Search

1. Visual abilities decline during normal aging, and many of these declines are due to neural changes in the retina or central visual pathways. We have begun studies of the primate visual system to investigate the location and nature of these changes as well as to answer general questions about the effects of aging on neural function. We began with the dorsal lateral geniculate nucleus (LGN) because it is the main structure through which visual information passes on the way to cortex and because the parallel parvocellular and magnocellular pathways, which may be affected differently by aging, are anatomically distinct there. 2. Single-cell recordings were made in the LGN of young adult (5-16 yr) and old (25-28 yr) rhesus monkeys. We made quantitative measures of a wide variety of response properties for a large number of parvocellular (n = 257) and magnocellular (n = 113) neurons in the two groups of animals. As a result, in addition to studying the effects of aging, we were able to make quantitative comparisons between parvocellular and magnocellular neurons using larger samples than have been studied previously and for some properties that have not been studied before. 3. We found that magnocellular neurons have significantly higher maximal response rates and signal-to-noise ratios than parvocellular neurons. However, response latencies to visual stimulation were similar for neurons in the two types of layers. In agreement with previous studies, magnocellular neurons had higher maximal contrast sensitivity and higher contrast gain than parvocellular neurons. However, the sensitivity difference occurred because nearly all of the neurons with low sensitivities (< 10) were in the parvocellular layers, not because neurons in the magnocellular layers had the highest sensitivities. 4. Neurons with the smallest receptive-field centers, the highest spatial-frequency resolutions, and the highest optimal spatial frequencies were found in the parvocellular layers. However, the overall distributions of each of these properties overlapped substantially for neurons in the two types of layers, and the mean values were not significantly different. The mean high temporal-frequency cutoff was significantly higher for magnocellular than parvocellular neurons, but the difference was small (only 3 Hz), and it occurred because many parvocellular neurons had lower cutoffs than any seen in the magnocellular layers, not because magnocellular neurons had the highest temporal-frequency cutoffs. Parvocellular neurons also had narrower temporal-frequency tuning than magnocellular neurons. However, there was no significant difference in optimal temporal frequency.(ABSTRACT TRUNCATED AT 400 WORDS)

The present study tested the hypothesis that nondominant-eye influences on lateral geniculate nucleus (LGN) neurons affect the processing of spatial and contrast information from the dominant eye. To do this, we determined the effects of stimulating the nondominant eye at its optimal spatial frequency on the responses of LGN cells to sine-wave gratings of different spatial frequency and contrast presented to the dominant eye. Detailed testing was carried out on 49 cells that had statistically significant responses to stimulation of the nondominant eye alone. Spatial-frequency response functions to nondominant-eye stimulation indicated that the responses were spatially tuned, as reported previously (Guido et al., 1989). Optimal spatial frequencies through the nondominant eye were significantly correlated with the optimal spatial frequencies through the dominant eye (r = 0.54; P less than 0.0001), and the optimal spatial frequencies were fairly similar for the two eyes. Nondominant-eye stimulation changed the maximal amplitude of the fundamental (F1) response to dominant-eye stimulation for only about 45% (22 of 49) of the cells that responded to nondominant-eye stimulation alone. The response vs. contrast function through the dominant eye was altered for 73% of the cells (51% independent of spatial frequency). Three types of effects were observed: a change in the initial slope of the response vs. contrast function (contrast gain), a change in the response amplitude at which saturation occurred, or an overall change in response at all contrasts. The incidence of these changes was similar for X and Y cells in LGN layers A, A1, and C (only four W cells were tested). Nondominant-eye stimulation had little or no effect on the sizes or sensitivities of the receptive-field centers or surrounds for the dominant eye. In addition, nondominant-eye stimulation had little or no effect on optimal spatial frequency, spatial resolution, or the bandwidth of spatial-frequency contrast sensitivity curves for the dominant eye. Possible functions of binocular interactions in the LGN are considered. The present results suggest a role in interocular contrast-gain control. Interocular contrast differences can occur before the acquisition of binocular fusion, when the two eyes are viewing different aspects of a visual stimulus. Psychophysical and physiological studies suggest that an interocular mechanism exists to maintain relatively constant binocular interactions despite differences in interocular contrast. The present results suggest that at least part of this mechanism occurs in the LGN.

