Does Musical Ability Matter?

This study examined the relation between musical ability and second-language (L2) proficiency in adult learners. L2 ability was assessed in four domains: receptive phonology, productive phonology, syntax, and lexical knowledge. Also assessed were various other factors that might explain individual differences in L2 ability, including age of L2 immersion, patterns of language use and exposure, and phonological short-term memory. Hierarchical regression analyses were conducted to determine if musical ability explained any unique variance in each domain of L2 ability after controlling for other relevant factors. Musical ability predicted L2 phonological ability (both receptive and productive) even when controlling for other factors, but did not explain unique variance in L2 syntax or lexical knowledge. These results suggest that musical skills may facilitate the acquisition of L2 sound structure and add to a growing body of evidence linking language and music. People exhibit substantial individual differences in secondlanguage (L2) proficiency. Learners’ age of immersion is known to influence their ultimate level of L2 ability, but even when this factor is taken into account, striking individual differences still exist, especially among people who started acquiring an L2 after childhood. Although some adult L2 learners attain near-native proficiency, others speak with strong foreign accents and frequent grammatical errors long after their immersion in the L2. Why do some adult learners acquire an L2 more successfully than others? What characteristics differentiate good L2 learners from not-so-good ones? One common answer to these questions (at least among laypersons) is that musical ability is an important determinant of such variation. According to this account, being skilled at music means having a ‘‘good ear’’ for analyzing and discriminating foreign speech sounds, so that musically talented individuals are better equipped than other people to pick up various aspects of an L2, especially pronunciations of L2 sounds. There are good scientific reasons to expect a link between musical ability and L2 proficiency. First, like language, music is a human universal consisting of perceptually discrete elements organized into hierarchically structured sequences (Patel, 2003; Sloboda, 1985). Second, neuropsychological evidence indicates that some brain regions often assumed to be language-specific (e.g., the inferior frontal gyrus, including Broca’s area) are also implicated in musical processing (Levitin & Menon, 2003; Maess, Koelsch, Gunter, & Friederici, 2001; Tillmann, Janata, & Bharucha, 2003), as are certain event-related potential signatures of language processing (Patel, Gibson, Ratner, Besson, & Holcomb, 1998). Third, musical ability can predict aspects of first-language (L1) verbal ability, such as reading ability in children (Anvari, Trainor, Woodside, & Levy, 2002; Atterbury, 1985). Given such evidence for a music-language link, it is surprising that little evidence is available regarding the hypothesized relation between musical ability and L2 proficiency. Skehan (1989) provided a detailed monograph-length review of individual differences in L2 acquisition, but had nothing to say about the relevance (or irrelevance) of musical ability to L2 proficiency. Moreover, several studies failed to find a clear link between self-ratings of musical ability and L2 ability (Flege, Munro, & MacKay, 1995; Flege, Yeni-Komshian, & Liu, 1999; Tahta, Wood, & Loewenthal, 1981; Thompson, 1991). Although two recent studies reported a positive link between musical ability and aspects of L2 pronunciation ability (Nakata, 2002; Tanaka & Nakamura, 2004), they did not control for effects of other potentially correlated variables. Thus, it is unclear whether musical ability makes a unique contribution to explaining variance in L2 proficiency. Address correspondence to L. Robert Slevc, UCSD Department of Psychology 0109, 9500 Gilman Dr., La Jolla, CA 92093-0109, e-mail: [email protected]. PSYCHOLOGICAL SCIENCE Volume 17—Number 8 675 Copyright r 2006 Association for Psychological Science Given this lack of evidence, it is tempting to conclude that the popular conjecture that musical ability matters for L2 learning is a myth. Drawing such a conclusion may be premature, however, because previous studies relied on subjective self-ratings, rather than objective, psychometrically validated measures, to assess musical ability. The aim of the current study was to test the musical-ability hypothesis more rigorously than has been done before. To this end, we tested native Japanese speakers who were not immersed in their L2 (English) until after the age of 11 (M 5 25.0 years, range: 11–47). No participant had achieved native-level L2 proficiency at the time of testing, but there was large variability in all four domains of L2 proficiency we tested (receptive phonology, productive phonology, syntax, and lexical knowledge). We focused on ‘‘late arrivals’’ because our goal was not to evaluate the intensively studied critical-period hypothesis, but rather to address a previously neglected question: What factors underlie individual differences in L2 proficiency among late learners? We assessed musical ability with several subtests from a wellknown standardized test (Wing, 1968) that has been shown to reliably predict teachers’ ratings of students’ musical ability and students’ grades in music (see Shuter-Dyson & Gabriel, 1981, for a review). We also measured other factors potentially relevant to L2 proficiency and used hierarchical regression analyses to examine whether musical ability could uniquely explain variance associated with the four L2 domains. Several of these additional measures have shown systematic relations to L2 proficiency: learners’ age of arrival (Hakuta, Bialystok, & Wiley, 2003; Johnson & Newport, 1989), length of residence in the L2 community (Flege & Liu, 2001; Flege et al., 1999), patterns of language use (Piske, MacKay, & Flege, 2001), and phonological short-term memory (STM) capacity (Baddeley, Gathercole, & Papagno, 1998; Ellis & Sinclair, 1996). If there is some truth to the popular musical-ability hypothesis, then musical ability should account for individual differences in L2 proficiency (especially pronunciation ability) even after controlling for the effects of other potentially relevant variables.