Pronunciation learning through captioned videos: Gains in L2 speech perception and production

Abstract

This doctoral dissertation examines the benefits of extended exposure to multimodal input through captioned videos for second language pronunciation development. It investigates the effects of TV viewing on both L2 speech processing and perceptual and productive accuracy in learners’ L2 phonology. It also analyses learners' eye-movements while watching captioned videos in order to relate the amount of on-screen text processing to pronunciation gains. Ninety Spanish/Catalan adult learners of English (EFL) were tested on speech processing skills (segmentation, speed of lexical access, and sentence processing) and phonological accuracy in perception (ABX discrimination) and production (accentedness ratings) before and after an 8- week treatment consisting of regular exposure to audiovisual materials. Participants were randomly assigned to four experimental conditions involving two viewing modes (captioned or uncaptioned) and two task focus conditions (focus on phonetic form or focus on meaning). The results revealed that exposure to authentic audiovisual materials in English can benefit the L2 pronunciation of post-intermediate/advanced EFL learners irrespective of viewing mode. Whereas previous studies had found larger benefits for captioned than uncaptioned viewing for speech processing and segmentation after short (e.g. single-session) exposures, the relatively long exposure treatment administered in the current study and inclusion of a form or a meaning-focused condition might have washed away potential advantages of one viewing mode over another. Viewing treatment benefits on L2 speech processing were larger on tasks assessing sentence-level than word- or segment-level gains. No significant benefits were found for phonological accuracy in perception. In production, the results revealed an interplay between viewing mode and task focus, indicating that a focus on phonetic form improved pronunciation only in the absence of captions, whereas captioned viewing led to pronunciation gains as long as there was no focus on phonetic form. Cognitive

overload might explain why no benefits were obtained when attention was directed to pronunciation in a captioned viewing mode. Although large individual differences characterized L2 learners' caption reading behaviour, which was influenced by material-related as well as learner-related factors, the results indicated a relationship between amount of subtitle processing and foreign accent reduction. Viewing mode moderated foreign accent reduction, as incidental learning of pronunciation occurred only through exposure to captioned viewing without a focus on pronunciation, whereas in the absence of captions gains were driven by an intentional focus on pronunciation. Taken as a whole, this dissertation suggests that enriching the limited L2 input learners are exposed to in foreign language settings through the viewing of authentic audiovisual materials in target language may enhance L2 pronunciation development.

Esta tesis doctoral examina los beneficios de la exposición prolongada a programas de televisión en inglés con subtítulos para mejorar la pronunciación. La tesis investiga los efectos de la exposición multimodal tanto en el procesamiento del habla como en la corrección fonético-fonológica a nivel perceptivo y productivo del inglés como lengua extranjera. Con este propósito, el estudio analiza los movimientos oculares de los estudiantes mientras visualizan videos subtitulados y poder así relacionar la cantidad del texto procesado en los subtítulos con posibles mejoras en la percepción y producción fonético-fonológica. Noventa estudiantes universitarios, estudiantes de inglés como segunda lengua, fueron evaluados respecto de sus habilidades para el procesamiento del habla en inglés (segmentación, velocidad de acceso léxico y procesamiento del habla en frases) y corrección fonológica en la percepción (discriminación ABX) y producción (nivel perceptivo de acento extranjero) antes y después de un tratamiento de 8 semanas de exposición regular a materiales audiovisuales (películas de video). Los participantes fueron asignados aleatoriamente a cuatro condiciones experimentales que implicaban dos modos de visualización (con subtítulos o sin subtítulos) y dos condiciones de enfoque de la tarea de visualización de los vídeos (un enfoque en el que se induce una atención a la forma fonética de las palabras u otro en el que se induce una atención al significado). Esta tesis demuestra que la visualización de materiales audiovisuales auténticos en la lengua extranjera puede, además de facilitar un mayor contacto con la lengua inglesa en contextos de instrucción sometidos a un input lingüístico mínimo, mejorar el desarrollo de la pronunciación en una segunda lengua.