Final Report Summary - COMINTENT (Information transmission in language: do infants perceive communicative intent?)
In this project, we explored when, and by what means, infants start perceiving the communicative role of verbal acts. We asked whether and under what conditions infants expect agents to share a language, and what inferences they draw if agents share such a communicative code. In the course of the project, we further broadened the scope investigations, asking how infants and adults learn about important carriers of communicative load in language, including words and their syntactic organization. We also investigated the low-level cues observers use to track and remember socially relevant actors.
Since the beginning of the project, we addressed these questions in about 300 infants of different ages, and at least 700 adult participants. Further, we started developing a software package for the analysis of eye tracking data using the R programming language, and developed several computational models of aspects of language acquisition. As the fellow acquired data until the end of his stay at the host institution, the results below are still preliminary, being analyzed and most are not yet submitted to scientific journals. However, we estimate that the data and models acquired during the fellowship duration will lead to at least 10 peer reviewed papers, two of which will be submitted to Science and the Proceedings of the National Academy of Sciences of the United States, while the others will be submitted to top journals in the field. In addition, the results from this work have triggered further questions that will be addressed both at the host institution and at the fellow’s new laboratory at City University London.
Since the beginning of the project, we addressed these questions in about 300 infants of different ages, and at least 700 adult participants. Further, we started developing a software package for the analysis of eye tracking data using the R programming language, and developed several computational models of aspects of language acquisition. As the fellow acquired data until the end of his stay at the host institution, the results below are still preliminary, being analyzed and most are not yet submitted to scientific journals. However, we estimate that the data and models acquired during the fellowship duration will lead to at least 10 peer reviewed papers, two of which will be submitted to Science and the Proceedings of the National Academy of Sciences of the United States, while the others will be submitted to top journals in the field. In addition, the results from this work have triggered further questions that will be addressed both at the host institution and at the fellow’s new laboratory at City University London.