Language Comprehension: A New Look at Some Old Themes
In the last 25 years, many researchers have heeded the call to look for those "devious" measures that would uncover the child's language knowledge through comprehension even before the child could produce language. New assessments of comprehension have been pioneered that have been highly informative (see Golinkoff, Hirsh-Pasek, Cauley, & Gordon, 1987, for a review). Generally speaking, however, many of these efforts have been plagued with methodological difficulties. In the commonly used picturepointing technique, for example, young children often are puzzled by attempts to depict dynamic episodes in a static representation (Friedman & Stevenson, 1975; Cocking & McHale, 1981). In tasks that require children to act out commands like "Make the cat chase the dog," young children often are disposed to act on the dog without regard to the particular request at hand. Whether or not the child understands the structure in question, these tests can be viewed as tests of compliance rather than of comprehension. Thus, although there have been a number of attempts to describe early
language comprehension, our methods have constrained us. First, most of the methods can be used only with older children (greater than 28 months of age) who are already language users. Second, the field has created measures that are better at assessing the comprehension of nouns than at assessing the comprehension of verbs or other linguistic descriptions of actions and events.