CORE Metadata, citation and similar papers at core.ac.uk Provided by Lancaster E-Prints Revisiting problems with foreign language aptitude* Anna Sáfár* and Judit Kormos*+ Abstract This study investigated three of the issues recently raised in connection with the traditional concept of foreign language aptitude: the relationship of foreign language aptitude and working memory and phonological short-term memory capacity, the role of foreign language aptitude in predicting success in the framework of focus-on-form foreign language instruction, and the stability of language aptitude and phonological short-term memory in the course of language learning. The participants of our research were 40 students of an English- Hungarian bilingual secondary school and 21 students in a regular Hungarian secondary school. Language aptitude was assessed both at the beginning and the end of the academic year. Our results support the existence of an effect of language learning experience on language aptitude. We also concluded that foreign language aptitude does not play a highly important role in communicative language teaching combined with focus-on-form instruction. 1. Introduction The traditional construct of language aptitude has come under a lot of criticism. The most important problems raised in connection with this cognitive determinant of language learning outcomes include issues of construct validity, in other words problems related to the fact that language aptitude lacks an appropriate definition supported by theories of language learning (Dörnyei, 2005; Sawyer & Ranta, 2001), as well as the relevance of the traditional 1 components of language aptitude included in the Modern Language Aptitude Test (MLAT) (Carroll & Sapon, 1959) in 21st century language classrooms, where language learning means more than just the acquisition of lexis and grammar (for a review see Robinson, 2005; Sawyer & Ranta, 2001; Skehan, 2002). Additional issues that emerged in connection with foreign language aptitude involve the relation of language aptitude to intelligence and working memory capacity, and the effect of language learning experience on aptitude (for a review of these issues see Sawyer & Ranta, 2001). Despite the voices of criticism, language aptitude tests such as the MLAT are still widely used to screen students for admission and streaming in language learning programs. Parallel to the growing concerns about the appropriacy of the traditional language aptitude test, new conceptualizations of language learning aptitude have been proposed, for instance Grigorenko, Sternberg and Ehrman’s (2000) CANAL-FT, which regards language aptitude as the ability to cope with novel experience in language learning, and Robinson’s (2001) aptitude complexes, which form a hierarchy of cognitive abilities relevant for second language acquisition (SLA). The aim of our research was to test some of the problematic aspects of the traditional concepts of language aptitude and some of the new proposals concerning the definition of language aptitude. We concentrate on three issues: the effect of language learning experience on language aptitude test performance, the predictive power of language aptitude in a one- year long intensive language training program, which primarily uses a communicative approach with focus-on-form instruction, and the relationship of working memory capacity and language aptitude. Our research uses a quasi-experimental design in assessing the effect of language learning experience on language aptitude as measured by the Hungarian version of MLAT (HUNLAT), and we applied a range of proficiency measures, which test more than just the acquisition of grammatical rules. The participants of our study were 15-16 year old 2 students of an English-Hungarian bilingual secondary school enrolled in a year of intensive foreign language training program, and a group of students with similar demographic characteristics from a regular secondary school, where languages are taught in four 45-minute lessons per week. We assessed phonological short-term memory capacity and language aptitude both at the beginning and the end of the academic year. Students’ verbal working memory capacity was measured on one occasion only at the end of the school-year. The language proficiency of the bilingual group was tested with a Cambridge First Certificate Exam at the end of the academic year. Our study used a correlational design to investigate the relationship of language aptitude, working memory and phonological short-term memory capacity and language learning success, whereas in order to test the effect of language learning experience on the performance on language aptitude and phonological short-term memory measures a quasi-experimental arrangement was used, with the bilingual group being the experimental and the students from the regular secondary school the control group. In what follows, we will outline previous research in the field of language aptitude and working memory, describe the research methods we used and the results obtained. We will conclude with a discussion and interpretation of the findings. 