by Rod Ellis
A feature of these models is the importance they attach to working memory. Working memory is generally conceived as a multi-component memory system. Baddeley (2003) distinguished four sub-components: (1) the central executive that controls attention; (2) the visuospatial sketchpad that stores and rehearses visual information; (3) the phonological loop that stores and rehearses oral information; and (4) the episodic buffer that combines information from different sources. The two components that have received the greatest attention are the phonological loop and the central executive. The phonological loop holds phonological information for short periods of time and is also able to silently rehearse the information to reactive fading memory traces. It constitutes a limited-capacity component. The central executive is a supervisory system that controls complex operations such as focusing attention and regulating the flow of information between short-term and long-term memory. It is also seen as a limited-capacity system.
It is not difficult to see why learners’ working memory capacity is so important for language learning. Learners with a larger capacity will be able to store more linguistic data, rehearse it more fully, and make links with information stored in long-term memory. Working memory is hypothesized to be especially important in implicit learning when learners are primarily focused on meaning. J. Williams (2005), for example, provided evidence to show that differences in phonological short-term memory (i.e. the ability to store auditory traces) predicted differences in learners’ ability to learn certain grammatical features such as gender agreement in Spanish implicitly.
However, the results of working memory research in SLA to date have not always produced easily interpretable results. As Ortega (2009) noted, research has only begun to ‘scratch the surface’ of the relationship between working memory and L2 learning. Nevertheless, there are strong theoretical reasons, supported by some research findings, to indicate that working memory is an important component of language aptitude. In particular, short-term phonological memory appears to play a significant role. Working memory is considered further in Chapter 7 and Chapter 8.
Summing up
Early work in language aptitude centred around the development of tests—such as Carroll and Sapon’s (1959) Modern language Aptitude Test (MLAT)—that were used to predict how easily learners would learn a second language.
The MLAT is a robust and useful instrument and continues to be used in research today. It has been shown to predict success in learning in both naturalistic and instructed contexts.
As language aptitude is comprised of a number of distinct abilities, it is possible that learners differ in the abilities they are strong in. Skehan (1986) proposed a distinction between analytic-oriented and memory-oriented learners, both of whom can achieve success.
Language aptitude was initially seen as a stable, trait-like construct, but this view was subsequently challenged. There is evidence to suggest that abilities change as a result of learning experience and therefore may be trainable.
There is a relationship between language aptitude and age. A high level of language analytical ability may be required to enable adult learners to achieve high levels of L2 proficiency (DeKeyser 2000). This ability, however, appears to be of less importance for child learners.
The abilities required for implicit and explicit learning may also differ. For example, phonological coding ability is more important for implicit learning while language analytical ability is more important for explicit learning.
The original conceptualization of language aptitude has changed. New models (Skehan 2002; Robinson 2002) have been developed that link specific abilities to stages in the process of L2 acquisition and to the requirements of different instructional tasks. Central to these new models is working memory, which is now seen as a key component of language aptitude.
Motivation
Like language aptitude, motivation is a complex construct. It involves:
The reasons a learner has for needing or wanting to learn an L2 (i.e. motivational orientation).
The effort a learner makes to learn the L2, the learner’s persistence with the learning task, and the impact immediate context has on these (i.e. behavioural motivation).
The effect that the learner’s evaluation of his/her progress has on subsequent learning behaviour (i.e. attributional motivation).
Like language aptitude, motivation is primarily seen as affecting the rate of learning and ultimate achievement. However, there have been only limited attempts to link the study of motivation to mainstream work in SLA by demonstrating how it is related to the processes responsible for acquisition. But see Tseng and Schmitt (2008) for an attempt to do this for vocabulary learning.
Following Dörnyei (2005), I will adopt a historical approach by outlining how the study of motivation has evolved over the last 50 or so years. I will then focus on a recent theory of motivation and the research it has generated.
The social-psychological period (1959–1990)
This period was dominated by the work of Canadian social psychologists, who were interested in the role that motivation played in language learning in a society that was divided into Anglophone and Francophone communities. The starting point was the recognition that learners’ motivation depended on their attitudes towards the other community and to the target language, and that these were socially determined.
Gardner and Lambert (1972) distinguished two broad orientations: an integrative orientation entails a desire to identify with the target-language culture and its speakers; an instrumental orientation arises when learners wish to learn a second language for functional purposes (for example, to pass an examination or obtain a job). However, initially, instrumental motivation was not fully theorized as they saw the integrative orientation as central in the Canadian context. Later, however, Gardner (2001) came to recognize that ‘there can be other supports for motivation not directly associated with integrative motivation’ (p. 7).
