Publication Type: Individual Poster
Abstract: Background and Aims. Recent work on infants’ imitation emphasizes their understanding of means and ends as separate and combinable components. For instance, Elsner and Aschersleben (2003) found that 15- and 18-month-old infants were sensitive to which of two modeled actions produced which of two effects, and we have found that 15-month-old infants readily apply an action learned with one effect to a novel toy yielding another. In the research reported here, we examined whether infants can learn multiple ways to generate the same effect. This form of flexibility is the complement of that shown in our earlier work, where infants used the same means to produce different effects.

Methods and Key results. The participants were 36 infants between 14 and 16 months old (M = 15.0 mo, SD = 17 days). The experimenter initially demonstrated opening a wooden box either with her hand (n = 18) or with a novel tool (n = 18), and then gave the materials to the infant for a turn. All but 3 of the 36 infants (X2 = 22.4, p < .0001) attempted to open the box by the method the experimenter had demonstrated. This was especially impressive for the novel tool, as most infants were not successful with it but nevertheless did not set it aside and use their hands instead.

The experimenter then demonstrated opening the box with the alternative means – with her hand for infants who previously saw her use the tool and with the tool for infants who previously saw her use her hand – and then gave the materials to the infant again. On this second trial, 26 of the 36 infants (X2 = 7.69, p = .0056) attempted to open the box by the method employed on the first trial. Thus infants who had initially seen and tried the tool strategy persisted with it even after observing the box opened by hand. Likewise, infants who had previously seen and opened the box by hand set the tool aside and opened the box by hand again, even though the experimenter had just demonstrated with the tool.

Conclusions. The results for the first trial support Tomasello’s and Gergely’s emphasis on infants’ attention to the means and the experimenter’s apparent intentions in imitation (cf., Carpenter, Call, & Tomasello, 2002; Gergely, Bekkering, & Király, 2002). The results for the second trial suggest that the first means learned to achieve an end is somehow privileged for infants. The initial means may establish a motor “attractor” following Thelen and Smith (1994) or it may later be primed by anticipating the effect associated with it earlier, as suggested by Elsner and Hommel (2001). Finally, means and ends may be subject to a mutual exclusivity constraint similar to that in early word-learning.