2. Review of literature Language aptitude is traditionally conceptualized as a stable ability that is not susceptible to training and is independent of previous language learning experience (Skehan, 1998). Carroll and Sapon (as most developers of language aptitude tests before and after them) followed the empirically-based psychometric approach to test development. Between 1953 and 1958 they administered a great variety of tests that seemed likely to predict 3 language learning success to about 5000 students. Based on the results, they selected the tasks which differentiated best between successful and unsuccessful language learners but did not correlate highly with each other (Carroll & Sapon, 1959). Such a pragmatic, assessment- based, and atheoretical approach is not without precedent in the research of cognitive abilities, in fact this is the approach most commonly adopted in the development of intelligence tests. Instruments devised in this manner may have appropriate psychometric qualities (by definition, in fact, since their development was based on measures of validity and reliability). Moreover, such instruments may be used to define the construct in question, by using complex statistical analyses to extract underlying factors. This was the approach followed by Carroll as well in defining components of language aptitude (see below). However, from a theoretical point of view, this process may be criticized for producing a construct which is, in fact, nothing more or less than what the test measures. This is a common critique of tests of intelligence, and according to Dörnyei (2005: 35) “the tacit understanding in the L2 research community has been that language aptitude is what language aptitude tests measure”. Carroll defined language aptitude as “some characteristic of an individual which controls, at a given point of time the rate of progress that he will make subsequently in learning a foreign language” (1974, quoted by Sawyer and Ranta, 2001: 310). Based on the results of factor analyses, Carroll (1981) identified four components of language aptitude: (1) Phonetic coding ability, which is “the ability to identify distinct sounds, to form association between those sounds and symbols representing them, and to retain these associations” (p. 105), (2) grammatical sensitivity, meaning the ability “to recognize the grammatical functions of words (or other linguistic entities) in sentence structures” (p. 105), (3) rote learning ability, which was defined as “the ability to learn associations between sounds and meanings rapidly and efficiently, and to retain these associations” (p. 105), and (4) inductive language learning, 4 that is, the ability “to infer or induce the rules governing a set of language materials, given sample language materials that permit such inferences” (p. 105). Correlation coefficients between various language aptitude tests and tests of foreign language proficiency range between 0.23 and 0.73 (Grigorenko et al., 2000). Ehrman and Oxford (1995) found that among individual differences it is language aptitude that correlates most closely with foreign language performance, explaining 25% of the variance. In a Hungarian context similar to that of our study, Nikolov and Ottó (2006) found that language aptitude was the best predictor of the achievement on proficiency tests at the end of the academic year in the case of beginner language learners (r=0.52 among students of English, r=0.39 among students of German). In the case of students who were not beginners when they started the preparatory language course, the best predictor proved to be language proficiency at the beginning of the training (r=0.78 and 0.74), and language aptitude was the second best predictor (r=0.44). One of the basic assumptions behind the concept of language aptitude is that it is a relatively stable characteristic, an endowment that is not modified by training or affected by previous experience (Skehan, 1998). However, not all researchers agree with this view. McLaughlin (1990:173) claims that “aptitude should not be viewed as a static personality trait; novices can become experts with experience.” Grigorenko et al. (2000: 401) assert that “language aptitude is a form of developing expertise rather than an entity fixed at birth.” Research evidence concerning the stability of language aptitude is ambiguous. In an early study Politzer and Weiss (1969, cited by Skehan, 1989) attempted to improve achievement on language aptitude measures by training, but they could not provide convincing evidence that this was possible. Sawyer (1992) did not observe correlations between previous language learning experience and aptitude among students taking part in short intensive language courses. Harley and Hart (1997) did not find higher levels of aptitude among early immersion 5 students having 12 years of exposure to a foreign language compared to late immersion students with only 4 years of exposure. Other studies, however, provide evidence for an effect of experience on language aptitude. A study conducted by Eisenstein (1980) showed that both bilinguals and students with previous training in foreign languages