CHAPTER 12 Attention deficits in communicatively-disordered populations are well documented (Blake, Duffy, Myers, & Tompkins, 2002; Erickson, Goldinger, & LaPointe, 1996; Fillingham, Sage, & Lambon Ralph, 2006; Murray, 1999; Myers & Blake, 2008; Peach, Rubin, & Newhoff, 1994) and are frequently described as a contributing factor to the language impairments that are observed in these groups. That is, some of the language problems of communicatively-impaired individuals are thought to be due, at least in part, to a reduction in attentional abilities that are needed to (a) focus a listener on incoming verbal information while excluding competing or distracting information, (b) maintain a continuous record of conversation and context to support the interpretation of new information, and (c) construct coherent verbal outputs from among a number of contending alternatives (Crosson, 2000; Chapter 8, this text). For example, attentional impairments have been associated with problems in listening and reading comprehension following aphasia (Coelho, 2005; Murray, 2002; Murray, Keeton, & Karcher, 2006; Sinotte & Coelho, 2007), with the conversational problems observed in individuals with Alzheimer disease (Alberoni, Baddeley, Della Sala, et al., 1992), right-hemisphere damage (Myers & Blake, 2008) and traumatic brain injury (Stierwalt & Murray, 2002; Ylvisaker, Szekeres, & Feeney, 2008), and in the lexical and discourse deficits found in the language production of individuals with aphasia (Hula & McNeil, 2008, Murray, Holland, & Beeson, 1998), Alzheimer’s disease (Kempler, Andersen, & Henderson, 1995; Neils, Roeltgen, & Greer, 1995), right-hemisphere damage (Myers & Blake, 2008), and traumatic brain injury (Ylvisaker et al., 2008). It is not surprising, then, that some recent approaches to the rehabilitation of such language disorders have emphasized the amelioration of a variety of attention deficits, including those concerned with selecting, sustaining, dividing, and alternating attention to both external and internal information. In such an approach, attention is typically conceived as a general pool of resources that are exogenous to but nonetheless critical for a variety of communicative behaviors. Treatments that stimulate and improve attentional processing might then be assumed to yield similar improvements in communication behaviors that depend on attention. Unfortunately, this has not been the case. While improvements in the specific skills necessary to perform a variety of tasks have been reported following attentional treatments, evidence for generalization of these improvements to related behaviors has been lacking (Park & Ingles, 2001; Sohlberg, Avery, Kennedy, et al., 2003). These observations have led some to suggest that, in order for attentional treatments to be effective with regard to behaviors such as language, these treatments should be performed within the language domain and under conditions that compete for attentional resources (Fischler, 2000; Hula & McNeil, 2008). The premise underlying this position is that attentional processing is apt to improve only when such endogenous resources are deployed in the service of specific language tasks that requires control over and coordination of multiple processes, such as semantic, syntactic, and phonologic. Inasmuch as previous treatment studies have not necessarily focused on attention for specific cognitive tasks like language under conditions that challenge patients’ attentional control, this may be one reason for the poor generalization of improved attention to untreated behaviors such as language. Such an approach is consistent with the perspective that language is not simply an object of attention but rather is, in itself, an attention-focusing mechanism (Langacker, 2008). According to Crosson (2000), intention (the preparation to use language) “affects attention because the intention to perform a particular activity determines the particular internal and external sources of information to which we attend” (p. 375). For example, lexical selection (for both open- and closed-class words) engages central attentional mechanisms (Ayora, Jannsen, Dell’Acqua, & Alario, 2009; Hula & McNeil, 2008) while grammaticalized elements (conjunctions, prepositions, bound morphemes, etc.) direct the listener’s attention to important aspects of sentences that convey the specific meanings intended by speakers or writers (Taube-Schiff & Segalowitz, 2005). In discourse, anaphoric constructions (i.e., those that use a noun or pronoun to refer to an entity that was previously mentioned) require that attention be directed to an antecedent occurring in an earlier sentence (Myachykov & Posner, 2005). It might be expected then that problems in adequately “windowing” attention (directing the distribution of attention over a referent scene in a specific pattern) (Talmy, 2003) for these and other language tasks can produce the types of communication disturbances identified above. Attention deficits are among the most widely reported cognitive problems following brain damage and contribute to the acquired language disorders seen in these patients. A brief overview of the attention and language problems associated with various neuropathologies is provided here (see also Chapter 3 and Box 12-1). Box 12-1 Functional Equivalency of Attention Deficits? Although discussions of the impact of attentional deficits on language functioning have traditionally been organized according to the specific neuropathologies underlying those impairments (i.e., stroke to one hemisphere or the other, dementing processes, traumatic brain injury), it should not be surprising that attentional deficits arise following most types of brain damage. Inasmuch as the neural network that underlies attention is distributed with bilateral cortical and subcortical contributions (Filley, 2002; Knudsen, 2007; Mesulam, 1990, 1998), attentional