Title: Paradiso M. Perceptual and neuronal correspondence in primary visual cortex.
Abstract: Paradiso M. Perceptual and neuronal correspondence in primary visual cortex. Curr Opin Neurbiol 2002;12:155–61. According to most models, processing of visual information within either the dorsal (motion) pathway or ventral (form) pathway is serial. From V1, information is sent to increasingly complex visual areas and, at some point upstream of V1, perception is achieved. This review article summarizes several lines of data that support the concept that primary visual cortex plays an important role in visual perception. The first type of evidence comes from experiments measuring response latencies in various visual areas. Strictly serial processing would predict that response latencies are greater in higher-order visual regions. Although this holds true for the earliest responses in V1 versus the earliest responses in upstream areas of the form pathway, the scatter of onset latencies in the form pathway would allow for some neurons in higher-order areas to be activated simultaneously with neurons in V1. In the motion pathway, the earliest onset latencies in areas upstream of V1, such as MT, are identical to those in V1. Second, some properties of primary visual cortex neurons argue for their role in perception. V1 has the highest resolution retinotopic map of any visual area and the magnification factor for V1 corresponds well with known visual acuities at varying eccentricities across the visual field. Thus, primary visual cortex may play a role in the perception of high acuity vision. The activity of some neurons in V1 is related to perception during binocular rivalry. The relative number of these neurons found in V1 is small as compared with higher order areas, but the absolute number may be higher as V1 is the largest visual cortical area. Mental imagery is the perception of images without retinal stimulation and is thought to be a higher order visual function. However, mental imagery of gratings has been shown to activate primary visual cortex in PET studies. Furthermore, single neuron recordings from V1 have shown that stimuli lying outside the classic receptive field can influence the response to stimuli within the classic receptive field in an orientation specific manner. Similarly, when a human subject viewing a stimulus is asked to imagine previously viewed flanking stimuli of a specific orientation, contrast threshold for the central stimulus is reduced. Two properties of this modulation in the human, imagined version of the task suggest that it is generated in early visual cortical regions. First, it is highly orientation specific, as are neurons in V1. Second, the effect is monocular-no reduction in contrast threshold is seen when the flanking stimuli to be memorized and later imagined are presented to one eye but the task is performed with the central stimulus presented to the other eye. Monocular cells are rare outside of V1. Finally, attention was previously thought to be a function of only higher order visual cortical neurons. It is now clear that attention modulates the responses of V1 neurons. This implies that primary visual cortex plays a greater role in vision than merely the detection and discrimination of visual stimuli. Taken together, these data support a role for primary visual cortex in certain types of perception, especially perception of features to which V1 neurons are highly tuned.