71, p = .489. Results of a within-subjects ANOVA yielded a main effect of display, F(2, 26) = 15.71, p < .0001, due to differences in mean number of manual actions produced in sequence to each of the displays. Pairwise comparisons (with LSD) suggested that the infants engaged in a reliably greater number of sequential manual gestures during the trial toward the impossible cube relative to the possible cube display, t(13) = 4.29, p < .001, and the perceptual controls, t(13) = 4.05, p < .001, as shown in Figure 2b. The mean impossible preference score was .68, which differed

significantly from chance, t(13) = 3.58, p < .003. Infants attempted an average of three additional sequential actions toward the impossible cube display above that of the possible cube display. The pattern of greater manual exploration toward the impossible cube was observed in 12 of the 14 infants, with two engaging in more reaching to the possible cube, Z = 3.01, p = .003. p38 MAPK inhibitor Results of a within-subjects ANOVA yielded a main effect of display, F(2, 26) = 13.40,

p < .0001, due to differences in mean number of instances of social referencing occurring during each of the displays. Pairwise comparisons (with learn more LSD) indicated that infants engaged in a reliably greater amount of social referencing overall to the caregiver and/or experimenter when presented with the impossible cube relative to the possible cube, t(13) = 2.87, p < .01, and the perceptual controls, t(13) = 5.27, p < .001, as shown in Figure 2c. The mean impossible preference score was .64, which differed significantly from chance, t(13) = 2.58, p = .02. On average, infants engaged in two additional instances of social referencing to the parent and/or experimenter during presentation of the impossible cube display above that of the possible cube Tangeritin display. This pattern of behavior was observed in 11 of the 14 infants, with two infants referencing equally and one infant referencing to a greater extent during the possible cube display, Z = 2.45, p = .015. Further analyses revealed that infants engaged in significantly more referencing behaviors toward the experimenter (relative to

the mother) during the presentation of the impossible cube display, t(13) = 3.47, p < .005. However, there were no significant differences in the amount of referencing behaviors to the mother relative to the experimenter during the possible cube display (p > .10), and infants’ first looks to either of the adults during both the possible and impossible cube displays did not differ from chance (p > .25). There was a main effect of display, F(2, 26) = 8.57, p < .001, due to differences in mean number of vocalizations emitted during each of the displays. Pairwise comparisons (with LSD) demonstrated that infants produced a greater number of vocalizations during the impossible cube display relative to the possible cube, t(13) = 3.15, p < .01, and the perceptual controls, t(13) = 3.57, p < .001, as shown in Figure 2d.