My guess would be that we associate memories (especially memories of objects) with specific locations and can remember them better when we're in those locations. One, easily testable, prediction of this theory is that returning to the original room will make the memory more easily accessible.
One, easily testable, prediction of this theory is that returning to the original room will make the memory more easily accessible.
They tested it in a recent paper.pdf).
In Experiment 3, the original encoding context was reinstated by having a person return to the original room in which objects were first encoded. However, inconsistent with an encoding specificity account, memory did not improve by reinstating this context.
http://www.springerlink.com/content/m6lq80675m22232h/
There's probably some deep implications to this I'm not qualified to plumb. But next time I'm concentrating on something, and need to get up from the computer and walk around a bit, I'm going to try avoiding doorways.