THE DISCRIMINATION OF ORIENTATION BY YOUNG INFANTS DAPHNE MACRER* and MYRNAMARTELLO Department of Psychology, McMaster University, Hamilton. Ontario. Canada L8S 4K I (Received 5 zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCB June 1979) Abstract-After being habituated to obliquely oriented stripes, %-week-old infants looked longer at stripes oriented along the opposite diagonal than at the stripes to which they had been habituated. However. they looked equally long at the ‘*habituated” stripes as at their negative (in which black stripes replaced the white stripes and vice versa). These results suggest the infants were processing orientation. and not just some regional difference between the stimuli. The results also suggest the visual cortex is functioning to some extent by 5-6 weeks. INTRODUfflON Young infants, including newborns, can discrimi- nate vertical from horizontal stripes (McKenzie and Day, 1971; Slater and Sykes, 1977). This ability seems to imply the visual cortex is functioning, since the visual cortex is the first structure in the visual path- way sensitive to orientation (cat: Hubel and Wiesel, 1959, 1962; monkey: Hubel and Wiesel, 1977; human: Campbell and Kulikowski, 1966; Campbell and Maf- fei. 1970; Maffei and Campbell, 1970), and since mon- keys with lesions of the visual cortex appear not to discriminate orientation (Weiskrantz, 1963). However, studies of young infants’ scanning suggest they might discriminate horizontal from vertical stripes on some basis other than orientation. Although we have no data on how young infants scan stripes, two aspects of how they scan other figures suggest this possibility: their eye movements are mainly horizontal (Salapatek, 1968; Salapatek and Kessen, 1966; but see also Hainline, 1978);t and they scan only a limited portion of a figure (Hainline, 1978; Leahy, 1976; Maurer and Salapatek, 1976; Salapatek, 1968, 1975; Salapatek and Kessen, 1966, 1973). Consider first how horizontal scanning would affect an infant’s discrimination of two striped squares like those used in previous studies. When he scanned a square filled with vertical stripes, his eyes would cross many transitions of black to white; when he scanned a square filled with horizontal stripes, his eyes would not cross any. He could detect apparent differences between the squares in patterning, contour or brightness, differences which even newborns can discriminate easily (Fantz er al., 1975; Hershenson, 1964). Such discrimination would not necessarily im- plicate the visual cortex, since a monkey with an * To whom reprint requests should be sent. t Note that if, as Hainline (1978) suggests, infants’ scan- ning is predominantly vertical, then the same interpretive difficulties arise. extensive lesion of the visual cortex can discriminate stimuli which differ in any of these respects (Hum- phrey, 1974; Pasik et al.. 1969; Weiskrantz 1963). In the following experiment we tested whether infants could discriminate between two circles filled with oblique stripes oriented along opposite diagonals (Stimuli A and C in Fig. 1). If an infant scanned these patterns horizontally, his eyes would cross the same number of black-white transitions on either pattern. Thus horizontal scanning would not allow an infant to discriminate between the two orientations on the basis of patterning, contour, or brightness. Our other concern was that infants might be able to discriminate between different orientations because they scan (or process) only a limited region of the stimuli, a region in which the stimuli differ in pattem- ing. contour, ot brightness. Suppose, for example, that an infant scanned only a region on the bottom left edge of the circles we used in this study. Then in one circle he would see a solid white bar; in the other, he would see a pattern of black and white bars. He could discriminate between the two oblique orientations on the basis of regional differences in contour. pattem- ing, or brightness. Infants could have discriminated between the horizontally and vertically striped squares used in previous studies in a similar way. To find out if it were likely that infants discrimin- ated the two oblique orientations on such a regional basis, we also tested the discrimination of two stimuli which were identical in overall contour and intensity but which differed regionally: between a pattern of stripes and its negative (where one stimulus had a white stripe, the other had a black) (see stimuli A and B in Fig. 1). If an infant were processing just one small region of these two stimuli, he would be exposed to a large difference in intensity; on one pat- tern he would see a region of white; on the other, a region of black. Since infants can easily discriminate black from white (Hershenson, 1964). we would expect them to discriminate the positive from the negative-but only if they were processing a suffi- zyxwvutsrq 201