Purchase this article with an account.
Chuan Hou, Melanie Palomares, Anthony Norcia; Texture-defined Figure/Ground Segmentation In Human Visual Development: A High-Density Electrical Mapping Study. Journal of Vision 2010;10(7):491. doi: https://doi.org/10.1167/10.7.491.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
The human visual system uses texture discontinuities and grouping mechanisms to segment visual scenes in objects and supporting background. To study the development of these processes, we used synthetic images portraying simple texture-defined figures and high-density VEP recording in 3-5 month old infants and in adults. We compared responses in two conditions: in one condition a set of disk-shaped regions appeared and disappeared from a uniform background. The uniform background consisted of horizontal 1-dimensional dynamic noise. The figures were created by rotating the texture within 9 small disk-shaped regions from horizontal to vertical at 1 Hz. In a control condition, the texture within the disks also rotated between horizontal and vertical at 1 Hz, but the disks remained segmented when the texture was horizontal. A difference in response profile between these two conditions was used to isolate responses that are sensitive to the global organization of the stimulus. We recorded strong, but delayed segmentation-related responses in the infants: their initial segmentation response started at about 200 msec after the onset of the figures versus 70-90 msec in adults. The infant response waveform was much simpler than that of the adults and their responses were restricted to the occipital midline at all latencies. This pattern contrasted strongly with that of the adults: after an initial peak at the occipital pole, the adult response spread to lateral electrodes over right and left lateral cortex by 150 msec. Activity over frontal electrodes occurred by 200 msec, followed by later activity on lateral sites at 250-300 msec. While infants are sensitive to the global organization of our stimulus, they process this information in a much more restricted range of cortical areas. In particular, we found no evidence of activity in lateral occipital and frontal areas that is prominent in the adult evoked response.
This PDF is available to Subscribers Only