Motivation, however, comprises more than the learner’s orientation. It is also influenced by the learner’s attitudes towards the learning situation (for example, attitudes towards the teacher and the instruction) and the actual effort that the learner puts into learning a second language. ‘Motivation’, then, is a composite construct involving orientation, attitudes, and effort. Gardner’s (1985) Socio-educational Model also emphasized the importance of the social and cultural milieu in which learning took place. This determined the cultural beliefs learners held, which in turn influenced their orientation and attitudes to the learning situation. Gardner also acknowledged the role played by language aptitude in determining learning outcomes, but saw it as only relevant in formal learning contexts.
The theory was enormously influential. Dörnyei (2005) commented that it was the dominant theory for three decades. It generated a large number of correlational studies. These made use of the instrument that Gardner developed to measure motivation—the Attitude Motivation Test Battery (AMTB) (1985), which included questions relating to motivational orientation, attitudes, and effort. The main findings of this research can be summarized as follows:
Integrative motivation is positively correlated with various measures of L2 achievement. Gardner (2005) reported a median correlation of 0.37 in a survey of studies of learners of L2 French conducted in seven different geographical areas in Canada.
Learners’ integrative motivation was also found to be related to the teacher’s and students’ classroom behaviours (for example, students’ voluntary responses to teacher questions).
Learners with an integrative motivation were less likely to abandon learning a second language (Ramage 1990).
In some contexts, however, integrative motivation was found to be negatively associated with achievement and that other motivations could be important: for example, Oller, Baca, and Vigil (1977) reported that Hispanic learners of English in California were more motivated by a ‘Machiavellian motivation’ (i.e. a desire to manipulate and overcome speakers of the target language).
/> Overall, instrumental motivation is a much weaker predictor of L2 achievement than integrative motivation (Masgoret and Gardner 2003).
However, instrumental motivation can play a bigger role in foreign language contexts where learners have little interest in the target language culture. Gardner and Lambert (1972) reported that this was the case with Tagalog learners of L2 English in the Philippines.
The benefits of an instrumental motivation are likely to wear off once the instrumental objective has been achieved as learners cease making any effort to learn. This is especially likely to occur in some foreign language contexts.
Gardner’s Socio-educational Model has been subjected to considerable criticism. It took no account of the impact that success in learning can have on a learner’s motivation. Gardner (1985) claimed that motivation was a causative variable, but this ignores the fact that learners can derive motivation from their actual or perceived success in learning a second language. Gardner’s theory paid scant regard to the fact that a learner’s motivation is not static but dynamic, continuously responsive to the learning conditions. Perhaps the most serious limitation, however, lies in how the role of social milieu was conceived. The key notion of integrativeness is of obvious relevance to a sociocultural context such as Canada where there are clear L1 and L2 communities, but is less clearly relevant to many contexts where the notion of the ‘target language community’ is highly problematic (for example, monolingual contexts such as Japan, or the complex multilingual and multicultural contexts found in the USA). Also, in Gardner’s model, the social milieu is seen as determining the motivational disposition of the learner and no recognition was given of the fact that learners have agency and that local social contexts are actively and continuously constructed by learners (Pavlenko 2002). In other words the ‘social’ component of the theory was under-theorized. See Chapter 9 for further comments on this limitation.
The Socio-educational Model was not the only socio-psychological theory of this period. Clement (1986) failed to find any relationship between integrativeness and English language ability in Francophone students studying at the University Ottowa. However, the learners’ self-confidence (i.e. the extent they believed in their own ability to learn a second language successfully) proved a strong predictor. Self-confidence constituted a social-psychological variable as it originated in the quality and quantity of the contact between members of the target and L2 communities. Clement argued that ‘frequency of contact and the concomitant self-confidence might be more important in determining second language proficiency than socio-contextual or affective factors’ (p. 287).
The cognitive-situated period
In this period, responding to criticisms levelled at the social-psychological approach, researchers turned to mainstream theories of motivation in cognitive psychology. In so doing, they broadened the scope of enquiry into the motivation for second lanaguage learning by examining factors that aroused intrinsic interest in learners and learners’ perceptions of the reasons for their success or failure.
