Individual Differences and the Shape Bias—Lessons From Late Talkers

• In Event: 3-036 - 30 Years of Shape Bias Research: Lessons for Word Learning, Atypical Development, Categorization and Beyond

Integrative Statement

Thirty years of shape bias research has taught us much about the developmental mechanisms driving word learning. Central among these lessons is that the very process of learning words can bias a child’s attention to properties that are relevant to category membership. For example, most typically developing (TD) children have vocabularies dominated by names for categories organized by similarity in shape and develop a bias to attend to shape (i.e., the shape bias). Some TD children, however, have vocabularies dominated by names for categories organized by material and develop a bias to attend to material (i.e., the material bias). In this talk, we ask how this relationship between vocabulary and shape bias development can help us better understand children with language delays and difficulties, particularly late talking (LT) children.

LT children are those who fall significantly behind in productive vocabulary for their age. As a group, LT are less likely than their TD peers to show a shape bias, though there is a lot of heterogeneity in their developmental trajectories. Some LT eventually “catch-up” to their TD peers, but many others continue to be delayed. In two studies, we use the shape bias as a tool for exploring this heterogeneity.

In study one, we examined the relationship between vocabulary structure (via the MCDI) and children’s attention to shape versus material during novel noun generalization in a sample of 66 16-30 month olds (half LT). We found that LT children were less likely to generalize novel names by shape, X2(1)=11.40, p=.0007. As can be seen in Figure 1, we also found a significant interaction between group and the shape-based nouns in a child’s vocabulary X2(1)=4.66, p=.031, such that the more shape-based nouns TD children knew, the more likely they were to attend to shape. There was much more heterogeneity among LT, and they did not show an effect of vocabulary structure.

In study two we examined how LT’s vocabulary structure related to whether or not they eventually caught up to a more typical vocabulary size. Participants were 899 children contributing MCDI scores at 2 time points to Wordbank (www.wordbank.stanford.edu). On average, children were 16 months old at time 1 and 27 months at time 2. At time 1, there were 605 TD and 294 LT. Of these, 97 LT continued to be LT at time 2, and 173 LT now had typical vocabularies. We compared the vocabulary structure of these two LT groups at time 1. Controlling for total vocabulary size, LT children who caught up tended to know more shape-based nouns at time 1 than those who stayed LT, X2(1)=9.37, p=.002. However, even those who caught up had fewer shape-based nouns in their vocabulary than those who were TD at time 1, X2(1)=106.3, p<.00001, suggesting that their “catching up” might not be as categorical as it seems.

Together, these studies bring new insight into sources of LT strengths and difficulties and are an important first step towards understanding the heterogeneity of this population.

Authors

• Lynn K Perry, University of Miami
  Presenting Author

• Sarah Kucker, Oklahoma State University
  Non-Presenting Author

• Jessica S Horst, University of Sussex
  Non-Presenting Author

• Larissa K Samuelson, University of East Anglia
  Non-Presenting Author