Appreciating language conventions: thirteen-month-old Chinese infants understand that word generalization is shared practice

Abstract

Language is conventional because word meanings are shared among different people. The present study examined Chinese infants’ understanding of the language convention that different people should generalize words in the same way. Thirteen-month-old Mandarin-speaking Chinese infants repeatedly viewed a speaker providing a novel label for a target object in the presence of a distractor object. Next, the objects changed colour and infants viewed the same speaker and a new speaker providing the label for either the different coloured target or distractor. They were also asked by both speakers to locate the correct referent of the label. Results revealed that infants expected both speakers to generalize the label to objects that belonged to the target category. This is the first evidence demonstrating that Chinese infants perceive word generalization as a form of shared convention.

Examining the female‐talker default in experimental language acquisition research

“…In these studies, the research question often involved the perception or detection of language features/speaker properties in diverse contexts (e.g. Fecher & Johnson, 2019) and so using speakers of different genders allows for testing of children's ability to generalise across input sources (Siying & Renji, 2019; ter Haar & Levelt, 2018). A small number of studies alternated speaker genders as a way of differentiating between the training and test stages of the experiments (e.g.…”

“…CD, child development; CL, child language; DP, developmental psychology; DS, developmental science; I&CD, infant and child development; IB & D, infant behaviour and development; LA, language acquisition; LL&D, language learning and development.detection of language features/speaker properties in diverse contexts (e.g Fecher & Johnson, 2019). and so using speakers of different genders allows for testing of children's ability to generalise across input sources(Siying & Renji, 2019;ter Haar & Levelt, 2018). A small number of studies alternated speaker genders as a way of differentiating between the training and test stages of the experiments (e.g Bulgarelli & Bergelson, 2021;Verdine et al, 2017)…”