I can pull specific, full and quality images and "scenes" with related emotions from memory. When prompted to answer the question, it's constructing the image

I can pull specific, full and quality images and "scenes" with related emotions from memory. When prompted to answer the question, it's constructing the image on the fly. A moving image that is partially created and then only remembered if I put energy into remembering it. The fidelty is one step lower than full cinematic lush glory, and above all the other answers.