Visual abilities decline during normal (non-pathological) aging. Many of these visual declines cannot be attributed to optical changes and must therefore be due to changes in the retina or central visual pathways. These include declines in visual acuity and spatial contrast sensitivity (especially under low luminance levels), suprathreshold contrast vision and contrast gain, temporal-frequency contrast sensitivity and resolution, spatial-temporal interactions, hyperacuity, binocular processing, and sensitivity to motion. Certain aspects of these vision deficits and comparisons with neurophysiological and lesion-behavior studies in monkeys suggest hypotheses about the nature and location (e.g. magnocellular vs parvocellular pathways, specific visual structures, and so on) of the neural deficits. Despite the well-documented psychophysical deficits, available anatomical studies in humans and monkeys suggest that aging has only relatively minor effects on the retino-geniculo-striate pathway. Retinal photoreceptor losses are relatively restricted to rods, and there is compensation among the remaining rods for those that are lost. Although some retinal ganglion cells appear to be lost, the loss is small relative to individual-to-individual variability. In addition, there appear to be no massive cell losses in the LGN or striate cortex. Physiological results in the monkey LGN suggest that the functional properties of LGN neurons, and therefore their retinal inputs, are not significantly affected by aging. Retinal pattern-evoked ERG studies in humans likewise suggest that the physiological properties of the retina are little affected by aging. Comparisons between pattern-evoked ERG and cortical evoked potentials in the same individuals suggest that some neural change occurs between the retina and striate cortex, but the location and nature of this change is not known. Thus, we are far from being able to answer the question, What are the neural bases of visual deficits during aging? There are several possible reasons for this: (1) The neurobiological methods that have been brought to bear on the question have been fairly limited. (2) Investigations of neural changes may not have been guided sufficiently by what is known about the psychophysical changes that occur with aging. (3) Existing studies may not have examined the correct locations in the visual system. (4) There is large individual-to-individual variability in the effects of aging and, with the small samples of individuals that typically are available in neural studies of aging, the variability could obscure detection of aging-related changes. Suggestions are offered for ways in which future research can solve these problems.(ABSTRACT TRUNCATED AT 400 WORDS)