RPF 11/12 Towards a better understanding of the needs of English Learners of foreign languages

Project: Research project (funded)Internal pump-priming

Project Details


Most people agree that it is hard to learn a new language as an adult, but relatively little research tries to link why language learning is difficult with how we can help learners do better. Of the research that there is, most looks at sentence structure, rather than speech sounds, and deals with the needs of people learning English, rather than the needs of native English speakers learning other languages.

This project will address this imbalance by highlighting the need for research on English learners of other languages: we will carry out a study, then, share our results with other researchers, within and beyond York, through a public conference and an inter-disciplinary workshop to generate further research.

Our study replicates an experiment which showed that French listeners can’t tell which syllable in a word is stressed (Dupoux et al 2001, 2007). We suspect that English listeners have the same difficulty, but for a different reason: in French, stress is always on the last syllable, so listeners don’t pay attention to stress at all when they learn new words; we think English listeners do pay attention to stress, but only ‘hear’ it if it sounds like English stress. If we’re right, then, there is a way to help, using methods already developed to train learners how to listen differently.

Key findings

The main findings from the study were that i) the phonetic realisation of stress does impact on the ability of English listeners to detect stress in a 'sequence recall' task (i.e. which requires encoding of the targets into working memory); but, ii) that English listeners also have difficulty in identifying the position of stress in targets (in a simple 'same/different?' AX task) when the correlates of stress differ from those used in English.

This means that English listeners show the same surface symptoms of 'stress deafness' as have been observed for French listeners, but the underlying cause is different; it appears that English listeners only encode stress into working memory representations if they can hear/identify it, and they only reliably hear/identify stress when the correlates match those used in English.

Follow-on applications
A funding application was made to the British Academy/Leverhulme Small Grant scheme in November 2012 [£9.922; R14917; not awarded].
Effective start/end date7/12/117/12/11


  • Unknown