This is chapter 1 of the textbook Understanding Vision: theory, models, and data Oxford University Press, 2014.
Abstract: This book's approach to understand vision is by linking neural substrates in the brain with visual perceptual behavior through the use of theory and models. This chapter proposes to view vision as containing three roughly successive stages: visual input encoding, visual information selection by attention, and visual recognition or perception by visual decoding. In contrast to traditional views, which decompose vision to low-level, mid-level, and high-level processes, the proposed three-stage view makes explicit the separate tasks in vision and highlights the role of vision selection, which admits a tiny fraction of visual input information through the attentional bottleneck for visual decoding. This chapter previews the topics in each of the three stages that will be covered in the rest of the book, noting that we have the least understanding about the visual decoding stage.
Keywords: visual encoding, selection, decoding, attentional bottleneck, visual perceptual behavior
Its figures in a pptx file