Processing of Contour Information by Human Infants Evidenced by Pattern-Dependent Evoked Potentials

Visually evoked potentials (VEPs) elicited by checkerboard stimuli varying in check size (from 20′ to 5°) or arrangement (redundant or random) were recorded from the occipital pole of 55-107-day-old infants, using 2 techniques of pattern presentation. Patterns were illuminated with either brief light flashes (tachistoscopic presentation, N = 33) or a light with intensity varying sinusoidally (modulated presentation, N = 7). Both techniques yielded similar results in that an inverted U-shaped function related contour density (as measured by the square root of the total of all lengths of black-white transitions in the pattern) to variations in the amplitude of the positive component peak (P2) which occurred 100 msec after light onset; p < .001, r = .88. The maximum P2 amplitude was found to shift toward stimuli with greater contour densities as chronological (CA) and neurological (NA) age increased where NA was based on the latency from stimulus onset of P2. This shift did not appear to be related to increases in gestational age (GA) where GA was based on reported due date. The inverted U-shaped function of VEP component amplitudes and shifts in function maximum toward increases in contour density as CA increases was identical with that reported by Karmel (1969a) for total time spent fixating comparable stimuli.