Jump to content

Demand type memory

From Wikiversity

Demand Type Memory

Theoretically, there are two ways that you can access memory, by content, and by place-code or address. Classically the whole concept of memory that we get from folk psychology is the place-code addressable form, in which you demand that the system give you the contents of a location, and it returns the contents. This idea has roots back in the hydraulic model of memory where memory was seen to be kept in a reservoir in the brain and addressed by opening valves to let it out.

The problem with demand memory as a model of how memory works, is that it doesn't adequately explain implicit memory, where memories are added to other memories, because they are seen to be related, but without the brain being aware that it even knows the information stored there. In demand memory you can't access a memory that you don't have an address for, so implicit memory lies outside the model.

Phenomenal memory where you can't isolate an address, as in the storage of memory in a neural network also lies outside the demand memory model. This has led some people to speculate that implicit memory and phenomenal memory might be the same thing. If so it would suggest that phenomenal memory is unavailable for demand/retrieval because we cannot know the address at which to look for a particular memory.