Abstract
Keywords
Introduction
During the last 50 years, there has been a significant shift in the goal of L2 pronunciation teaching from sounding native-like to becoming intelligible for the purpose of successful communication (Levis, 2005). Encouraged by the empirical evidence that adult L2 learners are often foreign-accented (e.g. Abrahamsson and Hyltenstam, 2009) and that being an effective communicator does not necessarily require fully native-like pronunciation (Munro and Derwing, 1995), the paradigm of intelligibility spurred research to identify the key constructs of intelligible pronunciation (Trofimovich and Isaacs, 2012) and learners’ individual differences (IDs) that characterise the course of L2 pronunciation acquisition (see Pennington and Rogerson-Revell, 2019, pp.74 ff for a comprehensible overview). According to Dörnyei’s (2009) socio-cognitive view, learners’ IDs are derived from three domains of human mental functions: cognition, motivation and affect (e.g. Matthews and Zeidner, 2004). Following this three-domain orientation, the present review provides a selective overview of the past research that has explored three types of learner IDs (foreign language learning aptitude and musical aptitude in the domain of
The Process of L2 Pronunciation Learning
Second Language Acquisition (SLA) scholars tend to agree that the pattern for first language (L1) speech acquisition (e.g. the fact that L1 perceptual development starts prior to L1 production development) can also be applied to that of L2 speech learning (e.g. Eckman, 2008; Flege, 1995; Hansen, 2004). Empirical evidence shows a relatively consistent link between perception and production in L2 segmental development (see Saito and Van Poeteren, 2018 for a review). What makes
In foreign language (FL) classrooms, compared to an immersion setting, the amount and quality of the available L2 experience appear to be relatively limited (Larson-Hall, 2008). Therefore, how much the learner can develop their pronunciation largely depends on the type of instruction (e.g. Norris and Ortega, 2000), the amount of recent classroom instruction (Saito and Hanzawa, 2016), and the amount of extra-curricular L2 learning (Muñoz, 2014). However, apart from the influence of experience, the research has revealed considerable differences in the amount and quality of L2 pronunciation development (e.g. Saito and Hanzawa, 2018). While social and contextual factors shape the learning environment (e.g. Toth and Moranski, 2018), SLA researchers attribute the differences in the rates and outcomes of learning to the impact of various learner factors (e.g. Baker-Smemoe and Haslam, 2013; Gonet, 2006; Muñoz, 2014; Saito and Hanzawa, 2016). According to Dörnyei (2009), learners’ IDs are the subcomponents of learners’ minds – their
Effect of Learners’ Individual Differences on L2 Pronunciation Learning
Cognitive Individual Differences
Foreign Language Learning Aptitude
One of the most researched cognitive ID variables in the field of SLA is aptitude. Aptitude has been conceptualised as the following set of abilities that enhance foreign language learning: phonemic coding ability (noticing and analysing unfamiliar auditory information), grammatical sensitivity, inductive learning ability, and memory (Ortega, 2013). Since the inception of Carroll and Sapon’s (1959) Modern Language Aptitude Test, various versions of aptitude batteries have been developed (see Skehan, 2016, for review) and utilized in SLA research (see Li, 2015 for a meta-analysis).
