Is a linguistic model needed to build abstract event representations?
Year: 2024
Type: preprint
Abstract: This project seeks to explore the abstract representation and complexity of two-participant events in comparison to one-participant events, particularly focusing on whether language plays a role in the abstract encoding of two-participant events Shukla and de Villiers (2021) addressed this question with a study involving infants (aged 12-24 months) and adults (in both a control and a verbally shadowing conditions) using an anticipatory looking time paradigm. While adults on the control condition demonstrated the ability to generalize across both one-participant and two-participant events, that was not the case for in verbally shadowing adults nor, critically, in infants. The authors concluded language mastery and use might essential to correctly form generalizations on two participant events but not on one participant events. We find these results intriguing since the oldest infants (24 month olds), already language users, did not display the same generalization ability. To further investigate the complexity of two-participant events, we plan to use a simplified imitation task, loosely inspired by Shuckla and de Villiers (2021) and de Villiers (2014) paradigm, with children as our initial focus. The task will involve two sets, A and B, each consisting of action figurines. Within each set, there will be four distinct tokens featuring the same animal. In the activity, each figurine from set A will be paired with a corresponding one from set B. We will demonstrate the figurine from set A acting upon the figurine from set B in three of these pairs. Subsequently, the child will be given the fourth pair and prompted to "repeat what they've seen." Besides the method (an imitation task instead of an anticipatory looking task) the task will differ from Shuckla and de Villiers in two key aspects: first it will only feature animate participants (Shuckla and de Villiers’ findings were conditional on the agent being animate), and second, it will introduce more variability to encourage generalization(different events and participants will be tested)
Institution: Harvard University
Language: English
Subfield: Developmental and Educational Psychology
Field: Psychology
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W7128920950