Hi,

me and a friend of mine had a brain-related discussion that we hope you can help us out with.

I argued, since I can visually "compose" a scene from my imagination, and answer color- and geometry-related questions about it, the visual center must necessarily hold a "rendered" version of that scene, i.e. in some preprocessed format but still containing the actual information needed to theoretically read it out and render it on a computer screen without accessing the memory.

He argued, that there's no point in actually composing the image since the visual center can process associations from the brain just fine, so it's enough to just hand the visual center the associations and let it analyze them on-demand - i.e. imagination is a hub of associations rather than a buffer of pseudo-image data.

So I guess the crucial question here is:

on "seeing", a scene is decomposed into objects, which are stored separately. Does this step happen before, in, or after the stage that handles imagination?

Thanks for your replies.

fMRI suggests that visual imagery involves the same visual association areas as visual perception

Whatever you focus on comes to the forefront of your mind, its not really like looking over a rendered scene that already exists like you would a picture. Its different when you're physically viewing a scene, and perhaps one day we will be able to record our lives in the first person. Viewing the brains 'hallucinations' is another matter though, as whatever you think about or focus on becomes the image. An experiment...in the next couple of seconds explain why you were thinking of a dancing Pink Elephant?!

Yes, but when I visualize a pink dancing elephant, I can actually gather geometric information about that (angles, relative lengths). Doesn't that imply that a rendered-out version of that elephant has to exist _somewhere_ around my visual center?