Every time I learn something new it pushes some old stuff out of my brain

So says Homer in one episode of The Simpsons. But maybe Homer actually "misunderestimated" (Copyright G W Bush) himself by falling for the metaphor of the mind as a container. It's a very tempting metaphor, because it seems to get across the way that we access and 'store' new information. Here's a thumbnail sketch of the idea from the essay 'Is there a problem in explaining cognitive progress?' by Aaron Ben-Ze'ev in Rethinking Knowledge: Reflections Across the Disciplines.

"In this view, the mind is an internal container, and cognitive progress is a quantitative increase in the amount of internal representations. In such a mechanistic paradigm, the cognitive system remains more or less stable, the only difference being that its empty shelves are gradually filled with more information. Cognitive progress is attained by adding a certain part to an existing system. When this mechanistic picture is applied to to the realm of scientific knowledge, science is conceived of as essentially taking pictures of the external world; the more pictures science has, the more adequate the science is. Hence there is always linear progression. Both the individual person and science as a whole are constantly marching toward a better understanding of their surroundings."

But is this really how our minds work? In the essay he suggests adopting a different approach - the schema paradigm. In this view the mind is made up of capacities and states. In the container metaphor, when you want to recall information presumably you send the little bloke in your head off into the information warehouse to retrieve what is required. In the schema paradigm, however, the container metaphor doesn't work because key elements - capacities and states - are retained rather than stored.

"The capacity to play the piano and the state of being beautiful are retained but not stored. Similarly, capabilities are not brought out of storage but are realised or actualised. The state of a car in motion is not stored in its engine when the car is stationary; rather the car has the capacity to repeat its state of being motion. And by the same token, when a squeaky toy does not actually squeak, it retains (rather than stores) its capacity to squeak."

This, he argues, also explains something about the organisation of the brain:

"In a storehouse, it makes very little difference how the items are disposed or organised. Something may be stored at the right or left side of the storehouse without being affected. However, in the schema paradigm, organisation is an essentia property, not a later addition. The importance of organisation and relations in memory can, for instance, explain that it is much harder to recall the months of the year in alphabetical order than in their chronologial sequence. A junkyard or tapre recorder model of memory is feasible and even natural in the container paradigm, whereas the schema paradigm stresses the importance of the relations and organisation among the carious items. Many phenomena indicating the sensitivity of memory to organisation attest to the greater suitability of the schema than the container paradigm for memory."

Notably George Lakoff (yes, him again) has been here already, and has identified the container metaphor as one the most prevalent. You can also see that it crops in other areas - what about set theory in maths for example? But as a way of understanding how we learn, perhaps it simply doesn't work. Sorry Homer.