ABSTRACT
Applying an eye-tracking technique, we tested early verb understanding in 48 infants aged 9 and 10 months. Infants saw two objects presented side by side and heard a verb that referred to a common action with one of these objects (e.g., eating relating to a banana). The verbs were spoken by the parent in an interrogative manner in order to elicit a looking behavior in the infant. Results showed that 9-month-old infants did not show recognition of our test words. However, 10-month-old infants were able to understand a number of the tested verbs. In the discussion, we relate our findings to the nature of early verb representations.
Acknowledgments
This study was part of the interdisciplinary project IP-13 on “Towards a model for the incremental co-development of linguistic and conceptual representation of actions“ at the Cluster of Excellence Cognitive Interaction Technology “CITEC”‘ (EXC 277), Bielefeld University funded by the German Research Foundation (DFG). We are grateful to the participating children and their parents, to Elena Justus for piloting the study and making the audio recordings, and to Marc Rohlfing and Malte Schilling for their data analysis scripts. We also wish to thank the readers of our poster at the ICIS 2016 for constructive discussions.
Notes
1 In a pilot study, we asked parents whether they used these verbs with their infants. The verbs were confirmed as being common. Two further verbs were reported as being problematic: the verb to drink because parents reported that they often use the verb to eat for drinking milk or for nursing; the verb to swing because their infants did not have yet any experience with a swing. We therefore excluded these two items from the final verb selection.
2 The assumption of normality was assessed with the Shapiro–Wilk test. The statistics for both the 9-month-olds (SW = .956, df = 27, p = .292) and the 10-month-olds (SW = .913, df = 21, p = .063) suggested that normality was a reasonable assumption.