Chasing, causality and goals: Conceptual understanding at 9 months and its relationship to later language use.

Samantha Durrant, Caroline Rowland, Franklin Chang, Jessop, Amy Bidgood and Michelle Peter, presented this paper at the Biennial International Conference on Infant Studies in New Orleans, LA, USA in 2016.


Language acquisition relies on the development of a basic conceptual understanding. For example, a child cannot learn the relationship between semantic (e.g. agent, patient) and syntactic (e.g. subject, object) roles in transitive sentences unless they are already able to identify agents and patients in actions. Thus, it seems logical to predict that children who start to identify the components of conceptual events earlier in life might have an advantage when it comes to learning language. However, we do not yet know whether there are early individual differences in the development of conceptual understanding, and whether these differences predict the speed of children’s later language gains.In this study, we tested the conceptual understanding of 80 9-month-old children taking part in the longitudinal Language 0-5 Project in three different tasks. In task 1, we used a preferential looking paradigm to test the children’s ability to distinguish basic chasing events (one dot on a screen chasing another) from random motion (c.f. Frankenhuis et al., 2013). In task 2, we used a familiarisation paradigm to test the children’s ability to distinguish causal and non-causal actions in inanimate objects (e.g. boxes and cones; c.f. Oakes & Cohen, 1990). In task 3, we used a familiarisation paradigm to test if the children would attribute goals to a novel non-human agent (e.g. cubes and cylinders; c.f. Luo, 2011). We use these data to test the predictions that a) the children’s performance on the three tasks will correlate, even when we partial out attention span, indicating a common basis of conceptual understanding, and b) the children’s performance on the three tasks will correlate with their concurrent language and gesture ability, as measured by the UK-CDI: Words and Gestures.   A preliminary analysis of data from the causality and goals studies for 51 of the 80 children replicated the results of the original studies: children looked longer to a novel causal action than a novel non-causal action F(1,51)=, p=.009, ηp²=.175 as in Oakes and Cohen (1990), and looked longer to a new goal than an old goal F(1,51)=44.4, p=.048, ηp²=.076 as in Luo (2011). However, these preliminary data did not show that performance on these two tasks was related, nor was there a relationship between performance and concurrent vocabulary (expressive and receptive) or gesture use.   The full dataset will be presented at the conference. We will also discuss plans to test the predictions that performance on the conceptual tasks will predict syntactic development at 30 months (e.g. performance on the causal task will predict children’s ability to comprehend agents in language at 30 months), and that a composite score from all three tasks will predict cognitive abilities (as measured by the Bayley’s Infant Development Scale) at 16-months. The implications for the role of conceptual understanding in later language and cognitive development will be discussed.