1. X, Y, and W cells in the A and C layers of the cat's dorsal lateral geniculate nucleus (LGN) were tested for responses to stimulation of the nondominant eye. The main purpose was to determine the incidence of nondominant-eye excitation and inhibition among different classes of cells and to examine the spatial-frequency tuning of responses to the nondominant eye. 2. Of 198 cells that were tested with drifting sine-wave gratings presented to the nondominant eye, 109 (55%) showed statistically significant responses. Four types of responses were observed: an increase in the mean discharge rate (F0 excitation), a decrease in the mean discharge rate (F0 inhibition), an increased modulation at the fundamental frequency of the grating (F1 excitation), and a decreased modulation at the fundamental frequency of the grating (F1 inhibition). Overall, 29% of the cells responded with inhibition, 24% responded with excitation, and 2% showed both excitation and inhibition, depending upon the spatial frequency and/or the harmonic response component. The relative incidence of excitation and inhibition was similar for X, Y, and W cells, for cells with on-center and off-center receptive fields, for cells with different receptive-field eccentricities, and for cells in each LGN layer. In addition, within layers A and A1, responses were similar for cells at different distances from the laminar borders. 3. Spatial-frequency response functions indicated that cells could have band-pass or low-pass spatial-frequency tuning through the nondominant eye. Band-pass cells tended to be narrowly tuned (less than or equal to 1 octave), and low-pass cells responded to a broader range of spatial frequencies. These properties were similar for X, Y, and W cells. Spatial resolution tended to be low (less than or equal to 0.8 c/deg for most cells), although a few cells responded to the highest spatial frequency tested (5.4 c/deg). Likewise, optimal spatial frequency was low (less than or equal to 0.2 c/deg) for most cells. These properties were similar for X and Y cells, and there was a weak tendency for X and Y cells to have higher optimal spatial frequencies and spatial resolutions than W cells. 4. In terms of absolute change in activity, responses to drifting gratings were weak. However, cells that were inhibited generally showed 20-60% decreases in activity to the optimal spatial frequency, and cells that were excited generally showed 40-100% increases. Response amplitudes were similar for X, Y, and W cells.(ABSTRACT TRUNCATED AT 400 WORDS)

In unequal alternating monocular exposure, each eye receives normal patterned input, but on alternate days and for unequal periods. This imbalance in stimulation produces a behavioral deficit for the less-experienced eye and alters the ability of that eye to activate cortical cells. To determine whether unequal alternating exposure also affects the sizes of cells in the lateral geniculate nucleus (LGN), we measured the cross-sectional areas of geniculate neurons in seven normally reared cats, 14 cats reared with equal alternating exposure, and 17 cats reared with unequal alternating exposure. We found that, in the LGNs of cats reared with unequal alternating monocular exposure, cells in layers that received their input from the less-experienced eye were smaller than those in layers that received their input from the more-experienced eye. This effect was restricted to the binocular segments of the nucleus, and the difference in cell size was a function of the imbalanced exposure, rather than the length of exposure per se. In control groups given balanced alternating exposure, cell size was not correlated with the length of daily exposure. In cats reared with unequal exposure, the change in cell size was greater in the nucleus ipsilateral to the less-experienced eye. Further, the size of the effect was correlated with the size of the imbalance imposed during rearing: Cats reared with a moderate imbalance (8 hours/day vs. 4 hours/day) showed less change in cell size than cats reared with a large imbalance (8 hours/day vs. 1 hour/day). These results are consistent with those of behavorial and physiological studies and strongly suggest (1) that unequal alternating monocular exposure affects the sizes of cells in the LGN by altering the normal competitive balance between the retinogeniculocortical pathways from the left and right eyes, and (2) that the contralateral pathway has some inherent advantage in this competition. We also found a slight shrinkage of cells in the LGNs of cats reared with equal alternating monocular exposure. Since this effect was restricted to the binocular segments of the nucleus, and was not related to the length of exposure given, it was probably caused by the imbalanced binocular competition that occurred during each day's monocular exposure.

