Caregivers as experimenters: Reducing unfamiliarity helps shy children learn words.

Matt Hilton presented a paper at the XX Biennial International Conference on Infant Studies, New Orleans, LA, USA. (May 2016)

Authors: Hilton, M., Twomey, K. E., & Westermann, G.

Abstract: Shyness, defined as inhibited approach and/or discomfort in social situations involving novelty (Putnam, Gartstein & Rothbart, 2006), affects early language development. Shy children speak less (Crozier & Perkins, 2002) and have a lower productive vocabulary than less-shy peers (Smith Watts et al., 2014). Hilton and Westermann (in prep.) investigated whether shyness affects the processes involved in language acquisition. An experimenter presented 24-month-old children with sets of objects, each consisting of one novel and two familiar objects. The experimenter then requested the novel object using a novel pseudoword. Less-shy children reliably chose the novel object, but shy children did not. After a five minute break, each novel object was presented alongside two other previously-seen novel objects, and children’s retention of the word-object mapping was tested by asking for the target.  Shy children did not retain any word-object mappings, while less shy children retained mappings above levels expected by chance. These results indicate that the unfamiliarity of the testing environment led to an aversion to the novel object in shy children. The current study examined whether reducing the unfamiliarity of the testing environment facilitated children’s learning of word-object mappings. Two-year-old children (N = 24) took part in a task identical to that of Hilton & Westermann (in prep.) except that the child’s caregiver acted as the experimenter, presenting each set of objects and asking for the target on each trial. To simplify the procedure, each novel object was labelled three times during each request (as opposed to six times in Hilton & Westermann; in prep); each set was therefore presented twice, to ensure that each label was heard the same number of times as in the original study. Children were classified as shy or less-shy according to a median split on their scores taken from the shyness scale of the ECBQ (Putnam et al., 2006). In contrast to the children in Hilton & Westermann’s (in prep.) study, shy children mapped the novel label to the novel object equally as successfully as less-shy children when their caregiver ran the experiment. Further, only shy children retained the word-object mappings that they had formed [ t (11) = 2.42, p = .03, d = .70 ]. Less-shy children did not retain these mappings above levels expected by chance [ t (10) = .48, p = .64, ns ]. Taken together with Hilton & Westermann (in prep.), the current findings indicate that shy children learn word-object mappings when those mappings are presented by a highly familiar adult but not an unfamiliar experimenter. Increasing the familiarity of the environment may help shy children to overcome their aversion to novelty, allowing them to attend to the novel word-object mappings sufficiently to retain them. The finding that less-shy children do not learn word-object mappings suggests that they benefit from some novelty in their learning environment. Overall, the current study offers converging evidence that shyness is an individual difference that impacts on the processes used to acquire language.