A look was coded if infants looked at the ottoman following the mention of a hidden object. A point was coded if infants looked and raised their arm in the direction of the ottoman. Both index finger and full-hand pointing were considered. Approaching the ottoman was coded if the baby looked at the ottoman and moved their body toward the ottoman. Videotapes of the sessions (representing 71% of the sessions) were then coded by a second coder who was blind to the hypothesis of the study and to the condition.

The coder was not blind to the position of the ottoman because it was partially visible on the tapes. Overall Gefitinib cell line agreement on the presence or absence of target behaviors was high (94%, Cohen’s kappa 0.88). Disagreements were resolved via discussion, and the experimenter’s YAP-TEAD Inhibitor 1 research buy initial judgments were used in the analyses below. The purpose of this experiment was to investigate why infants have difficulty orienting to a hidden toy’s location after having seen this toy in an adjacent room. We predicted that infants would perform at similarly high levels with the new and a familiar toy in the identifying feature condition. In the nonidentifying

feature and the no feature conditions, we predicted high performance with the new toy and poor performance with the familiar toy. Results are displayed in Figure 1. As a first step, to ensure that infants were equally attentive in the three familiar toy conditions, we analyzed the time they looked next at the object when the experimenter highlighted the object or its feature during the familiarization phase. Data from one participant in the identifying feature condition were excluded from this and all other analyses because the infant focused on the object more than 2.5 standard deviations longer than average. A one-way Welch ANOVA1 revealed no difference in how long infants looked at the object across the three conditions during the feature

introduction, F (2, 28.65) = 1.97, p = 0.16, (identifying feature: M = 9.53 sec, SE = 1.06, nonidentifying feature: M = 9.25 sec, SE = 0.71, no feature: M = 7.58 sec, SE = 0.64). Importantly, how long infants looked at the object during the familiarization did not predict whether infants responded or not to the familiar toy in the test phase (logistic regression, β = 0.003, p = 0.43). This suggests that any differences in infants’ responses to a familiar object across conditions cannot be explained by differences in their attention during the familiarization phase. Further analyses of infants’ responses in the test phase revealed no effects of gender, side, or toy order. Boys were as responsive as girls, and neither the side where a toy was hidden, nor the order of the familiar and the new toy conditions mattered for infants’ ability to respond. There was also no interaction between condition and order.