Self-determination Theory
Self-determination Theory (Deci and Ryan 1985) was built around the common-sense notion that people are motivated by both external factors such as rewards, grades, or the opinions of others and by internal ones such as personal interests, curiosity, or experiencing an activity as fun. The theory was applied to L2 motivation by Noels, Pelletier, Clement, and Vallerand (2000), who developed a model based on the distinction between intrinsic motivation and extrinsic motivation. They defined extrinsically motivated behaviours as ‘those actions carried out to achieve some instrumental end’ and intrinsic motivation as ‘motivation to engage in an activity because it is enjoyable and satisfying to do so’ (p. 61). Various sub-categories of each type were distinguished. For example, intrinsic motivation could be derived from (1) knowledge (i.e. the motivation derived from exploring new ideas and knowledge), (2) accomplishment (i.e. the pleasant sensations aroused by trying to achieve a task or goal), and (3) stimulation (i.e. the fun and excitement generated by actually performing a task). Noels et al. also included amotivation (i.e. the absence of any motivation to learn) in their model. In a study of Anglophone learners of L2 French, they found both extrinsic and especially intrinsic motivation were positively related to measures of perceived competence and intention to continue study while amotivation was negatively correlated with these measures. Noels et al. concluded that the more self-determined learners’ motivation, the greater their achievement.
The concept of intrinsic motivation accords more closely with how teachers conceive of students’ motivation. Crookes and Schmidt (1991) observed that ‘it is probably fair to say that teachers would describe a student as motivated if s/he becomes productively engaged in learning tasks, and sustains that engagement, without the need for continual encouragement or direction’ (p. 480). They also recognized that teachers see it as their responsibility to help generate intrinsic motivation in their students.
Attribution Theory
Attribution Theory (Weiner 1992) views motivation as deriving from the explanations that learners give for their progress in learning a second language. As Dörnyei (2005) put it ‘the subjective reasons to which we attribute our past successes and failures considerably shape our motivational disposition underlying future action’ (p. 79). There are three main types of attributions. First, they can be internal (i.e. learners explain their performance in terms of their own ability or lack of it) or external (i.e. learners place the blame for learning problems on external factors). Second, learners can perceive the outcome of their learning efforts as stable or unstable. In the case of the former, learners may be less inclined to make any further effort as they believe it will make no difference, but in the case of the latter, they may try harder. The third set of attributions concerns whether the factors influencing success or failure are seen as controllable or uncontrollable. Learners will be more motivated to improve if they perceive the cause of their difficulties lies within themselves rather than in other people (for example, a poor teacher).
In one of the few L2 studies based on attribution theory, Williams and Burden (1999) reported that British secondary school students tended to explain how well they were performing in terms of external factors. They attributed success to their own aptitude and the efforts of others and failure to the difficulty of the tasks they were assigned or poor teaching. In another study, Ushioda (2001) conducted interviews with adult Irish learners of French to investigate what factors enabled these learners to maintain a positive self-concept and belief in their capacity to learn French. She identified two attributional patterns that contributed to this: (1) attributing positive L2 outcomes to personal ability and/or effort and (2) attributing negative L2 outcomes to temporary shortcomings, such as lack of effort or inadequate opportunity to learn, which could be overcome.
Both self-determination theory and attribution theory recognized the importance of situation-specific factors. These theories also recognize the dynamic nature of motivation as motivation is not fixed, but rises and falls according to the situational conditions and the attributions learners form. However, neither theory fully captures the ebb and flow of motivation over time or within a single lesson.
The process-oriented period
During this period, researchers turned their attention to examining the dynamic character of motivation and the temporal variation that can occur both over the lifetime of a learner and within a single lesson. This phase in motivational research was associated with a number of important developments in the modelling of motivation:
the identification of phases of motivation, involving the reasons for learning a second language, deciding to do something, and sustaining the effort over time (Williams and Burden 1997)
the influence of group dynamics on learners’ motivation
the role of self-regulation in helping learners to maintain their motivation when faced with challenges
the motivational forces that arise as
a product of interacting with others in the second language.
The Process Model of L2 Motivation
The Process Model of L2 Motivation (Dörnyei and Otto 1998) constitutes the fullest attempt to represent the complex, dynamic nature of motivation. It proposes three phases:
Pre-actional phase. This involves goal-setting and the formation of an action plan. Motivational influences in this stage include ‘attitudes towards the L2 and its speakers’, as in Gardner’s model, but also a range of other factors (for example, ‘expectancy of success’ and ‘perceived coping potential’). Dörnyei and Otto refer to this stage as ‘choice motivation’.
Actional stage. This is when learners begin to implement their action plan. It involves ‘executive motivation’. Three basic processes come into play in this stage: implementation of the sub-tasks in the action plan; ongoing appraisal of the extent to which the goals of the sub-tasks have been achieved; and action control involving the use of self-regulatory strategies such as motivational maintenance strategies and language-learning strategies.