ABSTRACT
The application of first-language (L1)-based research methods to second-language acquisition (SLA) has led researchers to question assumptions of the methodologies, with respect, in particular, to their reliability and validity (for consideration of metalinguistic judgments, see, e.g., Birdsong, 1989; Chaudron, 1983). The connection between the data collection technique and a subject’s behavior independent of the data collection context is always one that should concern the researcher. This chapter considers one recent methodology, the psycholinguistic technique of elicited imitation, in which a subject is presented with a spoken string of words in the language (usually a grammatical sentence) and asked to repeat it. The investigator records facts about the repetition and makes inferences about the state of the learner’s knowledge of the language. The basic idea is that if the subject’s grammar corresponds to the grammar used in producing the string, the imitation is more likely to be accurate. Specific inaccuracies may point to specific differences between the subject’s grammar and the grammar of the target string. The technique has been used in studies both of child language development and of adult foreign-language learning (see Chaudron & Russell, 1990; Lust, Chien, & Flynn, 1987, for reviews), but the precise parameters of performance and valid interpretation of learner competence have not been fully determined. We regard it as premature to view elicited imitation as a proven method for inferring learner competence, because a considerable amount of research needs to be conducted to understand how performance under imitation conditions compares with other methods and with learner’s underlying knowledge.
