The Benefits of Active Experience in Action Understanding

• In Event: 1-145 - Mechanisms Underlying the Development of Action Understanding in Infancy

Integrative Statement

The ability to understand goals of others’ action plays a crucial role in social interaction and development. Prior research has provided insights into infants’ emerging sensitivity to the intentional structure of others’ actions (Woodward, 1998), and, further suggested that infants’ own actions are related to their understanding of others’ actions. First, individual variation in infants’ motor abilities correlates with their sensitivity to others’ goals (Sommerville & Woodward, 2005). Second, interventions that support infants’ engagement in new actions leads to increased sensitivity to others’ goals (Gerson et al., 2015; Sommerville et al., 2005). In the current study, we explored whether and how variations in infant motor development may interact with trained experience in supporting infants' action understanding.

We tested 80 infants (M = 8.90 months). In the Training session, infants were trained to use a cane to obtain an out of reach toy. Each infant was also tested in a goal-imitation paradigm (Hamlin et al., 2008) on their understanding of an experimenter's cane actions as goal-directed. Infants watched an experimenter reach toward one of two small toys with the cane. Then they were given the opportunity to select between the toys. Prior findings (Gerson & Woodward, 2012) have indicated that infants’ tendency to select the experimenter’s goal object reflects their understanding of the action as goal-directed. In the training-first condition (n=40), infants received training prior to the goal-imitation test, and in the training-second condition (n=40) the order of these tasks was reversed. We obtained parent report Early Motor Questionnaire (EMQ) ratings for each infant (Libertus, 2013).

Infants’ responses in the imitation paradigm varied depending on whether they had first received training with the cane: in the training-first condition, infants chose the experimenter’s goal more often than expected by chance (t(39)= 4.27, p<0.001, Figure 1), and this was not the case in the training-second condition (p>0.4). We next asked whether variability in infants’ motor skill, as indicated by their success in using the cane and their EMQ scores, was related to their goal imitation. Infants’ coordinated use of the cane during training was correlated with their goal imitation (r=.28, p<0.05, Figure 2). This relation did not differ reliably between the two testing orders and, if anything, was somewhat stronger in the training-second condition. That is, infants’ readiness to learn was a general predictor of their response to the experimenter’s cane actions. EMQ scores did not predict goal-imitation scores, but they did predict infants’ skill with the cane during pretest (prior to training) (r=0.27, p<0.05), and pretest scores predicted infants’ cane skill at the end of training (r=0.41 p<.001).

Cane training led infants to respond systematically to the goal-structure of an experimenter’s cane actions. In addition, our findings highlight that there is variability in infants’ individual readiness to learn and understand means-end tool use. Infants’ planful coordination of cane action was also a predictor of their responses to the experimenter's cane actions and goals. These results shed light on the benefits of active prior experience and individual readiness to learn in early development.

Authors

• Haerin Chung, University of Chicago
  Presenting Author

• Courtney Filippi, NIMH
  Non-Presenting Author

• Amanda Woodward, University of Chicago
  Non-Presenting Author