showed higher levels of aptitude than students without previous language learning experience. She also found that polylinguals tended to outperform bilinguals and that bilinguals who received formal education in a second language had an advantage over those who did not, although these differences were not statistically significant. In an experimental study Sparks, Ganshow, Fluharty and Little (1995) reported that instruction in Latin resulted in an increase in language aptitude scores in the case of both learning-disabled and non-learning-disabled high school students. It can be seen from the above that previous research does not provide conclusive evidence either for or against stability in language aptitude. Despite the lack of evidence, the assumption of stability is rarely called into question. In the words of Skehan (2002: 79), “the truth of this matter is that there is simply not enough evidence to argue for the stability of aptitude with any certainty, but for now, following Carroll, we will assume that aptitude does not change with the seasons!”. Robinson (2001, 2005) raised several concerns with the traditional concept of language aptitude, and he proposed that aptitude should be conceptualized as a complex and dynamic construct, comprising cognitive resources and primary abilities which combine into high- order abilities that are directly involved in various learning tasks. These abilities can in turn be grouped into aptitude complexes. Aptitude complexes play different roles under different conditions of exposure to input (incidental, implicit and explicit learning). The significance of Robinson’s research is that he investigates the aptitude-treatment interaction, conceiving of language aptitude as a dynamic construct. 6 The relevance of the Carollian concept of language aptitude in the communicative classroom was first questioned by Krashen (1981). Since then the role of aptitude in various language learning settings and under different learning conditions has been explored by a number of studies. Reves (1983, cited in Skehan, 2002), who investigated Arabic learners studying English in a classroom setting and acquiring Hebrew in the target language environment, showed that aptitude plays a role in predicting success in both situations. Skehan (1989) also argued that aptitude might in fact be more important in naturalistic SLA than in an instructional setting. Harley and Hart (1997) found that for young immersion learners, aptitude had little power in predicting competence in L2, but for older ones, for whom accuracy and academic language proficiency was a more focal point of instruction than for younger students, aptitude correlated highly with most proficiency measures. Horwitz (1987) and Ehrman and Oxford (1995) investigating older age groups, which were primarily taught with communicative methods, also found that certain sub-tests of language aptitude were related to L2 performance. On the other hand, Ranta’s (2002) research suggests that language analytic abilities play a relatively insignificant role in communicative classrooms. Likewise, Robinson (1997, 2002a) found that language aptitude did not influence incidental rule-learning under the meaning focussed condition. A number of researchers have argued that working memory capacity might be a cognitive ability that is just as important in language learning as foreign language aptitude. Sawyer and Ranta (2001: 340) pointed out that “working memory capacity may be the key to elaborating the concept of language aptitude itself and to clarifying its relationship with the second language acquisition (SLA) process.” Miyake and Friedman (1998) proposed the “working memory as language aptitude” hypothesis, claiming that working memory may be the central component of language aptitude. To our knowledge, it is only Robinson (2002a), who has tested the relation of working memory and foreign language aptitude. He found that 7 working memory as measured with a reading span test had a moderately strong correlation with language aptitude scores. The most widely accepted conceptualization of short-term memory today is the working memory model developed by Baddeley and Hitch (1974; Baddeley, 1986). While previous theories of memory systems focused on the storage function of memory, the new model, as its name suggests, adapts a more dynamic approach. This conceptualization of working memory combines storage with the processing and manipulation of information, thus in this view working memory plays a far greater role in cognitive activities such as comprehension, reasoning and learning than previously assumed (Baddeley, 2003). The most widely researched component of working memory is the phonological loop. This subsystem consists of a phonological store, which holds information for a few seconds, and an articulatory rehearsal process, which refreshes decaying information amongst other functions. The rehearsal process is analogous to subvocal speech and takes place in real-time, resulting in a limited span of immediate memory (after a certain number of items, the first one will fade before it can be rehearsed). Phonological