Shyness affects word learning: evidence from eye-tracking

Lancaster University PhD student Matt Hilton presented a poster at the Society for Research in Child Development (SRCD) 2015 Biennial Meeting in Philadelphia, PA,  19-21 March, 2015.

Authors: Matt Hilton and Gert Westermann, Lancaster University 

Abstract: Shyness predicts productive vocabulary size at 24 months of age (Prior et al., 2008), however, the mechanisms underlying this link are unknown. One process used by children to assign labels to objects is fast-mapping (Carey & Bartlett, 1978). In the lab, fast mapping tasks involve a child mapping a novel label to a novel object, based on the assumption that the novel label will not map onto any accompanying familiar objects for which a label is already known. Recent research (Hilton & Westermann, in preparation) found that shy children did not demonstrate fast-mapping: when presented with one novel and two familiar objects and asked for a novel pseudoword, shy children only selected the novel object at levels expected by chance, unlike their less-shy peers. It was not known, however, whether shy children did not form the mapping, or whether they made the mapping but had difficulty demonstrating their knowledge of it. In order to investigate this question, sixteen 20-month-old and sixteen 26-month-old children were tested on a fast-mapping task using eye tracking. Measures of each child’s receptive and productive vocabulary were taken alongside a parent-report measure of shyness taken from the Early Childhood Behaviour Questionnaire (ECBQ; Putnam, Gartstein & Rothbart, 2006). Children were presented with pictures of two familiar and one novel object on a screen, and either the novel or a familiar object was labelled. Labels were played three times (e.g. “Look, a blicket! Can you see the blicket? Wow, a blicket! ”). An eye-tracker (Tobii X120) measured where the child was looking during each trial. Analyses were run on proportion of time spent looking at the novel object after hearing the novel label during the short time interval [r = -.24, p = .40] or the long time interval [r = -.10, p = .72]. However, at 26 months shyness was negatively correlated with proportion of time spent looking at the novel object post-labelling during the long time interval [r = -.76, p = .001], but not during the short time interval [r = -.32, p = .26]. This result indicates that at this age, shy children  maintained less attention to the novel object than less-shy children during novel-label trials, suggesting that there may indeed be differences in the way that shy children attend to novel objects during word learning. Given that the effect of shyness on novel-object preference emerged between 20 and 26 months of age, a better understanding of the link between shyness and language learning processes  could lead to early interventions to offset the potential for shyness to negatively impact on later language development.