Learning foreign and native accents: the role of production and listening

Abstract

The present dissertation asks whether and to what extent producing, compared to listening to, accented words can contribute to accented word learning and accent learning more generally. Learning effects of accented speech production are compared with learning effects resulting from listening to accented speech. This is specifically asked in the global Question 1. By this comparison, conclusions can be drawn regarding the nature of learning and how learning mechanisms induced by listening and production relate to one another, which is asked in Question 2. Foreign accent learning is compared with native accent learning by presenting speech material recorded by L2 and L1 speakers. Moreover, the role of listeners’ native language background is investigated. The speakers presented during training and test are always different in order to test speaker-general learning. A further goal of this dissertation is to characterize the learning effects in terms of the processing levels where they are observed. This is done with different experimental paradigms. Reaction time and eye-tracking tasks investigate the effects of learning on online processing, and memory tasks look at the effects on memory recognition. The generality of learning with production is also tested by comparing learning with long-term familiar and unfamiliar accents. Further aspects that describe these learning effects refer to how long lasting they are and what the role of self-listening is. Finally, Question 3 scrutinizes the role of salience in accent learning and learning with production and listening.