338 ORGANIZATION OF VISUAL PERCEPTION AT BIRTH Alan M. Slater The research presented in this paper describes a series of studies in which an infant-controlled habituation procedure has been successfully used to describe some aspects of visual organisation at birth. The f i r s t experiments demonstrate that newborns w i l l reliably habituate to two-dimensional stimuli, and w i l l give strong novelty preferences when the habituated stimulus is paired with a novel stimulus that has changed either in both shape and hue, or in shape alone. ,~ / f A number of conclusionsJresult~rom the other experimental series that are described. Two studies, using monocular viewing conditions, demonstrate that habituation cannot be attributable merely to retinal adaptation. Additional studies give strong support to the model of habituation which assumes i t to be an exponentially decreasing process. Studies using movement show that newborns can extract shape, movement and orientation from moving stimuli, and can transfer such learned discriminations to stationary stimuli, demonstrating that some form of identity constancy is present at birth. The last experiments described show that newborns can readily discriminate between 2- and 3-dimensional representations of objects, but are unable to transfer learning across dimensions. The experiments described necessitate the view that the structural basis of visual perception is innate, and i t w i l l be argued that perceptual development is largely a matter of the modification and improvement of perceptual input that is organized at birth.