deficits should be expected from most any type of brain damage. While it may appear that the attentional impairments identified above have been assigned unique status based upon their underlying neuropathologies, all of the language deficits associated with these clinical groups can be associated with four processes that are fundamental to attention: working memory (including spatial working memory), top-down sensitivity control, competitive selection, and automatic bottom-up filtering for salient stimuli (Knudsen, 2007) (see Figure 12-1). According to Knudsen, working memory is a highly dynamic form of memory that operates over seconds and temporarily stores information for detailed analysis. Competitive selection is the process that determines which information gains access to working memory. Top-down sensitivity control regulates the relative signal strengths of the different information channels that compete for access to working memory while salience filters automatically enhance responses to infrequent or biologically-important stimuli. Despite the dissimilarity of the clinical populations that have been studied, it may be that the attentional impairments that are observed among these patients are functionally equivalent (Ylvisaker, Hanks, & Johnson-Green, 2003). When viewed through this lens, the lack of differences in attentional impairments demonstrated by patients with brain damage in contrasting regions of the same (Murray et al., 1997) or different cerebral hemispheres (Arvedson & McNeil, 1987; Coslett, 1999; Murray, 2000) should not be unexpected. And in the absence of evidence establishing that the attentional impairments on similar language tasks within and across different clinical populations are qualitatively different, the approaches taken to rehabilitation of communication problems secondary to attention deficits have continued to rely on treatments that address the fundamental processes described in patients with a variety of underlying neuropathologies (Coelho, 2005; Crosson, 2008, Crosson, Fabrizio, Singletary, et al., 2007; Dotson, Singletary, Fuller, et al., 2008; Helm-Estabrooks, Connor, & Albert, 2000; Murray, Keeton, & Karcher, 2006; Peck, Moore, & Crosson, 2004; Sinotte & Coelho, 2007; Youse & Coelho, 2009). In a sample of consecutive patients with a wide range of ages and lengths of hospitalization and an even distribution of left and right hemisphere stroke, Hyndman, Pickering, and Ashburn (2008) found high levels of attention deficits at the time these patients were discharged to the community. Fifty-one percent of the patients demonstrated divided attention deficits while approximately 37 percent of the patients demonstrated sustained and auditory and visual selective attention deficits. While some improvements could be detected at 6 and 12 months after discharge, there was no clear recovery of attentional abilities in these patients. Knopman, Roberts, Geda, et al. (2009) investigated the relationship between cognitive impairment and stroke in a population-based sample of elderly individuals with mild cognitive impairment (MCI) and no evidence of dementia. MCI was classified as either amnestic if there was evidence for memory impairment or nonamnestic if there was no memory impairment. Participants underwent nursing, neurological and neuropsychological evaluations to assign a clinical diagnosis. Stroke history was then obtained from the study participants and verified by medical records and the findings obtained in the neurological evaluation. Logistical regression analyses demonstrated a higher risk of MCI in study participants with a history of stroke. With regard to MCI subtype, the association of stroke was greater for nonamnestic versus amnestic MCI. A history of stroke was also found to be associated with lower functioning in each of the cognitive domains that were tested except for memory. The association was found to be strongest for attention and executive functioning. Arvedson and McNeil (1987) compared accuracy and oral response times from left-hemisphere damaged aphasic (LH), right-hemisphere nonaphasic (RH), and non-brain-damaged (NBD) individuals for two focused attention tasks (semantic judgment and lexical decision) under binaural listening conditions. For semantic judgment, the LH group was less accurate and had significantly longer response times than the NBD group. The RH group did not differ from either of the other two groups with regard to accuracy or response time. For lexical decision, both brain-damaged groups performed more poorly than the NBD group although there were no group differences for overall response times. Arvedson and McNeil (1987) concluded that the performance deficits observed in both groups were consistent with problems in attention and resource allocation. Coslett (1999) investigated verbal processing following left and right hemisphere strokes as a function of the side of space to which subjects directed their attention. The study was motivated by reports of improved performance in stroke patients on a variety of sensory and motor tasks when stimuli were presented in the ipsilesional versus the contralateral hemispace. Unlike previous studies, however, this study sought to assess (in addition to other goals) the degree to which hemispatial influences affect language processing, a behavior that doesn’t appear to have a critical dependence on spatial representations. Because of the documented role of the parietal lobes in spatial functioning, Coslett predicted that patients with parietal lobe lesions would perform best when stimuli were presented in the ipsilesional space. Further support for the spatial registration hypothesis might be found in a study by Ansaldo, Arguin, and Lecours (2004). In their longitudinal investigation of recovery from Wernicke aphasia in a patient with a left parietotemporal stroke, they demonstrated that improved lexical