A new method for evaluating the grammatical complexity of preschool natural language corpora is introduced. In the Index of Productive Syntax, occurrences of 56 syntactic and morphological forms are counted, yielding a total score and subscores for noun phrases, verb phrases, questions/negations, and sentence structures. Development of the index and analyses of its reliability and age-sensitivity when applied to language samples of 2- to 4-year-olds are described. Some advantages and limitations of the index as a research and clinical instrument are also discussed.

A set of procedures called referential mapping has been used to assign meaning to spontaneous vocalizations of an African Grey parrot. These spontaneous vocalizations were combinations and phonological variations of specific vocal English labels that the bird had previously acquired in a study on interspecies communication. These recombinations were not necessarily intentionally used to describe or request novel objects or circumstances, even though our earlier data demonstrated that the bird could use vocalizations referentially and intentionally to identify, request, refuse, and categorize various objects. The procedures for dealing with this bird's spontaneous modifications neither attempted to nor needed to evaluate the intentionality of his behavior. Rather, the procedures instructed his trainers to respond to the novel speech acts as though he were intentionally commenting about or requesting objects, actions, or information. The success of such procedures in attaching functional significance to his spontaneous vocalizations and in encouraging subsequent innovation suggests that the technique might be applicable to intervention programs for humans with specific communicative deficits.

Environmentally contaminated speech errors occur when material derived from the speaker's environment but irrelevant to the speaker's intended utterance is erroneously incorporated into speech. Such material may include the names of objects at which speakers are looking, words that speakers are concurrently reading, segments of speech that speakers overhear, or properties of objects in the environment. It was found that although environmental contaminations display less phonological facilitation than other speech error types, they display some semantic facilitation. If the target item and interfering item are from the same syntactic category, a word substitution is more likely to result than a word blend, whereas if the items are from different categories, a blend is more likely. It is hypothesized that environmental contamination occurs at a high level of processing, but with a relatively late insertion point.

The relatively unexamined question of what kinds of narratives are memorable as whole units of discourse was addressed by studying memory for 288 children's personal narratives. Each narrative was assigned a structural category in story grammar analysis and Labov's highpoint analysis. Narratives were also rated for the presence or absence of sensational content. An incidental free recall of these narratives was then requested from 116 subjects. As predicted, an analysis of variance of recall frequencies revealed a main effect of sensational content; sensational narratives were recalled more frequently than nonsensational ones. Although there was no main effect of highpoint structure, that variable did interact with sensational content; in the absence of sensational content, highpoint structural sophistication predicted enhanced recall. Contrary to prediction, there were no main effects of highpoint structure or story grammar. The study is interpreted as providing partial validation of highpoint analysis, as well as revealing the independent, dominant contribution of content to memory for others' narratives as whole units.

Children in the preschool years develop a linguistic repertoire of narrative skills as they adapt their ways of representing events to different occasions of language use. The present study examines the abilities of primary school children to draw upon their repertoire of narrative skills in the service of language tasks. Children in grades K-2 were shown a shortened version of the silent film, The Red Balloon, and were asked to perform three narrative tasks: (a) produce an on-line narration of a 3-minute segment from the film, (b) recount the film's events as a news report, and (c) recount the film's events as an embellished story. The narrative texts produced for each task were subjected to analyses of linguistic markers of genre differences. The findings revealed very subtle distinctions between the narrative texts produced for the three genre tasks, leading to the conclusion that primary grade children have only nascent ability to apply their genre knowledge to school language tasks.