My Current research is focused on the organization of the primary
visual cortex (V1). V1 is the first station of call for visual information
reaching the cerebral cortex from the eyes. Despite its primacy in the
visual system, there remain many features of its anatomy that defy our
current understanding. Among these is one of the first discovered
anatomical features of V1, the ocular dominance column. Inputs from the two
eyes are segregated in V1 into an interdigitating mosaic of stripes that
tile the surface of the cortex. They are present in many higher species
with binocular vision, yet their precise function remains a tantalizing
puzzle. My most recent work has shed some light on the mechanisms that
generate and modify the pattern of ocular dominance columns during normal
visual experience.

I discovered that the pattern of blood vessels in the eye is mimicked in the appearance of ocular dominance columns. Further experiments demonstrated that the representation of the retinal vasculature was caused by anatomical re-wiring of brain connections induced by tiny shadows cast by the blood vessels. The shadows deprive small regions of the retina enough to influence normal activity dependant development. The process underlying this re-wiring is of particular importance because it is the same mechanism that causes blindness in the disease amblyopia. Amblyopia occurs when one eye is faced with a large scale deprivation, such as a congenital cataract. The deprived eye's ocular dominance columns shrink and good vision is lost irreversibly. Amblyopia causes blindness in the whole eye, whereas the shadows cast by retinal blood vessels cause narrow channels of blindness in both eyes (called angioscotomas). Our experiments show that amblyopia is the result of a normal developmental process that has been thrown out of balance by abnormal visual experience.

The discovery of a new phenomenon opens up a wealth of possible experiments and analyses. We have used the representation of retinal blood vessels in V1 to explore a number of avenues into the organization of the early visual system. For example, by matching blood vessels in retinal photographs to their representation in the brain, we were able to construct a complete and precise retinotopic map of V1. This provided us with enough data to answer a long standing controversy: Is every retinal ganglion cell allotted the same cortical surface area? Our unique data-set allowed us to set the record straight and demonstrate that more cortical territory was given to foveal retinal ganglion cell than those in the periphery.

While looking for the representation of retinal blood vessels in V1, we made another interesting discovery. Ocular dominance columns were present in some individual squirrel monkeys in our study, but absent in many others. This startling revelation forced us to the heretical conclusion that ocular dominance columns might not have any functional significance. After all, how can a structure be doing an important job if it is absent in 1/3 of the normal population? With hindsight, our conclusion is not so surprising. The function of ocular dominance columns has been sought for more than 30 years, yet no theory has even reached the level of general acceptance. Perhaps the most likely explanation is that we have been searching for a function that does not exist.