loop capacity is often measured by tasks involving immediate serial recall of numbers (digit span) or words (Baddeley, 2003). One of the most widely used tests of phonological short-term memory capacity is the non-word repetition test, where participants have to repeat non-words of various length. Non-words are words that do not exist in the given language but conform to its phonotactic rules. Participants’ short-term memory capacity may then be expressed in terms of the non-word span, which is the highest number of syllables the participant could repeat in at least 50% of the cases. Other tests of phonological short-term memory include alphabet span (Craik, 1986) and forward digit span (Botwinick & Storandt, 1974). Reading and listening span (Daneman & Carpenter, 1980) as well as the backward digit span task are considered complex verbal memory tasks. The backward digit span task is also 8 part of the Wechsler-IV intelligence test for children (Gathercole & Alloway, in press) and is hypothesized to be strongly related to general fluid intelligence (see Engle, Kane & Tuholsky, 1999). The backward digit span and the reading and listening span tasks are frequently regarded as instruments testing more than just the phonological short-term memory: they are claimed to assess the capacity of complex verbal working memory including the functioning of the central-executive, which is responsible for regulating attention (Gathercole, 1999; Hale et al., 2002). We have to note, however, that there is also evidence, mainly from studies using factor analysis, that tests measuring phonological short-term memory capacity and instruments assessing complex verbal working memory capacity are in fact diagnostic tools that tap the same underlying construct (for a recent study on this issue see Colom et al., 2006). Complex working memory has been found to influence mathematical abilities (Logie et al., 1994, cited by Gathercole, 1999) as well as intellectual abilities such as following directions, note-taking, writing and reasoning (Engle et al., 1999). In a recent paper Gathercole and Alloway (in press) argue that working memory “acts a bottleneck for learning” (p.12). Language learning is in this respect very similar to the acquisition of literacy and arithmetic skills and also requires that children maintain information in working memory while engaging in various cognitive activities. Phonological short-term memory was found to be related to the ability to acquire new words in L2 (Cheung, 1996; Papagno & Vallar, 1995; Service, 1992; Service & Kohonen, 1995; Speciale et al., 2004). Ellis and Sinclair (1996) presented experimental evidence that rehearsing foreign language material has beneficial effects on both comprehending and learning foreign language words, metalinguistic knowledge of grammar, accuracy in pronunciation and productive grammatical fluency and accuracy. O’Brien et al. (2006) showed that there was a link between phonological memory and oral production skills and that the nature of relationship between measures of 9 phonological short-term memory and various assessments of oral performance was different between proficient and less-proficient L2 speakers. Research on complex working memory and language skills is less abundant than studies investigating the role of phonological short-term memory. Most of these studies use the reading span task, the validity of which as a measure of verbal working memory capacity is questionable (Waters & Caplan, 1996; Caplan & Waters, 1999). To our knowledge, no studies in the L2 field have used the backward digit span despite the fact that considerably fewer validity issues have been raised concerning this test than the reading span test (Hale et al., 2002). Harrington and Sawyer (1992) found that L2 reading span scores showed high correlations with performance on the grammar (r=0.57, p<0.01) and the reading and vocabulary (r=0.54, p<0.01) sections of the TOEFL exam. Miyake and Friedman (1998) obtained evidence for the causal role of working memory in certain aspects of second language proficiency (e.g. syntactic comprehension) using path analysis. In our study we investigate three of the issues recently raised in connection with the traditional concept of foreign language aptitude: the relationship of foreign language aptitude and working memory capacity, the role of foreign language aptitude in predicting success in the framework of communicative language teaching combined with focus-on-form instruction and the stability of language aptitude in the course of language learning. As can be seen from the review of literature, all of these issues have been investigated in previous research, but never in the framework of a single study. Moreover, no studies examining the role of aptitude in communicative language teaching have used a comprehensive language testing instrument that measures attainment in productive and receptive skills as well as linguistic competence. Finally our study is novel in applying a quasi-experimental design to test the effect of language learning experience on language aptitude. We would like to point out here that in 10