The content of infants’ object representations is influenced by social context

• In Event: 2-031 - New advances in the study of object representations in infancy: Formation, structure, and content

Integrative Statement

Infants form representations of hidden objects, but are limited in their ability to maintain information about the objects’ specific features (e.g. shape). For example, 6-month-old infants who viewed two featurally distinct objects hidden sequentially behind two occluders consistently remembered the featural identity of only the last-hidden object, and failed to encode the features of the first-hidden object (Kibbe & Leslie, 2011, 2016, in press). Outside of the laboratory, infants often observe objects as the target of other agent’s actions, intentions, and communication. An agent may express a desire for the red object, or attempt to teach the infant a label for the round object, making some features more socially relevant than others. Here, we asked whether the social relevance of object features impacts how and when infants encode those features.

In Experiment 1, we asked whether 6-month-olds would preferentially encode an object’s features if the features were relevant to an agent’s goal. We first familiarized infants to an agent who, following an auditory cue, preferentially reached for one object over another, featurally-distinct object (after Woodward, 1998). At test, we then occluded the objects sequentially (after Kibbe & Leslie, 2011, 2016, in press), always hiding the goal object first. Then, following the auditory cue, we removed one of the occluders to reveal either the original hidden object (control outcome) or the other object (swap outcome). Half of the infants saw the first-hidden (goal) object revealed (which infants typically forget in non-social contexts), the other half saw the last-hidden (non-goal) object revealed (which infants typically remember in non-social contexts). We used two measures to assess infants’ representations of the objects. First, we measured whether infants would make predictive gazes to the location of the hidden goal object before an object was revealed. Second, we measured infants’ looking times to each outcome (control or swap) following the removal of an occluder. We found that infants made predictive gazes to the goal location significantly above chance, but tended to look equally at both outcomes in both the first-hidden and last-hidden locations. These results suggest that 6-month-olds may prioritize representing relationships between agent and action over encoding objects’ specific features, a strategy that may be adaptive given infants’ limited working memory.

In Experiment 2, we asked whether 6-month-olds would preferentially encode an object’s features if the features were relevant to a pedagogical communication from an agent. We familiarized infants to an agent who pointed to and labeled one object over another, featurally-distinct object (“Wow! A blick!”). At Test, we hid the objects sequentially (as in Experiment 1), with the labeled object hidden first. We then removed the occluder hiding the first-hidden (labeled) object (which infants typically forget in non-social contexts) and revealed either the control outcome or the swap outcome. We found that infants tended to look longer at the swap outcome versus the control outcome, suggesting that they encoded the feature of this object.

Together, these experiments suggest that different social contexts differentially impact what is encoded in the object representation.

Authors

• Jessica R. Beal, Boston University
  Presenting Author

• Melissa M. Kibbe, Boston University
  Non-Presenting Author