Learning: the science won't help you, if you don't know how it works.

There's a common misconception of science, namely that scientific experiments are used to figure out how the world works. Actually, scientific experiments are used to reject theories about how the world works. Where do the theories come from? That's an interesting question.

Today, in learning, I often wonder which is worse: the science or the crackpots. Often the two go hand in hand.

Here's an example of what I'm talking about: imagine I put together an experiment in which some people lick an object and some don't. It turns out that if you spend ten minutes licking a BMW 3-series you remember much more about it than if you merely spent an equivalent period in the same garage. Perfectly true. Hey presto the 'licking theory of learning' is born: ‘The basic mechanism of learning is Licking, according to research.’ I can't wait to get stuck into the instructional design.

Bizarre as this may sound this is precisely what routinely happens in learning: take Atkinson & Schiffrin's multi-store model for example. They found that people who repeated words were more likely to remember them that people who didn't and based on this they proposed that the basic mechanism for memory (a prerequisite for learning) is repetition. Repeating things transfers them from short to long-term memory. Utter tosh. Never mind that most things people learn they learn from single experiences, never mind that many things (such as school hymns and pesky phone numbers) prove near immune to repetition. The list is long. Learning styles - really? Odd how most people seem to cope just fine with Google or YouTube (or university), and that the vast majority of learning is 'on the job'. Learning styles are no more real than 'digestion styles': sure, I prefer chocolate ice-cream over strawberry, but we both digest it in pretty much the same way. And it’s not just stuff from the 70s – I read something equally ridiculous about ‘multi-modal learning’ published recently, based on some perfectly respectable science around Working Memory.

Today there are otherwise capable people mired in approaches with experimental support but no basis in reality. Learning: if you don't figure it out, the science may simply mislead you.

Judith Harris's book 'The Nurture Assumption' is a good example. At the heart of her book is the argument that there is no shortage of research into child rearing – which has all been used to support the wrong theory. Turns out people learn from peers, not parents according to Judith. But for years the research was used to support the contrary view.

In defense of science, there's some better stuff around now that there was way back when our learning theories were first being developed. Now, we have the likes of Damasio, Kahneman and the beginnings of an understanding of how the brain works. But the people responsible for developing theories of learning are us - the community of learning professionals.

Theory is a product of deep engagement; working with a thing, working with people who are working with a thing, thinking about it, discussing it, reading relevant material – some of which will be research - and most of all noticing when things aren't working rather than convincing oneself that they are. This kind of messy, ‘system1’ expertise then builds to the point where we have a good intuition as to how something works – an intuition that we can then put to the test scientifically. That is how science works.

Otherwise we are just pouring coffee-grounds into a diesel engine and declaring it to be a cappuccino-maker.