While existing research has found that L2 grammar, vocabulary, and collocation learning could significantly benefit from greater aptitude in memory, researchers also extended their interest in the aptitude effect to L2 pronunciation learning (Baker-Smemoe and Haslam, 2013; Saito, 2017; Saito and Hanzawa, 2016; Saito, Suzukida et al., 2019). With data from Japanese learners of English, Saito and Hanzawa (2016) reported that overall LLAMA aptitude test 1 scores (Meara, 2005) showed positive correlations with better segmental, word stress, and speech rate performance as evaluated by native judges. Baker-Smemoe and Haslam (2013) found that accurate pronunciation, lower foreign accentedness, and better fluency of L2 learners in English as a second language (ESL) and English as a foreign language (EFL) settings were best predicted by sound discrimination ability but that was not the case with higher comprehensibility. Saito (2017) also demonstrated a strong impact of aptitude on various dimensions of pronunciation. Drawn from spontaneous speech samples elicited from Japanese learners of English, the results illustrated the multifaceted role of aptitude in L2 speech development. The learners’ phonemic coding ability was associated with pronunciation and grammatical accuracy; their rote and associative memory contributed to articulation rate and grammatical complexity; and their grammatical inferencing ability was correlated with vocabulary richness. However, sound sequence recognition, which was assumed to measure learners’ implicit learning ability, did not correlate with any linguistic variables measured in the study, indicating that while implicit learning aptitude is relevant for unintentional, incidental learning, it may not be influential in the context of intentional, explicit learning – a characteristic of learning in FL classroom settings. These findings have been re-examined and extended from a longitudinal perspective by Saito, Suzukida et al. (2019), who reported that in the first semester of one academic year, explicit learning aptitude (rote memory and phonemic coding ability) appeared to enhance learners’ global comprehensibility through the improvement of fluency and prosodic aspects of L2 speech. In the latter stage of the academic year, the learners with higher implicit learning ability (sound sequence recognition) achieved higher comprehensibility scores due to the impact of incidental learning aptitude on the refinement of segmental accuracy. Therefore, the existing studies suggest that while greater aptitude appears to facilitate L2 learners’ phonological development, different types of aptitude enhance different aspects of pronunciation.
Musical Aptitude
In relation to foreign language learning aptitude, musical aptitude (superiority in perception of various aspects of music) has been considered to be a crucial factor that affects L2 speech development (e.g. DeKeyser, 2013; Li and DeKeyser, 2017; Moyer, 2014; Piske et al., 2001). Musical aptitude is broadly conceptualised as the potential to learn or achieve in music and is measured through one’s sensitivity to various aspect of sounds such as pitch, tone, and rhythm (e.g. Gordon, 1965; Seashore, 1939; Wing, 1970).
Due to the similarities between the processing of music and language, researchers have investigated whether language learning can benefit from superior sound sensitivity (see Nardo and Reiterer, 2009 for an exhaustive research summary; also see Kraus and Chandrasekaran, 2010 for a review of neuroscientific evidence). Concerning L2 pronunciation acquisition research, Kempe et al. (2015), for instance, have demonstrated the mediating effects of tonal and pitch sensitivity to the better processing of non-native sounds (vowel sound discrimination). Scholars have found that the learners’ sensitivity to tones, pitch, and rhythm appear to predict better L2 pronunciation learning in an ESL context (e.g. Kempe et al., 2015; Richter, 2018; Slevc and Myiake, 2006) and in an FL classroom context (e.g. Milovanov et al., 2010). Li and DeKeyser (2017) found that native speakers of American English with high sensitivity to pitch could perform well in perception and production tasks after learning how to differentiate Mandarin words with contrasting tone patterns. Saito, Sun et al. (2019) examined the impact of musical aptitude on the speech development of Chinese learners of English in the UK, demonstrating that rhythm sensitivity significantly predicted the learners’ improvement in perceived speech rate. Working with adult learners of English in an EFL context, Milovanov et al. (2010) demonstrated that general musical aptitude was strongly and negatively correlated with segmental errors made in a word reproduction task. Furthermore, recent research in domain-general auditory processing suggests that learners’ basic ability to process basic auditory information (e.g., frequency, intensity, and duration) is linked to successful L2 learning (e.g., Kachlicka et al., 2019; Saito et al., 2020; Sun et al., 2020; Zheng et al., 2020). Although further research is needed to determine how musical aptitude and domain-general auditory processing influence acquisition of various aspects of L2 speech (i.e. segmental, prosodic, and temporal), the evidence reviewed here indicates that these factors may interact with L2 linguistic ability and enhance L2 pronunciation development.
