and we almost never check whether it comes back when the question looks unfamiliar. Which is a problem, because the unfamiliar question is the only one that counts. Life doesn't ask you things in the shape you learned them, and neither does any real exam paper.

The name for the thing you actually want is transfer, meaning being able to use what you learned on a problem you've genuinely never seen. Not repeating it. Using it. And everything else, the highlighting, the rereading, the deck you can rattle off in order, is only worth anything if it eventually travels.

Why memorised knowledge stays at home

When you learn something by heart, you usually learn more than the idea. You learn the idea plus all the wallpaper around it: the exact phrasing, the order of the steps, the diagram that sat next to it, the worked example it arrived in. Your memory ties the knowledge to that particular setting, so when the setting changes, the thread you pull has nothing on the end of it.

Psychologists call this being bound to the surface, meaning stuck to the outward appearance of a thing rather than the idea running underneath it. You learned that photosynthesis turns light into sugar as a sentence, so you can produce the sentence on demand. Put a graph of plant growth under coloured lamps in front of you and the sentence is no help at all, because you never actually learned the why beneath it. The knowledge got shaped like the question you practised, and this question is a different shape.

Which is why cramming feels so productive and delivers so little. It's superb at making knowledge available in the precise form you rehearsed it, and close to useless the second that form shifts at all.

Near and far, the two distances an idea can cross

Not all transfer is equally hard, and naming the two ends of the range helps (Barnett and Ceci, 2002).

Near transfer is a short hop. The new problem looks a lot like the practised one, same topic, same shape, a few numbers swapped out. Drill twelve quadratic equations and a thirteenth is near transfer. Useful, but easy, and easy to fake as well, because pattern-matching against something familiar will carry you over a gap that small without much real understanding underneath.

Far transfer is the long jump. The idea turns up somewhere genuinely new: a different subject, a real situation, a question that deliberately hides which tool it wants from you. Using something from economics to make sense of a queue at a hospital is far transfer. It's the hard and valuable kind, and it's the kind that makes knowledge feel like it's yours rather than something you rented for the duration of an exam.

And here's the humbling bit the research keeps turning up, which is that far transfer is rare. People are considerably worse at it than we assume, and they don't do it unprompted simply because they once knew the material. In the classic study people were given a story whose solution held the exact key to a puzzle they were then handed, and most of them didn't use it, because they never noticed the two things shared an underlying structure (Gick and Holyoak, 1983). The knowledge was sitting right there. It just wouldn't cross the surface difference.

Analogies work, but only if you saw the idea rather than the picture

So how does an idea ever cross a long gap? Usually through an analogy, taking the shape of one thing and laying it over another. Electricity flows like water through pipes. The atom is a bit like a tiny solar system. A good analogy is a bridge: it lets a familiar shape guide you through an unfamiliar one.

But analogies only carry you if you got hold of the structure underneath rather than the picture sitting on top. Gick and Holyoak found people transferred a solution far more reliably once they'd seen the same idea in two different stories and could pull out what the two had in common, meaning the shared skeleton with the surface stripped off both.

One example teaches you the wallpaper. Two examples of the same idea in different clothes teach you the idea. That's the whole trick, and it's worth stating plainly: you build transfer by meeting the same idea in more than one form, until you can see what stays constant while everything on the surface changes.

What actually builds knowledge that travels

Pull the research together and it points one direction (Bransford, Brown and Cocking, 2000; Perkins and Salomon, 1992). Transfer isn't a talent some people are born holding. It gets built, and four things build it.

• Understanding the why and not only the what. You can't carry a rule somewhere new if all you have is the rule, because the reason it works is the handle you carry it by.
• Meeting the idea in varied forms. The same thing reworded, redressed, coming at you from different angles, until your memory gives up on tying it to any one surface.
• Using it rather than recognising it. Applying an idea to build something forces you to reconstruct it, which is a far deeper act than nodding along to it.
• Being tested on situations you haven't seen. Not the practised question again but a fresh one, where you have to notice which idea applies before you can apply anything. That noticing is most of the actual skill.

All four take effort, and that isn't a flaw in the approach, it is the approach. The strain of pulling an idea out of its comfortable setting and forcing it somewhere else is precisely what loosens it from the surface, which is why difficulty is the point rather than an obstacle to route around (Bjork and Bjork, 2011).

How Norudit is built for travel

Worth being honest here, because transfer is easy to promise and hard to earn. Norudit isn't built to help you recite things. It's built, at nearly every stage, to loosen ideas off the surface they were learned on so they can move.

The Feynman gate comes first, and everything else stands on it. To pass it you explain a topic simply, in your own words, and you can't do that from memorised phrasing, because your own words are themselves a change of surface. If all you have is the wallpaper, the gate exposes that straight away. Passing means you found the idea underneath, which is exactly what transfer runs on.

Concept-varied review is the meet-it-in-varied-forms principle built directly into the deck. Every card comes back as a fresh version of the same idea, so a fact returns as a question that makes you use it and a concept returns reworded. You never get the chance to memorise the card, because the card refuses to sit still.

The debate phase makes you hold an idea from a side you didn't rehearse, arguing a position, answering objections, using the knowledge under pressure instead of reciting it in peace. Transfer practice under a different name.

The capstone is the most direct of all of them, an open far-transfer test that hands you a problem the topic never showed you and asks you to apply the whole thing to it. An exam checks whether the material came back. The capstone checks whether you can take it somewhere new.

And the exam engine reaches into your review deck, finds whichever ideas you're weakest on, and rebuilds them into fresh situations, so a mock is never simply the practised question coming round again. It's your shakiest ideas in clothes you haven't seen before, asking to be recognised.

None of that makes studying easier. It's a sequence built specifically so that what you learn doesn't stay in the room where you learned it, which was the point of studying in the first place.

Start your first class →

Sources

• Barnett, S.M. and Ceci, S.J. (2002) 'When and where do we apply what we learn? A taxonomy for far transfer', Psychological Bulletin, 128(4), pp. 612-637.
• Bjork, E.L. and Bjork, R.A. (2011) 'Making things hard on yourself, but in a good way: creating desirable difficulties to enhance learning', in Gernsbacher, M.A. et al. (eds.) Psychology and the Real World. New York: Worth, pp. 56-64.
• Bransford, J.D., Brown, A.L. and Cocking, R.R. (eds.) (2000) How People Learn: Brain, Mind, Experience, and School. Expanded edn. Washington, DC: National Academy Press.
• Gick, M.L. and Holyoak, K.J. (1983) 'Schema induction and analogical transfer', Cognitive Psychology, 15(1), pp. 1-38.
• Perkins, D.N. and Salomon, G. (1992) 'Transfer of learning', in International Encyclopedia of Education. 2nd edn. Oxford: Pergamon Press.

Read more

• Why explaining it simply proves you understood it: The Test You Can't Fake
• Concept-varied review, and why the card never sits still: The Deck That Rewrites Itself
• Inside the exam engine, five question types and real mark schemes: How Norudit Examines
• Why a private tutor puts an average student in the top 2%: The 2 Sigma Problem