Mind and Context in Adult Second Language Acquisition: Methods, Theory, and Practice

6. Attention and Awareness in SLA

six Attention and Awareness in SLA ronald p. leow melissa a. bowles key words Attention ■ awareness ■ detection ■ limited capacity processor ■ noticing ■ operationalization ■ verbal reports. 1. Introduction The 1990s witnessed several major theoretical approaches to the roles of attention and awareness in second or foreign language (L2) learning, mainly in the formal classroom setting (e.g., Robinson, 1995b; Schmidt, 1990, 1993, 1994, 1995, 2001; Tomlin and Villa, 1994). This chapter presents first a brief report of major models of attention in cognitive psychology, followed by a concise review of the important tenets of the major attentional frameworks in second language acquisition (SLA). We then discuss methodological issues associated with the operationalization and measurement of attention and awareness in SLA research methodology, followed by a brief report of several studies that have employed verbal reports to operationalize the constructs of attention and awareness before statistically addressing the effects of attention and awareness on classroom L2 development. 2. Theoretical Models of Attention Almost all theories of SLA posit some role for attention, but the construct is especially emphasized in cognitivist accounts, where it has been proposed that “attention appears necessary for understanding nearly every aspect of second and foreign language learning” (Schmidt, 2001, p. 6). This view is based in large part on a long-standing and empirically supported position in cognitive psychology that maintains that attention to stimuli is needed for longterm memory storage and that little, if any, learning can take place without attention (Carlson and Dulany, 1985; Carr and Curran, 1994; Nissen and Bullemer, 1987; Posner, 1992; Reber, 1967, 1976, 1989, 1993). A number of theoretical models and hypotheses have been postulated about the specific underpinnings of attention. First, the major models from cognitive psychology 179 will be reviewed, followed by those that have been proposed specifically for SLA. 2.1. models from cognitive psychology In the classic cognitive psychology literature on attention, the common view was that human attention is of limited capacity by nature and that humans are limited capacity processors (e.g., Broadbent, 1958; Kahneman, 1973). Because this view held that there is a limited supply of attention, it further maintained that attention must be selective. There was disagreement over the exact nature of that selectivity, however. Filter theories of attention proposed that there is a sort of filter or bottleneck that allows only some stimuli to be processed (Broadbent, 1958). This was an “early selection” model, since stimuli are either selected or not early in the process (they are either allowed through the bottleneck or not). This view furthermore held that information is processed serially, in such a way that attention to one message blocked attention to another. Later research (especially dichotic listening studies) led researchers to propose an alternative theory that all information in the input stream is processed in parallel and that selection of what will be attended to happens late. This was a “late selection” model (Deutsch and Deutsch, 1963). Both the early and the late selection models just discussed share the common view that the human processor is a passive recipient of information. However, more recent models have stressed a more active role for information processors. For instance, Kahneman (1973) proposes that attentional allocation is effortful and although attention is a limited resource, it can vary as a function of arousal. That is, for Kahneman, two tasks can be performed simultaneously as long as there is sufficient arousal and task demands are not too high. Similar to Kahneman, Wickens (1980, 1984, 1989) assumes that the allocation of attention is effortful. However, his model proposes that there are multiple pools of attentional resources rather than a single pool. Wickens contends that the difficulty of performing two tasks simultaneously depends on whether the necessary attentional resources are being drawn from the same pool (serial processing) or from different pools (parallel processing). According to Wickens, it would be more difficult to perform two tasks drawn from the same pool (e.g., talking to someone and writing an essay) than to perform two tasks drawn from different pools (e.g., driving a car and listening to the radio). He posits that two tasks drawn from the same pool of resources can be performed simultaneously only when one of the tasks has become automatized.1 In contrast to the limited capacity theories discussed up to this point, some recent attentional theories contend that human attentional resource capacity is 180 internal factors in fact unlimited (Neumann...