Socio-Psychological Individual Differences
Motivation
Motivation research has revealed a positive correlation between L2 learning motivation and L2 learning success (e.g. Schmidt and Watanabe, 2001). Regarding L2 pronunciation, research revealed that L2 motivation could be an indicator of reduced foreign accent (Elliott, 1995; Flege et al.,1995; Moyer, 1999, 2004; Purcell and Suter, 1980) and comprehensibility (Saito et al., 2017). Moyer (1999) reported that the pronunciation accuracy of native English learners of German in Germany was rated higher for learners who had higher motivation to pursue a native-like quality of pronunciation. Focusing on learners’ L2 integrativeness, instrumental motivation, and metacognition about L2 learning, Saito et al. (2017) examined how such factors affect the development of L2 speech comprehensibility and accentedness over one year. Among the motivational factors, learners who were highly motivated to study English for their future career development as a vague and long-term goal, and who had a strong desire to improve their comprehensibility, received higher comprehensibility ratings than those who were not motivated to study English and those who did not have desire to produce more comprehensible L2. In contrast, the learners’ accentedness was not associated with any motivational factors.
Another line of research has investigated the link between L2 pronunciation development and the L2 Motivational Self System. The L2 Motivational Self System (Dörnyei, 2009) comprises three components that are measured by a questionnaire: the ideal L2 self (i.e. motivation to learn the target language stemming from one’s aspirations), the ought-to L2 self (i.e. a belief that one ought to learn the target language to avoid possible negative outcomes), and L2 learning experience (Dörnyei and Csizér, 2002). Although the number of L2 pronunciation studies that have incorporated the Self System is limited, the amount of evidence supporting it has been gradually increasing (Nagle, 2018; Saito et al., 2018). Saito et al. (2018) explored the links among L2 speech comprehensibility, motivation (ideal L2 self, ought-to L2 self), and emotion (L2 learners’ enjoyment of learning foreign language in an FL classroom, foreign language classroom anxiety) in a context involving Japanese high school learners of English. While a cross-sectional investigation revealed that higher comprehensibility was associated with stronger aspiration for becoming their ideal self as an L2 user (i.e. ideal L2 self), greater FL learning enjoyment, past learning experience in preschool (i.e. amount of hours the participants engaged in learning English before they started elementary school), the amount of L2 use during the regular English classes at their high school, and lower anxiety, a regression analysis showed that comprehensibility was impacted mainly by anxiety and strength of the aspiration for becoming an ideal L2 user. In a three-month longitudinal study, the researchers found that comprehensibility gains were correlated with the degree of the learners’ aspiration for becoming an ideal L2 user, how positive students felt about learning English in the classroom, lower anxiety, and current L2 experience such as L2 learning at a cram school (e.g. a private-tutoring school where tutors help students prepare for university entrance exams) and conversations with native and non-native speakers. Thus, evidence suggests that L2 learners’ strong desire to narrow the gap between their perceived actual selves and ideal selves may be a predictor of successful L2 pronunciation learning.
Anxiety
Anxiety has been recognised since the 1970s as a factor that negatively affects the process of L2 learning, especially in the classroom, (e.g. Horwitz et al., 1986; Scovel, 1978). This type of anxiety is further divided into three components: (a) communication apprehension, (b) fear of negative evaluation, and (c) test anxiety. Anxiety is assumed to manifest in learners as reluctance to use complex structures of the foreign language, a lack of confidence in role-play activities, and being unable to remember learnt language items when taking examinations (Liu, 2006). In addition, it appears to be triggered by particular feedback from the teachers and lesson styles where L2 learners have to stand up and speak the target language in front of peers (see MacIntyre, 2017). Accordingly, a body of research has revealed that anxiety impedes various aspects of L2 achievement (see Teimouri et al., 2019 for an extensive review).
Regarding L2 pronunciation learning, since learners’ perception of their own foreign accent tends to cause embarrassment in the classroom setting, it is recognised as the most anxiety-provoking aspect of spoken language performance (Baran-Łucarz, 2013; Price, 1991; Saito et al., 2018). Relatively recently, Baran-Łucarz (2016) has conceptualised pronunciation-specific anxiety, which consists of four main components: fear of negative evaluation, pronunciation self-efficacy and self-assessment, pronunciation self-image, and the learners’ set of beliefs related to pronunciation. According to Baran-Łucarz (2016), fear of negative evaluation involves worry and apprehension caused by negative responses or assessments from interlocutors/listeners (e.g. classmates, teachers, and native/non-native speakers of target languages). Pronunciation self-efficacy and self-assessment measures learners’ self-perception and comparative assessments of their own prospects of learning accurate pronunciation of the target language versus those of other learners around them. Pronunciation self-image deals with learners’ self-image of how they look and sound when they pronounce the target language. The learner’s set of beliefs related to pronunciation measures a learner’s attitude and perception of the importance of learning the correct pronunciation of the target language. While pronunciation-specific anxiety has not been examined as a predictor of L2 pronunciation learning success, the constructs offer insight into the development of less anxiety-provoking L2 pronunciation instruction (Baran-Łucarz, 2013).
