In contrast to the qualitative, hypothesis-raising research of Part II, the studies in Part III single out certain variables for more controlled investigation. The main variables include the cognitive skills involved, the type of lexical unit being learned, and the method of instruction employed.

Chapter 7, by Lynne Yang, reports on a study of the acquisition of L2 vocabulary under controlled experimental conditions. Three cognitive skills were measured over the course of learning: conscious word translation, word recognition, and semantic priming. Twenty-nine monolingual speakers of English were taught a miniature artificial L2 in a controlled laboratory setting, using computerized tutorials over a 5-week period. The three tests were administered periodically over the learning period, for both the L2 and LI (control) vocabulary. The results suggest a different time course for each of the three developing skills. Conscious translation ability developed most rapidly. In contrast, automated skills at both the meaning level (semantic priming) and the phonological level (word recognition) lagged considerably behind. This is consonant with cognitive studies of the development of automaticity. The results do confirm, however, that real changes at the cognitive level can occur during a relatively short learning period; indeed, at the end of this 5-week learning period, the manner of processing of L2 vocabulary for these learners began to approximate that of the L1 control.

Chapter 8, by Pierre (J. L.) Arnaud and Sandra J. Savignon, describes the passive knowledge of rare English words and complex English lexical idioms by 236 francophone learners at four levels ranging from first-year university majors to teachers in mid-career.