How We Learn

We don’t remember information or events that we experience. Instead, we remember how those things made us feel, and we use those reactions to reconstruct them.

This is the central claim made by the Affective Context Model, a framework to explain how we remember, learn – and to some extent how we think.

As a theory, it has unrivalled explanatory and predictive power. It explains why memory is notoriously unreliable, as well as the peculiarities of your everyday recollections. It enables us to design education more effectively, and understand why it doesn’t work today.

For organisations tying to improve the efficiency and effectiveness of their learning programmes it makes possible a step-change in outcomes as they re-think the way they design, develop and deploy programmes. (These new techniques are outlined in part1, part2 and part3 of ‘Disrupting Learning’).

The expression ‘neurons that fire together, wire together’ is fairly well known. What is less well known is what all that firing and wiring is doing. In the context of memory we often assume that neurons somehow ‘store information’ about events for later retrieval – like a somewhat unreliable computer. This is not true. Instead, our neurons encode our subtle affective reactions – ‘micro-reactions’ if you like – to the things that we experience. This affective context can then be used to reconstruct an event. Your reactions to events are as rich and as subtle as your visual experiences. They develop in a way that is unique to you – so that two people experiencing the same event will react differently - and therefore remember it differently. Rather than encoding the semantic features of your experiences, your brain registers the imprint in your emotional life: a bit like a footprint in the sand.

This explains why for example you may struggle to remember a boring drive to work on any given day, but if you have a car accident you will be able to reconstruct events in some detail. It explains why you are unlikely to remember much of your high-school algebra or geography lessons – and far more likely to remember school dinners, friends, embarrassments and enthusiastic or intimidating teachers. It even suggests why people who are depressed recall the things that they do – and struggle to recollect them when they recover. It explains why when people chat, they like to tell stories.

When, in 1974, Elizabeth Loftus found that her participants would remember events differently depending on whether the word used to describe the car accident was ‘bump’ or ‘smash’ she concluded that the words were altering the memories. This is not quite correct: the memories were never 'encoded' (in the conventional sense). Instead, participants’ reactions were stored and the confabulated ‘memories’ constructed from these, together with affective elements from the current context (such as the words used to describe the event).

In layman’s terms, if you have a boring day you won’t remember much of it. On the other hand, sometimes we imagine that a story we heard was actually an event from our own lives. The line between so-called 'episodic' and 'semantic' memory is blurred.

That we have persisted for so long in ignoring the obvious in our pursuit of the theoretical is a remarkable testament to the fallibility of science, which is forever prone to finding ways to reinforce its own affective biases: in this case the desire to find ourselves rational rather than emotional beings. The legacy has been a corrupting notion of learning as 'knowledge transfer' and decades of misguided attempts to force people to remember stuff. It is a reminder that while science may enlighten, it may also obscure.

As Doris Lessing once said "Think wrongly, if you please, but in all cases think for yourself."