particular age group of children. It is also probable, that the children were more familiar with
the use of the particular artifacts in their everyday life than with animals and thus, had
mapped them to the relevant semantic domain.
The above line of interpretation is also supported by relevant research to other domains.
Braisby and Dockrell (1999) found that young children are sensitive to the semantic
distinction between natural kind and colour terms and that the difference can be explained
on semantic grounds. For example, natural kind and colour terms differ semantically, with
natural kind terms having sharp, clear boundaries and colour terms having unclear or variable
boundaries.
The findings can have important implications about the development of a theory of word
learning from context which could take into account the semantic domain the target words
belong to.
8.3.3 Nature of the input
Another focus of the present study was to what extent the nature of the input influences word
learning from context. It was found that children used the linguistic input to infer the
meanings of novel words. Those findings are discussed to a greater extent in the following
subsections.
8.3.3.1 Children can use the linguistic input to infer the meanings of the novel
words
The present study demonstrated that children can Ieam novel words from listening to stories.
The results obtained support and extend previous studies (Leung and Pikulski, 1990; Eller
et al. 1998; Elley, 1989; Robbins and Ehri, 1994; Senechal and Cornell, 1993) in various
ways. Both experiments indicated that children performed better in those tasks where input
and assessment matched. For example, during Experiment 1 the children in the Inference
condition (input) performed better on the inference task (assessment) than the children in the
other conditions. The same pattern was also found for the children in the Analogy, Definition
and Lexical contrast conditions and their performance on the corresponding tasks. In a similar
way, Experiment 2 demonstrated that the children in the Lexical contrast group performed
better on the contrast task than the children in the other groups, as well as that the children