Postnatal development of receptive field surround inhibition in kitten dorsal lateral geniculate nucleus

Abstract

We recorded the responses to visual stimulation of single neurons in the A-layers of the dorsal lateral geniculate nucleus (LGNd) of 4- to 5-wk-old kittens and adult cats. Visual stimuli were generated on a cathode-ray tube (CRT) display and consisted of circular spots and annuli whose contrast was twice the threshold for each neuron and was modulated about a background luminance of 28 cd/m2 at 0.5 Hz. Neural responses were collected as interspike intervals and displayed as instantaneous firing rates for individual trials. From the responses to a series of sizes of spot stimuli, area-response functions were constructed and used to derive a quantitative measure of the strength of the receptive field (RF) surround inhibition of each neuron, the spatial density minimum ([SDmin[). To separate neural from optical factors that affect measurements of surround inhibition, published values for the posterior nodal distances of the kitten and adult eye were used to scale stimuli in terms of the retinal area subtended. Of 153 kitten and 95 adult LGNd neurons studied, the responses to a complete series of spot stimuli of different sizes (areas) were obtained for 52 kitten neurons [44 with linear spatial summation (L) and 8 with nonlinear spatial summation (NL)] and 45 adult (24 X-and 21 Y-) neurons. In addition, intracellular recordings were made from 30 of the kitten neurons that were filled iontophoretically with horseradish peroxidase (HRP) and were evaluated structurally. In the adult, neurons were classified as X-or Y-cells on the basis of a battery of physiological properties, including linearity of spatial summation, latency to electrical stimulation of the optic chiasm, and ability to respond reliably to rapidly moving stimuli. Kitten neuronal responses allowed them to be clearly identified as exhibiting linear or nonlinear spatial summation, but application of additional criteria produced ambiguous results for classification into X-or Y-categories. Kitten L or NL neurons showed differences typical of adult X-and Y-cells on some [e.g., RF center size (P less than 0.01)] but not other [e.g., latency to stimulation of optic chiasm (P greater than 0.40)] properties. In addition, by direct comparison of morphological features with these physiological responses, some kitten cells with adult X-cell physiological properties on these tests were found to have typical adult Y-cell somadendritic structure.(ABSTRACT TRUNCATED AT 400 WORDS)