Measured Language: Quantitative Studies of Acquisition, Assessment, and Variation

10. The Differential Role of Language Analytic Ability in Two Distinct Learning Conditions

141 10 ■The Differential Role of Language Analytic Ability in Two Distinct Learning Conditions NADiA MifkA PRofoziC University of Zadar, Croatia ■ THIS PAPER EXAMINES WHETHER the role of language analytic ability in L2 acquisition may change relative to different types of corrective feedback provided in a communicative language classroom. Two groups of high school students in New Zealand (n = 18 in each group) that were learning French as a foreign language were subjected to different learning conditions while working on three communicative tasks over two weeks. In one group, learners were asked to self-correct following the teacher’s request for clarification whenever they committed an error in the production of the target structures; the other group received recasts as a type of input-based corrective feedback. The target structures were the French passé composé and imparfait, which are past-tense structures. Along with language tests, the participants completed a test designed to measure language analytic ability (Ottó 2004). Correlations between gain scores of accurate production and analytic ability were different for the two groups. Analytic ability was significantly correlated with oral and written production gains for the group that received clarification requests, whereas the learners who received recasts had to rely on their analytic ability only to retain long-term gains of passé composé in oral production. Lack of correlation with gains for the recasts group suggests that recasts assisted all learners irrespective of their ability, whereas strong correlation of analytic ability with gains in the clarification requests group means that these students learned no more than their analytic ability would predict. Language analytic ability has been identified as a component of language aptitude , which refers to a person’s “capacity to analyze or process language input” (Skehan 1986, 202). Analytic ability correlates with general intelligence (Sasaki 1996), particularly with those aspects of intelligence known as fluid intelligence (Ehrman and Oxford 1995; Sternberg 2002). Analytic ability is essential for making inferences and generalizations, and thus contributes to all aspects of abstract, formal, logical, or mathematical thinking. In the most complex and widely used aptitude battery to date, the Modern Languages Aptitude Test (Carroll and Sapon 1959), language analytic ability is partially measured by the grammatical sensitivity subtest (the words in sentences section of the MLAT). As a counterpart to the top-down processing associated with grammatical sensitivity, the bottom-up types of analyses 142 Nadia Mifka Profozic assisted by inductive language learning ability also contribute to language processing . Hence, inductive learning ability is also considered to be part of language aptitude (Carroll 1962), although it is not measured by the MLAT. In Skehan’s (1998) framework of his information processing model, grammatical sensitivity and inductive language learning ability together constitute language analytic ability, which may facilitate analyses performed during the processing phase of language learning. Language Aptitude As repeatedly emphasized in literature on individual differences in language learning (Ellis 2004; Robinson 2001; Sawyer and Ranta 2001; Skehan 1989; Sparks and Ganschow 2001; Sternberg 2002), analytic ability is only one component of language aptitude; the other parts are related to memory and phonemic coding ability . It has been suggested that different components of aptitude may be linked to specific areas of language learning. For example, Nagata, Aline, and Ellis (1999) examined how different aspects of aptitude influenced different areas of L2 learning, and showed that analytic ability as measured by the MLAT grammatical sensitivity subtest was closely related to comprehension but not to vocabulary learning. Harley and Hart (1997) found stronger correlations with memory measures for early starters in L2 immersion settings, whereas for late starters, there was a significant correlation with analytic ability. Strong support for the claim that the role of analytic ability is clearly related to the age of L2 onset has been provided by DeKeyser (2000), who was able to show that analytic ability as measured by the MLAT grammatical sensitivity subtest correlated significantly with scores on a grammaticality judgment test (GJT) for those L2 speakers who arrived in the United States after the age of seventeen. The GJT scores of young arrivals—those who arrived in the United States before they reached the age of puberty—did not correlate with their analytic ability, thus providing further evidence for the critical period hypothesis (Johnson and Newport 1989). However, adults are different from young learners who still have access to their implicit learning mechanisms; when faced with the challenge of learning an L2, adult learners have to rely on their general cognitive resources, and, therefore, their analytic...