Elicited imitation as a window into developmental stages

In the second language acquisition literature, data of naturally occurring language use are considered the most ideal data to make statements about second-language (L2) development. This study examines to what extent experimentally elicited data can provide an equally valid basis for determining L2 development, by testing predictions derived from Processability Theory regarding the L2 acquisition of the German case system. Using naturally occurring language data, previous research on L2 German case acquisition has uncovered three developmental stages. The present cross-sectional study investigates whether the same stages occur in data obtained from an experimental task (i.e., a computer oral elicited imitation task (OEIT). Thirty-six university L2 learners of German participated in the study. The results show that the elicited data prove comparable to the naturally occurring data. As such, this study corroborates a previous validation study on developmental stages in L2 English, which demonstrated the comparability of naturally occurring and experimentally elicited data. In addition, concerning methodological advancement of the OEIT design, the present study proposes to include a direct measure of comprehension.