Selective imitation of complex sentences in a preschool setting.

Abstract

An experiment was conducted to determine the conditions under which selective imitation of relative clauses could be produced in preschool. Seventy-two preschoolers took part in a 2 (modeling, imitation training) X 2 (instructions, no instructions) X 2 (reinforcement, no reinforcement) X 3 (baseline, Probe 1, Probe 2) factorial design. Imitation training was more effective than modeling alone, and instructions and reinforcement were effective in an additive manner. Relative clause usage increased significantly after imitation training but not after modeling or no treatment. The results are discussed in terms of the efficiency of various teacher techniques in increasing language usage. Implications for the role of imitation in language development are also discussed.