Abstract: Humans and animals are able to orient themselves to an environment. Previous studies have demonstrated that young children’s reorientation behaviors are mainly guided by the geometric information of their environment. Children often reorient themselves according to the shape of the testing room and not in accordance with the room’s non-geometric properties. While previous studies typically used rectangular testing rooms that provided unique geometric information, we used a trapezoidal testing room that allowed us to explore the manner in which children use geometric information, including the angle of the corners as well as the dimensions of the walls.
In a preliminary experiment, a rectangular testing room (1.9 × 1.2 × 1.2 m) was used. In Experiment 1 (where the participants were children) and Experiment 2 ( where the participants were adults), one of the walls of rectangular room was shortened from 1.2 m to 1 m to form a trapezoidal testing room. In Experiment 3, the lengths of the two hemlines of the trapezium were altered to 0.9 m and 1.5 m. Participants included children who were 2–4 years old and adults (N = 20) who were university students. In each experiment, the participants’ reorientation abilities were tested through a procedure adapted from Hermer and Spelke (1994, 1996). First, the participants were allowed to observe the environment. Thereafter, they watched the experimenter hide an object in one of the landmark boxes positioned in the four corners of the room. Following one of the two experimental manipulations, the participants were required to retrieve the object. In oriented search trials, the participants retrieved the object after turning 360 degrees in one spot four times, with their eyes open. This served as a control condition that probed whether the participants remembered where the experimenter had hidden the object. In disoriented search trials, the participants retrieved the object after turning 360 degrees, four to six times in one spot, with their eyes closed; this condition was found to induce a state of disorientation in children of this age (Hermer and Spelke, 1996).
The results of the preliminary experiment indicated that children search geometrically appropriate locations more often than geometrically inappropriate locations, but did not search proximate locations more often than they do distant locations. In particular, the proportion of the search at the correct corner and at the geometrically equivalent opposite corner (in diagonal direction) did not differ. This result was consistent with the findings of previous studies.
When children were tested in a trapezoidal room, they still went to the correct corner or the opposite corner. The video recording revealed that when children faced the two hemlines of the trapezium, they frequently and directly went to the correct corner or the opposite corner. However, when they faced the two lateral walls of the trapezium, they infrequently and indirectly went to the correct corner or the opposite corner.
The present study provides converging evidence suggesting that children do not rely solely on geometric information about the dimensions of the walls of their environment. Further, their reorientation behaviors can be affected by egocentric information. It is most probable that children use information of the principal axe of the trapezoidal testing room in the task of reorientation

Key words: spatial representation, reorientation, trapezium, principal axe