The construct of attention has a long and occasionally tortuous history. Though regarded as central to psychology and as fundamental to human experience by James (1890), Wundt (1973), and other founders of psychology, its inability to be characterized unitarily has led many to regard it as theoretically incoherent (Cohen, 1993). Fischler (2001) has discussed the multicomponential nature of attention and the information-processing, factor-analytic, and brain systems contexts in which these processes are studied. The processes or components of attention are frequently characterized as (1) overall arousal, (2) orienting to novel stimuli, (3) selectivity to endogenous or exogenous stimuli, (4) division among concurrent tasks, (5) executive control of attention (resource allocation), and (6) vigilance or sustained attention. Though conceptualized as independent processes or components, the demonstration of this modularity has been difficult to instantiate, and studies designed to do so often remain confounded. For example, selective focus of one's conceptual and perceptual systems to external stimuli requires a mechanism for inhibiting some stimuli while allowing passage and activation of the intended stimuli. This notion invokes the distinctions between top-down and bottom-up processing, resource-versus data-driven processing, and controlled versus automatic processing. It also invokes the notions of selective versus divided attention, as well as an executive system that is capable of directing or allocating mental effort toward specific stimuli or actions. These attentional processes accomplish this in finitely timed intervals and in controlled amounts. Indeed, the models of attention are complex, and the study of attention is untidy. However, the struggle has produced a large and continuous flow of theoretical and experimental evidence supporting its validity as a field of study. The importance of attention in theories of consciousness, cognition, and brain dysfunction justifies the pursuit.

In his important treatise on attention and effort, Kahneman (1973) specified some defining attributes of attention. Among his attributes, he suggested that attention is a limited capacity commodity (whether viewed as a single or multiple pool system). Attention is mobile and can be shifted either through mechanisms of orienting, enduring dispositions, or through the executive control system. The distributor of processing resources allots attention according to a policy that (1) is biased toward novel stimuli, (2) has the ability to allocate attention to a particular domain or message, and (3) operates as a function of externally generated arousal levels. That attention is limited in capacity has been a central organizing principle for much of the research in attention and has given rise to the “dual task” paradigm, a widely used research method for investigating attention. The dual task is an experimental procedure whereby two tasks are performed concurrently and some aspect of each task is manipulated independently. The tasks are frequently manipulated by having the subject voluntarily allocate different percentages of attention or effort to each task (e.g., 50%/50%, 25%/75%, 75%/25%, 100%/0%, 0%/100%). If attention is shared between the two tasks, a trading of performance levels is expected and is typically expressed as a performance operating curve (POC). While the validity of the voluntary allocation part of the design has been challenged (Gopher, Brickner, and Navon, 1982), an alternative method is frequently used in which the inherent difficulty of the two tasks is manipulated parametrically. Again, a trading of performance levels is expected if the two tasks share a common pool or source of attention, and a POC is plotted and measured in order to test this hypothesis. The dual task paradigm, however, is not the only or even the most widely used approach to study attention. Without doubt, the Stroop test (Stroop, 1935) is the most widely researched attention task. In this task, unwanted intrusion of information is assessed through the rapid identification of colors or words of stimuli that are either congruent (e.g., written word “red” in red print) or incongruent (e.g., written word “red” in blue print). In this task, the subject is required to identify either the word or the color, and accuracy and response times are measured. Naming the color of a written word in the incongruent condition produces poorer accuracy and longer response times than in congruent conditions, indicating a competition for activation and inhibition of linguistic and nonlinguistic intentions and stimuli. N400 evoked potentials (Kutas and Hillyard, 1980; Bentin, 1987; Holcomb, 1988) and functional imaging (e.g., Just et al., 1996) are also common methods used to assess the role of attention in language processing.

