820

INFANTS’ REASONING ABOUT OBJECT IDENTITY IN MOVING EVENTS WITH STATIC ENDPOINTS: THE NATURE OF THE MAPPING PROBLEM

Teresa Wilcox Department of Psychology, Box 19528, University of Texas, Arlington, TX 76019

Renee Baillargeon University of Illinois, Champaign-Urbana

Recent research suggests that infants can use featural information to reason about object identity. However, whether or not this ability is revealed depends on the specific paradigm used. Experiments conducted with the size comoarison paradigm have yielded positive results with 7.5- month-old infants (Wilcox & Baillargeon, 19951, whereas experiments carried out with the obiect mappinq paradigm have produced positive results only in infants aged 11.5 months and older (Xu & Carey, in press).

In one experiment that used the size comparison paradigm (Wilcox & Baillargeon, 19951, 7.5-month-old infants watched a ball disappear behind one edge of a screen; after a brief interval, a box appeared at the other edge. For half the infants the screen was too narrow to hide the ball and box simultaneously (narrow-screen condition); for the other infants the screen was sufficiently wide to hide both objects (wide-screen condition). The event repeated until the end of the trial. The infants looked reliably longer at the narrow-screen than wide-screen event. These and control data indicated that the infants (a) concluded that two distinct objects were involved in the event; Lb) judged that both objects could fit behind the wide but not the narrow screen; and hence (4 were surprised by the narrow-screen event.

These findings contrast with those of a similar experiment conducted with the object mapping paradigm (Wilcox 81 Baillargeon, 1995). Infants 9.5 and 11.5 months of age saw a wide- screen event similar to that described above except that half the infants saw a ball rather than a box appear at the other edge of the screen. After several repetitions of the ball-box or ball-ball event, the screen was lowered to reveal a static array containing one ball. The 11.5-month-old infants looked reliably longer at the final array after seeing the ball-box as opposed to ball-ball event. The younger infants looked equally at the final array regardless of the event seen initially, as though they were unable to judge whether the one ball mapped onto the object(s) seen in the preceding ball-ball and ball-box event. However, when the ball and box were presented only once on each side of the screen, 7.5- to 8.5-month-old infants looked reliably longer at the final array after seeing the ball-box event. Together, these findings suggest that the multiple emergences of the ball and box on either side of the screen made it difficult for the younger infants to map the final array onto the object(s) seen in the preceding event. It is possible, however, that younger infants have a general difficulty reasoning about object identity in moving events with static end states.

To investigate this possibility, 7.5-month-old infants were tested with a size comparison task that ended in a static state. Infants saw a narrow-screen event similar to that described above except that the large ball was replaced by a small ball. The width of the narrow screen was more than twice the width of the small ball, yet the screen was too narrow to occlude both the small ball and the large box simultaneously. Thus, the infants had to compare two different sized objects with the size of the screen. After the box’s first emergence it stopped and remained stationary until the end of the trial. The infants were surprised by the narrow-screen event. These and control data suggested that the infants (a) concluded that two objects were involved in the narrow- screen event and Lb) understood that the screen was too narrow to occlude both the small ball and large box simultaneously. These results confirm our earlier findings that 7.5-month-old infants can use featural information to reason about the number of objects involved in an occlusion event and provide further evidence that infants compare the size of the object seen on each side of the screen to the size of the screen. In addition, these results suggest that younger infants’ failures on the object mapping task are not due to a general difficulty with reasoning about moving events with static end states but to a more specific mapping problem.