Development of stimulus selectivity and functional organization in the suprasylvian visual cortex of the cat

Abstract

We have recorded from single neurons in the medial bank of the middle suprasylvian sulcus (PMLS) of anaesthetized and paralysed cats aged between nine days and eight weeks. Visual responses were assessed qualitatively, by using conventional projected stimuli, and quantitatively for drifting, high-contrast gratings of optimum spatial and temporal frequencies, but varying in orientation and direction of drift. At 9 days of age, some cells in the PMLS were spontaneously active but in three long penetrations only one visually responsive neuron was isolated. Between 9 and 15 days there was a rapid increase in the proportion of responsive units, which first appeared in small clusters in the lower layers (IV, V, VI). During the second and third postnatal weeks, spontaneous activity and the strength of visual responses increased to adult levels, and the proportion of cells showing rapid habituation to visual stimulation decreased. Even before two weeks of age, at least 85% of responsive cells in the PMLS were selective, by quantitative criteria, for image motion along one particular axis, and a majority of these were clearly direction-selective (responding to movement in one direction significantly more strongly than to that in the opposite). By the end of the third postnatal week the proportion of units with strong direction preference reached adult levels. The selective cells were initially more broadly 'tuned', on average, for the direction of motion of a grating (mean half-width in animals of 10-12 days was 32.6 degrees), but the sharpness of tuning improved to reach the adult level (ca. 23 degrees) during the third postnatal week. In animals younger than three weeks a slightly smaller proportion of cells than in adults (but always more than one third of all visually responsive cells) responded to stationary, contrast-modulated gratings. The majority of these cells showed clear selectivity for the orientation of a flashed grating. A few 'non-selective' cells were found in the youngest animals but by the end of the third postnatal week virtually all cells responsive to stationary gratings displayed orientation selectivity. There was always good agreement between the preferred orientations for stationary and drifting gratings. Even before two weeks of age, when responsive cells occurred only in small clusters, there was a clear tendency for neighbouring neurons to have similar or opposite preferred directions, just as in adult cats. By 2-3 weeks of age there were clear progressive shifts in stimulus preference along oblique or tangential tracks.(ABSTRACT TRUNCATED AT 400 WORDS)