3D Direttissima: A V1 Heresy

The heretical conjecture offered here is that a 3-space scene reconstruction of sequences of visual fixations is constructed directly in V1 in a head-centered frame. "Direttissima" is a climbing term designating the shortest, most direct route to the objective of a climb. The "heresy" here is the claim that the construction of the first 3D representation of the scene before ones eyes takes place in V1 rather than further up the visual pathways. The inputs to the neural process proposed for such construction are the V1 retinotopic projections from each eye and the vergence angle. Absolute disparities are not encoded as a separate data set in a distinct set of neurons, but rather are emergent properties of the neurons that implement the construction of the 3D representation of one or a sequence of fixations on elements of the scene. In this conjecture, the conscious "picture" we take as vision is a projection of the 3D construction, not a view of the retinal image. One of the critical psychophysical clues to the nature of such a mechanism is the well known visual size constancy phenomenon. Another property of the proposed mechanism is that it is plausibly fast enough to enable such spatially and temporally demanding tasks as a human batting a fastball or a bobcat swatting the head of a striking rattlesnake.