Pedagogical Implications
As shown, different types of IDs (cognitive vs. socio-psychological) differentially influence various aspects of L2 pronunciation development. In order to make the most of the learners’ varied ID profiles and engender successful L2 pronunciation acquisition, there are several approaches which practitioners can incorporate into their regular language classrooms. First, research suggests that learners with beneficial cognitive ID factors (language learning aptitude and musical aptitude) are likely to be able to efficiently pick up on the L2 phonetic and prosodic features in the available linguistic resources due to their sensitivity to L2 sounds. In order to facilitate L2 pronunciation learning for those learners with relatively lower degrees of such aptitude as well as those with higher degrees of aptitude, teachers can help them increase the L2 experience qualitatively and quantitatively and introduce pronunciation-focused lessons in syllabi. Second, L2 learners’ motivational and emotional states have been found to influence both the amount and quality of their experience and the quality of L2 pronunciation development. In order to help learners enhance their motivation and induce positive emotional states, teachers can provide pronunciation learning activities and use different types of instruction and feedback during language lessons. In the following sections, I detail the approaches suggested by SLA studies.
Facilitating Cognitive Processing
As indicated by theoretical and empirical evidence, reinforcing L2 learners’ perception of L2 sounds will likely contribute to better L2 speech production. For instance, Bradlow et al. (1997) have demonstrated that the perceptual training of /r/-/l/ with Japanese learners of English improves their segmental production accuracy (see Sakai and Moorman, 2018 for an extensive review of perceptual and production training). Providing the training and feedback on specific pronunciation features may be beneficial as it can help learners direct their awareness towards key aspects of pronunciation and notice less accurate aspects of their performance (see, e.g. studies by Couper, 2011, on epenthesis; Lee and Lyster, 2016, on English vowels; Saito and Lyster, 2012, on English /r/; Sardegna, 2011, on linking). Other studies have examined the effectiveness of both segmental and suprasegmental instruction for reducing pronunciation errors (Couper, 2003) and improving global intelligibility, comprehensibility, and accentedness (Derwing et al., 1998; Derwing et al., 2014). For instance, Derwing et al. (1998) provided 20 minutes of production instruction to ESL learners for 10 weeks: one group received segmentally focused instruction, while another group received suprasegmentally focused instruction (with a control group who did not receive any pronunciation-focused instruction). Performance measured via a sentence reading task revealed that both experimental groups demonstrated improvement in accentedness and comprehensibility. By contrast, only those participants who received suprasegmental instruction showed progress in their extemporaneous performance, which was measured using a picture description task. While the study indicated that the accuracy of suprasegmental features may improve more quickly than for segmentals, it proved that 20 minutes per day of instruction of either type can make a significant difference in the learners’ pronunciation performance.
In addition to the explicit learning and deliberate practising of segmental and suprasegmental features, production activities and peer interactions appear to be effective for developing learners’ comprehensibility (e.g. Saito and Akiyama, 2017; Trofimovich et al., 2009). A study by Saito and Akiyama (2017) explored the effect of different types of instruction for Japanese learners of English over an academic semester. The control group received instruction focusing on learning grammar and lexical aspects of English, while the experimental group participated in online interaction activities with native speaker interlocutors in the US. The activities were completed in pairs of Japanese L2 learners and native speakers. The native speakers were instructed to give corrective feedback in the form of recasts whenever they felt the learners’ speech was difficult to understand. The results revealed a significant improvement in the learners’ comprehensibility. In EFL classroom settings, production and interaction opportunities effectively increase L2 speech comprehensibility. In summary, it has been suggested that teachers incorporate various pronunciation-focused types of instruction and activities systematically to help L2 learners train their perception and production as well as attend to the pronunciation features that are most crucial for communicative success.
