I think there is a confusion that visualisation must be photo-realistic. I can visualise the entire city in which I live in my head, and fly around its streets,

I think there is a confusion that visualisation must be photo-realistic. I can visualise the entire city in which I live in my head, and fly around its streets, parks, and landmarks in a handful of seconds. The map is fully accurate and the broad-brush image is. But if I stop and focus on any one detail, I cannot always remember exactly what lies there. Clearly my memory depends on filling this in 'as-if' to sustain my vision, and I think this is true in general. We don't have memories that can zoom in to ever greater microscopic detail, and we don't have every detail of a picture, but this doesn't stop us from imagining. It's similar to the way in which when we're presented with words that are jumbled up apart from their first and last letter, our brain fills in what's happening. Vision in general is not just the mechanical encoding of sense data from light. It is as much psychological as physiological. We see the world in stable colours and contrasts because our mind controls for illumination and processes the colour of one detail relative to the whole picture. I think our mind's eye depends on similar kinds of psychological support.