Individual Differences in L2 Speech Perception: The Role of Phonological Memory and Lipreading Ability

Abstract

The current study investigated the effects of individual differences in lipreading ability, working memory (WM), and phonological short‐term memory (PSTM) on second language (L2) speech perception. An L2 French vowel identification task was administered to 32 Australian English learners of French in audiovisual, audio‐only, and visual‐only conditions. Participants also completed tasks measuring their first language (L1) lipreading ability at the phoneme, word, and sentence level; a listening span task assessing their WM capacity; and a nonword repetition task assessing their PSTM. Overall, speech perception was significantly better in the audiovisual and audio‐only conditions, and large individual variability was observed in the lipreading, WM, and PSTM tasks. Results revealed a significant effect of PSTM on vowel perception scores in all modalities of presentation, but no effect of WM capacity was found. In addition, lipreading ability at the word level was found to be a predictor of accurate L2 vowel perception in the visual modality.

Language Experience and Subjective Word Familiarity on the Multimodal Perception of Non-native Vowels

“…One reviewer suggested that the lack of audiovisual (AV) benefit in this study might be due to the order of presentation (i.e., AV before auditory-only (A-only)). This possibility is, however, unlikely as the lack of AV benefit was also observed in two other studies (Inceoglu, 2016(Inceoglu, , 2019 where the presentation order of the AV and A-only conditions were counterbalanced across participants (see also, Pereira, 2013). 6.…”

“…Recent work has also started to shed light on how non-native speakers perceive nasal vowels (Detey & Racine, 2015;Inceoglu, 2016Inceoglu, , 2019Marquez Martinez, 2016). Inceoglu (2016) was the first study to examine the role of visual information in the acquisition of nasal vowels.…”

Metacognition in second language speech perception and production

“…Yet, even experienced L2 speakers might show a dissociation between the two domains, when the tasks used to assess L2 perception and production tap into di erent processing levels (e.g., Hao & de Jong, 2016;Melnik & Peperkamp, to appear;Peperkamp & Bouchon, 2011). For instance, while no relationship has been reported between L2 speech perception and production at the prelexical, acoustic-phonetic level, when using acoustic measures for L2 production accuracy (Hattori & Iverson, 2010;Schertz et al, 2015), a moderate-to-strong relationship was revealed at the phonological/lexical level, when more global perceptual (e.g., subjective) measures of perception and production accuracy (e.g., intelligibility ratings, see Evans & Alshangiti, 2018;Flege et al, 1999;Inceoglu, 2019) or tapping into similar levels of processing Flege, 1993;Flege et al, 1997Flege et al, , 1999Flege & Schmidt, 1995;Melnik & Peperkamp, to appear;Nagle, 2018) were used. In the current study, we carefully controlled for input/stimuli in exposure and during the tasks, aiming at assessing non-native perception and production at a similar prelexical, i.e., a phoneme, level; yet, the results revealed no association between the two domains.…”

“…Relatedly, previous research in L2 learning revealed no relationship (at the prelexical level) between L2 speech perception and production, when using acoustic measures of L2 production Schertz et al, 2015), as compared to when using subjective measures, e.g., intelligibility ratings, that assess L2 pronunciation beyond segmental accuracy, i.e., supra-segmental, coarticulation, etc. (Evans & Alshangiti, 2018;Flege et al, 1999;Inceoglu, 2019). Future research needs to examine ne-grained representations underlying L2 sound perception and production and their development and alignment with experience.…”

“…Although the dominant theoretical perspectives attribute L2 speakers' di culties in production to a lack of accurate perceptual representations for L2 sounds and predict a tight relationship between the two domains (Best, 1995;Best & Tyler, 2007;Flege, 1995), research provides no evidence for a consistent relationship between L2 speech sound perception and production (Bradlow et al, 1997;Flege, 1995;Flege & Eefting, 1987b;Hanulíková et al, 2012;Hattori & Iverson, 2010;Kartushina & Frauenfelder, 2014;Nagle, 2018;Okuno & Hardison, 2016;Peperkamp & Bouchon, 2011;Sheldon & Strange, 1982). The strength of the relationship is modulated by a number of factors, ranging from L2 experience and pro ciency Jia et al, 2006;Rallo Fabra & Romero, 2012), to the level of linguistic processing explored Hao & de Jong, 2016;Melnik & Peperkamp, to appear;Peperkamp & Bouchon, 2011), L2 sound di culty (Bohn & Flege, 1992;Evans & Alshangiti, 2018;Hao & de Jong, 2016;Levy, 2009;Levy & Law, 2010;Nagle, 2018) and L2 production accuracy measures (Evans & Alshangiti, 2018;Flege et al, 1999;Hattori & Iverson, 2010;Inceoglu, 2019). Therefore, in order to have a better understanding of the relationship between L2 perception and production, strict and systematic control of the tasks, measures to access L2 production, participants' experience and L2 sounds' di culty is required.…”