Facilitating Positive Motivation and Emotion
The studies of motivation and its link to SLA have revealed that strongly internalised motivation (i.e. a strong aspiration to reduce a self-assessed discrepancy between a learner’s current self as an L2 user and the learner’s ideal self) plays a pivotal role in increasing L2 learners’ opportunities to use the L2 and in helping them attain a higher level of pronunciation. However, goals and motivational orientations may differ considerably among learners, and they often exhibit varied strength of aspiration to achieve their ideal self as L2 users. Particularly in the context of EFL classrooms, many students tend to see studying English as an obligation and appear to have prevention-oriented motivation (i.e. they have a strong orientation to study English to avoid negative consequences such as obtaining low grades and failing to meet their parents’ expectation or curriculum requirements) (Li, 2014). Thus, different motivation-stimulating activities such as ones that have a specific goal and that have multiple subtasks in which the learners can see their tangible progress can be introduced into language classrooms (Dörnyei and Kubanyiova, 2014).
Other research (Guilloteaux and Dörnyei, 2008; Moskovsky et al., 2013) has revealed that teachers’ behaviour may influence their L2 students’ motivational orientations. Moskovsky et al. (2013) demonstrated that teachers’ use of motivational strategies (e.g. showing respect and warmth to the learners, knowing the learners well, sharing personal experience and interest with them) in the classroom led to a positive change in the learners’ L2 motivation profiles.
Concerning anxiety, in order to alleviate the fear of speaking in the L2 and to avoid possible fear-inducing situations, teachers may aim to foster learning enjoyment and to create an atmosphere in which teachers and peers all have a respectful, understanding, and tolerant attitude towards making mistakes. This can be achieved by creating a friendly and pleasant classroom environment in which learners are advised to use the target language frequently and freely (Dewaele et al., 2018). Anxiety may be also mitigated by providing learners with a number of communicative activities in low-risk situations in which they complete a task in pairs rather than in a potentially more anxiety-inducing setting involving large groups of peers and observation by teachers (e.g. Baran-Łucarz, 2014).
Lastly, L2 learners’ self-confidence can be compromised under certain conditions. This may happen, for example, when communication breakdowns with peers and native speakers are felt to be the result of flaws in their use of language, or when they receive explicit corrective feedback from the teachers in front of their peers in a classroom (e.g. Teimouri, 2017). Therefore, teachers may want to consider using different types of feedback (e.g. positive vs. negative feedback) depending on the students’ needs and attitudes, when reacting to their pronunciation errors. Furthermore, researchers have suggested that teachers’ support for learners’ effective self-training opportunities could help improve the learners’ self-confidence and reduce the degree of anxiety they feel while using the target language. This can be done by encouraging them to practise pronunciation by themselves and by providing the learners with various pronunciation learning strategies which they can use outside the classroom (e.g. Sardegna, 2011).
Future Directions
Following Dörnyei’s three-domain orientation of learner IDs (i.e. cognition, motivation, and emotion), the current review focused on empirical studies concerning the roles of aptitude, motivation, and anxiety. By introducing the theoretical framework of
Although it is difficult to tailor instruction to fit each learner’s ID profile and pronunciation proficiency level, various approaches which help learners make the most of their IDs for effective pronunciation learning can be introduced. In addition, it is crucial for teachers to frequently draw learners’ attention to pronunciation features not only in speaking-focused classes but also in listening-focused classes. In doing so, teachers can help learners have regular opportunities to enhance their perception and production of L2 segmental and suprasegmental features. In summary, I recommend that teachers take the following actions:
increase learners’ amount of L2 use;
increase the L2 input learners receive;
include perception-focused activities;
incorporate explicit instruction on segmental and suprasegmental features of pronunciation;
provide less anxiety-provoking activities (e.g. peer interactions) and motivation-stimulating activities that enable learners to practise the pronunciation features;
show a positive and tolerant attitude towards learners, especially when they make mistakes; and
share the personal experience of learning an L2 with your students.
