Biol. Cybernetics 34, 187-193 (1979) Biological Cybernetics 9 by Springer-Verlag 1979 Development of Orientation Selectivity in the Primary Visual Cortex of Normally and Dark Reared Kittens I. Kinetics Yves Fr6gnac Laboratoire de Neurophysiologie, Coll6ge de France, Paris, France Abstract. Kinetics of the development of orientation tuning are inferred from quantitative analysis of ex- tracellular recordings in the primary visual cortex of normally and dark reared kittens. 712 visual cells were classified in three functional groups: a) non-specific cells, and b) immature cells which are not as orien- tation selective as c) specific cells. Power regression and covariance analysis indicate that the "critical period" begins before 19 days and that the kinetics of the immature pool are the same in both rearing con- ditions. A catenary process of development of orien- tation selectivity is proposed, the immature compart- ment being a transit pool between non-specific and specific cells. Two sequential stages occur: 1) the realisation of an intrinsic programme of maturation, by which cortical specificity appears at eye opening and increases independently of visual experience 2) a phase of "epigenesis" beginning at 19 days, during which functional modification depends on visual experience. 1. Introduction After the pioneering work of Hubel and Wiesel (1963), it is now accepted that orientation selectivity is an inborn feature of some, but perhaps not all, visual cortical cells (Blakemore and Van Sluyters, 1975; Buisseret and Imbert, 1976; Sherk and Stryker, 1976). A recent study of developmental properties of cells recorded in the primary visual cortex of kittens gives evidence that before 3 weeks of age, most orientation selective cells respond preferentially to horizontal or vertical orientations aiad are mainly driven by the contralateral eye (Fr~gnac and Imbert, 1978). However, after 4 weeks of age, visual experience is necessary to maintain and develop orientation speci- ficity at the cortical level. At this age, specific neurones are found for all orientations in normally reared kittens and most cells are driven by both eyes. In contrast, in dark reared kittens of the same age, most visual cells, although binocularly activated, are no longer orientation selective. The electrophysiological findings of different wor- kers together suggest that the monocular "horizontal and vertical detectors" recorded during the first 3 weeks of postnatal life, which probably receive X-cell afferents, might form an initial pool of innately de- termined neurones which can be altered to a lesser extent by the absence of visual experience than the cortical neurones influenced by Y-cells (Leventhal and Hirsch, 1975, 1977 ; Fr6gnac and Imbert, 1978 ; Hirsch and Leventhal, 1978 ; Pettigrew, 1978). However, there is still a lack of quantitative analysis in the literature concerning the kinetics of orientation tuning in the presence or total absence of visual experience. Is there a preestablished specificity which disappears with di- suse if not verified by a normal visual input, or does visual experience guide the maturation of the visual cortex during a later period of postnatal development? A comparative study of the effects of dark and normal rearing has already been undertaken by Imbert and colleagues but the data were pooled in 3 age groups, the borders of which were arbitrarily chosen, and the authors were merely interested in the simila- rities between groups relative to orientation preference and ocular dominance. The 3-state analysis presented in Part I of this paper has been performed on the same recordings and describes the time constants of the specification and despecification processes, precisely enough to answer the 3 following questions: 1) What is the degree of functional specialisation of the visual cortex at eye opening, i.e. around 7 or 8 days? 2) At which stage of postnatal development does visual experience play a role in the maintenance or the 0340-1200/79/0034/0187/$01.40