Eight experiments tested the abilities of 3-4-year-old children to reorient
themselves and locate a hidden object in an open circular space furnished
with three or four landmark objects. Reorientation was tested by hiding a t
arget object inside one of the landmarks, disorienting the child, observing
the child's search for the target, and comparing the child's performance t
o otherwise similar trials in which the child remained oriented. On oriente
d trials, children located the target successfully in every experiment. On
disoriented trials, in contrast, children failed to locate the object when
the landmarks were indistinguishable from one another but formed a distinct
ive geometric configuration (a triangle with sides of unequal length or a r
ectangle). This finding provides evidence that the children failed to use t
he geometric configuration of objects to reorient themselves. As in past re
search, children also did not appear to reorient themselves in accord with
non-geometric properties of the layout. In contrast to these findings, chil
dren successfully located the object in relation to a geometric configurati
on of walls. Moreover, adults, who were tested in two further experiments,
located the object by using both geometric and non-geometric information. T
ogether, these ten experiments provide evidence that early-developing navig
ational abilities depend on a mechanism that is sensitive to the shape of t
he permanent, extended surface layout, but that is not sensitive to geometr
ic or non-geometric properties of objects in the layout. (C) 2001 Published
by Elsevier Science B.V. All rights reserved.