Figure 12.8. Physiological measurements of spatial vision in amblyopes. A, Distributions of cortical eye dominance obtained from five populations of macaque monkeys. The eye dominance scale is that of Hubel and Wiesel (1962), but only for the normal animals is the eye assignment based on group 1 being contralateral and group 7 being ipsilateral. The distributions for the other four groups are keyed so that the amblyopic eye corresponds to group 1 in all cases. The data for normal subjects and for the three right-hand groups of amblyopes are from Kiorpes et al. (1998) and Movshon et al. (1987). The data for the monocularly deprived animals are from LeVay et al. (1980). B, Comparisons of physiological and behavioral data for 10 amblyopic monkeys studied by Kiorpes et al. (1998) and Movshon et al. (1987). Each axis represents the ratio of the indicated performance value between the amblyopic and fellow eye. For behavioral measures, the values compared are the peak spatial frequency and the peak contrast sensitivity from data like those shown in Fig. 12.7B. For physiological measures, the values compared are the geometric means of measured values for populations of cells tested monocularly through the amblyopic and fellow eyes.

significance. There was also no consistent effect of amblyopia on overall visual responsiveness, orientation tuning, or temporal tuning.

Thus, neuronal correlates of amblyopia are evident in the spatial properties of cells in V1, but the observed deficits do not fully explain the spatial losses in amblyopia—that is, the physiological losses are relatively small compared to the behavioral ones. A qualitative account of the visual loss in amblyopia might be constructed by combining the losses in spatial sensitivity with the alterations in eye dominance and binocularity (Fig. 12.8A), but we are aware of no quantitative model that supports this conjecture.

It is notable that, just as vision in an amblyopic eye resembles the vision of a younger normal eye, so, too, do the properties of cortical neurons driven by the amblyopic eye resemble the properties of neurons driven by a younger normal eye. However, this similarity is unlikely to reflect similar mechanisms in the two cases. We have already argued that the properties of developing cortical neurons are largely determined by the development of the foveal cone mosaic, but there is no reason to believe that retinal development is abnormal in amblyopic animals. LGN cell responses are quantitatively very similar in normal and amblyopic eyes (Blakemore and Vital-Durand, 1986b; Levitt et al., 2001). The disruption of cortical receptive fields in amblyopia must, therefore, result from changes in intracortical or intercortical circuits and not from degraded peripheral inputs. The changes observed are consistent with a broadening and blurring of the structure of cortical receptive fields, reminiscent of the far more extensive changes reported to result from complete binocular form deprivation (Blakemore, 1990). Like our data on V1 development, these results seem to favor a permissive view of the role of the environment in development, but with the added feature that not only visual experience, but the right kind of visual experience is required for normal development. In the animals raised with blurred vision in one eye, our experiments can be seen as a selective case of visual deprivation in which cells preferring the highest spatial frequencies are the most penalized by the experience of continuously blurred vision. Perhaps it is only natural that these cells would be the ones most affected, or even lost, resulting in distributions of preferred frequency that are shifted in the way that we have observed (Kiorpes et al., 1998; Movshon et al., 1987). But this account is incomplete—it does not suggest an explanation for the effects of strabismus, which does not cause image blur or the consequent loss of high spatial frequency stimulation.

It seems significant that the relationship between cortical signals and behavioral responses is consistent across development and amblyopia, even if the relationship is quantita tively imperfect. It is therefore natural to wonder about the course of development and the effects of visual experience in cortical areas outside V1.

0 0

Post a comment