Damage to visual cortex (areas 17-19) in kittens or adult cats produces severe retrograde degeneration of neurons in the dorsal lateral geniculate nucleus (LGN). However, some neurons survive in otherwise degenerated portions of the LGN after a visual cortex lesion at any age. Previous studies have shown that there are well-defined differences in potential retinal inputs, soma size, synaptic connections, outputs, and physiological properties of output targets of the surviving LGN cells in cats that received visual cortex damage at different ages. The present experiment investigated the relationships between these differences and the responses of surviving LGN neurons to visual stimulation. Recordings were made from surviving neurons in the degenerated A- and C-layers of the LGN in cats that had received a visual cortex lesion on the day of birth, at 8 weeks of age, or as adults (survival was 11.5-36 months). Normal adult cats were studied for comparison. The visual receptive field was mapped, and tests were carried out to classify each cell as X, Y, or W. In addition, quantitative methods were used to assess response amplitude, strength of receptive-field surround inhibition, spatial-frequency tuning to drifting or counterphased sine-wave gratings, and response to nondominant-eye stimulation for each cell. We found that surviving cells in all LGN layers respond to light, have normal receptive-field organization, and have normal eye dominance following a lesion at any age tested. In addition, gross retinotopic organization of the LGN is normal. However, 2 main abnormalities were observed following a lesion at all 3 ages. First, there is a reduction in the percentage of X cells in the A layers, from 62% in normal LGNs to about 15% in degenerated LGNs. Second, many surviving cells in both the A- and C-layers have abnormally large receptive-field centers. Other differences that were observed between normal A-layer cells and surviving A-layer cells could be attributed to the loss of X cells. These results indicate that cells within a structure that shows severe retrograde degeneration after brain damage can maintain relatively normal function and can take part in potentially important residual neural pathways. Previous studies indicate that these residual pathways can show both anatomical and physiological compensation for the brain damage, and the present findings bear on the consequences and mechanisms of this compensation.

1. Previous studies have shown that rearing with monocular visual deprivation (MD) produces a loss of Y-cells and a reduction in spatial resolution among X-cells in layers A and A1 of the cat's dorsal lateral geniculate nucleus (dLGN). However, there have been no studies of the effects of visual deprivation on the function of the retinogeniculate W-cell pathway, which terminates in the C layers of the dLGN. It also is not known if Y-cells in the C layers are affected by MD in the same way as Y-cells in the A layers. These questions were addressed by the present experiment. 2. Single-cell recordings were made from the C layers of 5 normal adult cats (112 cells) and from the nondeprived (94 cells) and deprived (95 cells) C layers in 10 cats monocularly deprived by lid suture for 3-7 yr. The cells were classified as X, Y, or W on the basis of their receptive-field properties and responses to electrical stimulation of the optic chiasm. In addition, quantitative measures were made of responses to sine-wave gratings of different spatial frequencies. 3. Receptive-field organization, receptive-field center size, spatial and temporal linearity to counterphased sine-wave gratings, and latency to optic chiasm stimulation were similar for C-layer cells in normal cats and in the deprived and nondeprived layers of MD cats. On the basis of these properties, 23% of normal layer-C cells were classified as Y-cells and 72% were classified as W-cells. The Y-cells tended to be located in the magnocellular division of layer C and most (though not all) W-cells were in the parvocellular division. Normal layers C1 and C2 contained almost exclusively W cells. The incidence of Y and W cells was similar to normal in the nondeprived and deprived C-layers of MD cats. 4. In normal cats, W cells typically had the lowest amplitude first-harmonic (F1) response rates to drifting sine-wave gratings. However, many W cells gave quite brisk responses and, overall, there was no significant difference between F1 response amplitudes of Y and W cells. Response amplitudes of Y- and W-cells in the deprived and nondeprived C-layers of MD cats were not significantly different from normal. 5. Normal Y- and W-cells tended to have low optimal spatial frequencies (0.2 c/deg or lower) and spatial resolutions (generally 0.4-1.6 c/deg) to drifting sine-wave gratings.(ABSTRACT TRUNCATED AT 400 WORDS)