Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)


Linguistics (Interdepartmental Program)

First Advisor

George Yule


Proposals regarding the beneficial effects of elements of modified interaction such as clarification requests and confirmation checks on SLA have been taken up by a number of researchers who have found evidence for their existence in discourse involving NNSs. These investigators have assumed that the presence of such interactional features are beneficial to language learning, yet there is little empirical evidence to support a causal relationship between the presence of these discourse features and change in the performance of the learner toward target-language forms. Furthermore, the potential benefits of modified interactions have not been evaluated against other classroom practices. This study examines the English pronunciation of Chinese L1 learners following four classroom-like interventions reflecting current pedagogical practice: teacher-led drill; directed self-study in a language lab setting; time along for revision and reflection; and interactions involving clarification requests. NS naive listeners judged whether the L2 learners' pronunciation was more or less target-like before, immediately after, and at a later point in time after one of the four learning events. In addition to the question of whether conversational modifications could be shown to affect the spoken performance of ESL learners, four effects of the different input types on the spoken output of the learners were found: immediate improvement; delayed improvement; residual improvement; and restructured improvement. There was no overwhelming evidence for the effect of one learning event over another, which prompted the investigation of the effects of input type in terms of individual leaner behavior. While no generalizations are made in terms of group effects the "no difference" result makes one point clear--there is no evidence of acquisition of native-like phonological form as a result of language use in modified interaction. While this result has obvious implications for the current SLA theory regarding the effects of negotiated interaction, it has also led to some observations about the classroom learning events examined which are discussed. In addition to making these general observations, this study addresses some of the limitations of the study and suggests how they might be accommodated in further research.