Experimental paradigms are not the only source of evidence that attention is a construct worthy of study. Introspection also has provided a motivation for entertaining the notion of attention and its relationship to language. Indeed, most adults have had the experience of having read several pages of written material only to discover that nothing of what was read was remembered because the mind had wandered and focused on reviewing yesterday's particularly puzzling diagnostic or a previous argument with a colleague or the dean, or had focused on planning an upcoming holiday, course lecture, or treatment plan. Likewise, most have discovered the need to turn the radio off when encountering a difficult driving condition or courteously requesting the children to refrain from their backseat banter while formulating a response to the patrolman approaching the car following an apparent traffic violation. Although it is quite intuitive that attention to both internal and external stimuli plays an important role in many (perhaps all) language tasks, major syntheses and analyses of the general attention literature (e.g., Lang, Simons, and Balaban, 1997; Pashler, 1998) and the neuropsychological deficits of attention (e.g., Cohen, 1993; van Zomeren and Brouwer, 1994) have failed to address the role of attention in language processing. Indeed, not one of these major texts devoted to attention have addressed the role of attention in developmental disorders of language such as specific language impairment (SLI), or in acquired disorders affecting language processing, such as those of aphasia or traumatic brain injury. Only very recently has this subject received space in edited books on language and aphasia (e.g., Nadeau, Rothi, and Crosson, 2001) and only relatively recently have theoretical formulations (McNeil, Odell, and Tseng, 1991) of how attention might account for language impairments, and summary reviews (Murray, 2000; Crosson, 2001a; Fischler, 2001) offered explanations or hypotheses of how attention might interact with language impairments.

Language knowledge is characterized by the information that is represented in the brain along with the rules that govern it. Linguistic theory attempts to account for the structure of the information (rules and representations) that is stored in memory. Psycholinguistic theory attempts to account for conditions under which the rules and representations are stored or accessed and the various ways in which the different components are combined to produce or comprehend sounds, morphophonemes, words, phrases, sentences, and discourse. Informing and directing the linguistic system requires each individual language user to engage in a finitely tuned interplay between internally generated intentions, linguistic knowledge, and a massive amount of sensory information that is continuously available, in addition to the selection (planning), programming, and execution of appropriate responses. This temporally demanding interplay creates an astonishing array of factors that have to be sorted and managed at all instances in time and on a continuous basis. It is the domain and role of attention and resource allocation to account for the gating (inhibition) and activation of endogenous intentions and exogenous stimuli involved in the formulation, comprehension, and production of language. Indeed, the role of attention in normal language processing has a long history, and evidence supports the conclusion that all levels of language processing require and compete for attentional resources with other language processes and with the processing of nonlinguistic information. For example, the attentional demands placed on language processing have been illustrated for lexical/semantic processing through priming paradigms (Neely, 1977), through evoked potentials (Kutas and Hillyard, 1980), and through dual-task studies (Arvedson and McNeil, 1987; Murray, 2000). Attentional demands for language have also been demonstrated in dual-task studies for syntactic processing by Blackwell and Bates (1995), for phonemic processing by Tseng, McNeil, and Milenkovic (1993), for auditory prosody processing by Slansky and McNeil (1997), and between language and nonlinguistic tasks by LaPointe and Erickson (1991).

Disorders of language are common and account for a sizable proportion of all communication disorders. Within the various classification systems for language disorders, it is widely recognized that there are multiple causes. Most systems acknowledge deficits at the representational level, including the rules used to govern these representations. Deficits at this level are often referred to as deficits of linguistic competence. A variety of performance factors are also recognized that can cause an otherwise competent or intact linguistic system to malfunction. Examples of performance deficits include disorders of linguistic-specific memory processes (Baddeley, 1993; Crosson, 2001b) and slowed perceptual or cognitive mechanisms (Tallal, Stark, and Mellits, 1985). Disorders of various aspects of the attentional system include orienting of attention (Robin and Rizzo, 1989), selective attention (Petry et al., 1994; Murray, Holland, and Beeson, 1998), inability to engage or disengage attention (Posner, Snyder, and Davidson, 1980), and resource allocation (McNeil, Odell, and Tseng, 1991). The construct of attentional deficits underlying language deficits is neither new (e.g., Kreindler and Fradis, 1968) nor restricted to aphasia. Campbell and McNeil (1985), for example, illustrated attentional deficits in an acquired pediatric language disorder population, and Barkley (1996) has applied the construct to attention deficit-hyperactivity disorder. However, restricting the discussion to the language impairment in aphasia, the past decade has seen a renewed interest in various aspects of attention, but primarily in the allocation of processing resources.

While skepticism remains apparent in some circles, it is widely recognized that explanations of language and other domains of cognition (e.g., memory, learning, executive function) that fail to account for attentional phenomena will remain incomplete. This is especially true for those areas of cognitive dysfunction resulting from brain damage (congenital or acquired, regardless of the time or cause of the injury) and developmental disabilities.