semantic processing, as assessed by a lexical decision task, was correlated with presentation in the left visual but not right visual hemispace. An interaction between the grammatical class and the imageability of the lexical stimuli presented to the left visual hemispace suggested that the right hemisphere contributions to the observed recovery were linguistic. But correlations between the patient’s global (non-lateralized) lexical and attentional performance, as assessed by the Nonverbal Stroop Test, suggested that recovery was mediated by attentional factors as well. While the authors conclude that the findings associated with presentations to the left visual hemispace likely represented premorbid language abilities of the right hemisphere, the observed language facilitation may just as well have been associated with the lexical highlighting that resulted with presentations in the ipsilesional hemispace. Murray (2000) evaluated the influence of attentional deficits resulting from left-hemisphere (aphasic) and right-hemisphere brain damage (RBD) on word retrieval following stroke. Subjects completed phrases that were either highly (responses are drawn from a limited or closed set of choices) or minimally (responses are from open set with many plausible choices) constrained under a series of conditions with increasing attentional demands. In the single-task condition, subjects completed the phrases or discriminated tones (high versus low) in isolation (i.e., without distraction). In the focal attention condition, phrase and tone stimuli were presented simultaneously but the subjects completed one task only (phrase completion or tone discrimination). In the divided attention condition, subjects heard the phrase and tone stimuli simultaneously, discriminated the tones, and then completed the phrases. For both LH and RH groups, word retrieval accuracy was influenced by attentional demands. Neither group performed differently from a non-brain-damaged group in the single-task condition. However, in the focused and divided attention conditions, non-brain-damaged subjects performed significantly better than LH and RH subjects. Murray interpreted the lack of differences in the word retrieval abilities of the two brain-damaged groups as evidence that the deficits were not of purely linguistic origin. Instead, the results provide support for a negative interaction between attention and language processing in both aphasic and RBD subjects. Murray, Holland, and Beeson (1997) found evidence for attention and resource allocation deficits on auditory-linguistic listening tasks in stroke patients with mild aphasia. Patients with frontal versus posterior lesions were compared to normal listeners on semantic judgment and lexical decision tasks during three listening conditions: isolation, focused attention, and divided attention. In the isolation condition, subjects performed either task without distraction; in the focused attention condition, subjects listened to competing primary and secondary stimuli but completed the primary listening task only. In two divided attention conditions, subjects again listened to competing primary and secondary stimuli but were required to complete both tasks. The type of distraction in the secondary task was either verbal (semantic judgment and lexical decision competing with each other) or nonverbal (semantic judgment or lexical decision competing with a tone discrimination task). While aphasic and control subjects performed comparably during isolation conditions, both aphasic groups performed less accurately and more slowly than the normal group during the focused and divided attention conditions. The performance differences between the groups increased as the complexity of the listening conditions increased. Greater dual-task interference was observed in aphasic and normal subjects when the secondary task was verbal versus nonverbal. Murray et al. (1997) concluded that (a) the differences between aphasic and normal subjects on auditory-linguistic tasks are more quantitative than qualitative and support the concept of attention allocation inefficiency as an explanatory construct for aphasic performance; (b) the underlying cause of these inefficiencies is damage to a diffusely-represented attentional network that involves both frontal and posterior components; and (c) the greatest decrements in aphasic performance should be expected when linguistic processing demands competition for verbal attentional resources. Murray, Holland, and Beeson (1998) found that these attentional impairments also negatively influence the spoken language production of individuals with mild aphasia. In this study, the morphosyntactic, lexical, and pragmatic characteristics of picture descriptions produced by normal and aphasic speakers were assessed under conditions imposing increasing demands on attention allocation (isolation, focused attention, and divided attention). The distracter in this study was the tone discrimination task. When compared to normal speakers, aphasic subjects produced significantly fewer well-formed utterances and significantly more simple versus complex sentences in the divided attention condition. They also produced significantly fewer words, more word-finding errors, and fewer correct information units (Nicholas & Brookshire, 1993) as attentional demands increased. The number of unsuccessful utterances (failure to communicate accurate and novel information or failure to follow directions) also increased significantly in the divided attention conditions when compared to isolation or the focused attention condition. In a retrospective chart review of a large inpatient rehabilitation unit, Blake and colleagues (2002) found relationships between attentional and other cognitive deficits and the presence of acquired pragmatic disorders associated with hypo-responsiveness, hyper-responsiveness, and interpersonal interactions. In