In the earliest experiments 
using techniques of this kind, 
psychologists found that 
infants preferred to look at 
some colors more than others, 
from which it was inferred, 
reasonably enough, that 
infants can perceive colors. But 
Robert Fantz realized that the 
same technique could be used 
to study form perception in 
infants. If infants could indeed 
discriminate among forms, and 
if they preferred to look at one 
pattern more than at another, 
then they would spend more 
time gazing at that pattern. The 
results of this kind of research 
on human infants can be 
summarized as follows: In the 
first month of life, infants 
show no preference for one 
simple geometrical shape, such 
as a circle or a cross, over 
another. But they do spend 
more time looking at a complex 
pattern (such as a 
checkerboard) than at a simple 
one (such as an outline square). 
They prefer to look at drawings 
of the human face rather than 
at drawings of the same overall 
shape and size, but they show 
the same preference for face 
configurations in which the 
features are so thoroughly 
scrambled that one must 
wonder if the figures have any 
quality at all that makes them 
facelike. After the first month, 
and increasingly so with 
advancing age, infants show a 
preference for a normal as 
opposed to a scrambled face and 
for some simple shapes over 
others.

    To ascertain with greater 
precision exactly what region 
within a figure an infant is 
looking at, more sophisticated 
devices for recording eye 
movement and position have 
been used. With one such 
apparatus, Philip Salapatek and 
his associates at Yale 
University were able to show 
that newborn infants, in 
looking at a triangle, tended to 
direct their gaze at its vertices. 
Further research suggests that 
infants tend to seek out areas of 
the greatest change or 
discontinuity.