Isabelle Darcy

Plenary talk: Improving pronunciation in spontaneous speech? A comparison of two instructional methods

Abstract

Pronunciation instruction is still underemphasized in many language programs and in teacher training curricula. One possible reason behind this resistance is the uncertainty whether pronunciation instruction “works” and why. While mounting evidence from research suggests that pronunciation instruction “works” to improve specific pronunciation features (Lee, Jang & Plonsky 2014), it remains unclear how much these improvements also impact global comprehensibility and fluency (Thomson & Derwing, 2014). In addition, research publications often remain superficial in their explanations of how exactly pronunciation instruction was implemented. The lack of clearly defined principles behind the reported effectiveness of methods, combined with the fact that only few such studies are classroom based, makes it difficult for teachers to envision ways to apply these methods in their own classrooms. This presentation first reviews two fundamental components of pronunciation improvement (automatization and generalization to spontaneous speech), and presents their psycholinguistic underpinnings. Second, we compare the effectiveness of two teaching methods in terms of how well they foster these components: explicit instruction (e.g. Saito, 2012) and communicative approaches that go beyond explicit instruction. The results help outline clear psycholinguistic principles behind the effectiveness of specific pronunciation instruction methods.

Kazuya Saito

Plenary talk: How research helps assess, develop and teach comprehensible L2 pronunciation: An overview

Abstract

Over the past 20 years, there has been a recognised paradigm shift in the field of second language (L2) pronunciation teaching research. The new proposal has emphasized the importance of instruction not as a de-contextualized, nativist form of language development (reminiscent of audio-lingualism), but as a way to help L2 learners attain “comprehensible” rather than “nativelike” pronunciation. In this talk, I would like to provide a state-of-the-art review on a range of studies in instructed L2 speech learning including my own work, targeting to correspond to two major issues in the field. First, I will discuss which pronunciation features are relevant (or irrelevant) to listeners’ successful understanding of L2 speech, and thus should be taught as a priority (i.e., what-to-teach issues). Second, I will introduce a series of quasi-experimental studies which have aimed to find an optimal way to teach L2 pronunciation via a range of communicative focus-on-form techniques, such as explicit instruction, focused tasks and interactional feedback (i.e., how-to-teach issues).

Mª Luisa García-Lecumberri

Plenary talk: Segmental Foreign Accent

Abstract

A foreign language (FL) phonetic system is known to be particularly influenced by the native language (NL) resulting in what is called a foreign accent (FA). In this talk we will examine segmental FA from a production and a perceptual point of view. Ideally, as learning progresses, learners' productions would adapt to the target categories by separating from those of the NL as they approximate the FL. Using a bi-directional English-Spanish corpus we show an interesting interplay of NL and FL influences, showing various levels of adaptation to the FL, i.e., degrees of deviation from the FL. In order to analyse the perceptual effects of different deviations from the norm, we examine acoustically manipulated foreign accented segments and their saliency for listeners from various linguistic backgrounds. Differences and interactions are found amongst segments and listeners in the degree of deviation required to be considered foreign-accented. Results are discussed in terms of functionality and perceptual saliency of foreign accented segments.