addition, basic deficits in expressive and receptive language functions were found in approximately one quarter of their patient sample. Myers and Blake (2008) suggest that attentional deficits impair RBD patients’ ability to (a) appreciate visual and verbal cues within the context of communication, (b) shift attention during conversations, and (c) sustain attention to the communication environment and filter distractions. They also suggest that attentional deficits place more demands on cognitive resources, especially as interactions become more complex. These contribute to difficulties in forming and maintaining inferences about the meaning of verbal communications. As a result, RBD patients often miss the overall theme, or central point, of narratives. According to Myers and Blake (2008), communicative interactions increase in complexity when they require participants to form elaborative inferences. Elaborative inferences require an individual to attend to and make predictions about the emotions or motives of a conversational partner and therefore go to the implied rather than literal meaning of a communication. Doing so requires integration of multiple cues or selecting from several possible interpretations. Selective attention deficits may interfere with the ability to filter extraneous information and recognize important contextual cues that may manifest as seemingly irrelevant interpretations of, and responses to, the communication environment (e.g., attending to and/or describing isolated details of an integrated scene, failing to appreciate jokes or indirect requests, or producing inappropriate statements based on faulty inferences during discourse). These problems appear to arise in discourse when revision of an initial inference is required for correct interpretation. Two different theories of faulty lexical-processing have been described to account for these deficits. One suggests that RHD patients activate diffuse semantic fields in response to lexical inputs that include distant and unusual semantic features; the fields are shaped by context and modulated by attention and time course. These large semantic fields provide course interpretations of words that are frequently ambiguous but important for understanding natural language (Jung-Beeman, 2005; Tompkins, Fassbinder, Scharp, & Meigh, 2008). It has been suggested that the problems RHD patients have with comprehending inferences may stem from difficulty sustaining distant activations that are needed to correctly interpret non-literal interpretations of a discourse (Tompkins, Scharp, Meigh, & Fassbinder, 2008). A second theory suggests that RHD patients maintain initial interpretations even when the context indicates that they are inappropriate (Tompkins, Baumgaertner, Lehman, and Fassbinder, 2000). That is, RBD patients appear to activate both initial and revised inferences, but have difficulty suppressing an irrelevant or incompatible interpretation once it has been activated. Deficient suppression therefore might be understood as an impairment of inhibition (as conceived in models of selective attention) resulting from the demands suppression places on attentional resources (Tompkins et al., 2000; Tompkins, Blake, Baumgaertner, & Fassbinder, 2002). More recent evidence has confirmed that RBD patients generate and generally maintain inferences similarly to non-brain-damaged individuals (Blake, 2009) although these inferences did not concern ambiguities as in previous work. Attentional impairments following Alzheimer disease (AD) are now well known (Belleville, Chertkow, & Gauthier, 2007; Foldi, Lobosco, & Schaefer, 2002; Levinoff, Saumier, & Chertkow, 2005). In their review of attentional functions in AD, Parasuraman and Haxby (1993) suggest that attentional deficits may appear coincidentally with the memory deficits that arise in the earliest stages of the disease. While selective attention may be spared in some patients, disengagement of attention, attentional switching between spatial locations and stimulus categories, divided attention for auditory, visual, and motor tasks, and sustained attention in conditions requiring effortful processing have been found to be impaired in even mildly affected individuals. While the potential for attention deficits to disrupt language is recognized (Foldi et al., 2002), few studies have investigated this relationship directly. Alberoni et al. (1992) found that patients with AD have difficulty following even simple conversations and that this problem becomes amplified when the conversation involves multiple participants and/or they move from one location to another. They attributed this impairment to divided attention deficits that make it difficult for these patients to shift and refocus attention as well as track and remember the individual contributions and locations of each participant to a conversation. Deficits in attentional control (i.e., the focusing on and recall of important information) may also contribute to these conversational difficulties (Castel, Balota, & McCabe, 2009). Kempler, Andersen, and Henderson (1995) assessed naming in patients with AD and found highly variable performance that was associated with attentional performance. Participants who consistently erred on the same items over two occasions were thought to demonstrate a deficit in lexical semantic representations while those who inconsistently erred on stimulus items were thought to demonstrate deficits in lexical access. The less consistent participants were significantly more impaired on tasks assessing attention. Patient severity (disease, anomia) could not account for the results. The authors concluded that impaired attention, along with deficits in lexical knowledge, contribute to anomia in AD. Neils, Roeltgen, and Greer (1995) investigated the spelling abilities of people with mild AD. Participants completed tests of direct and delayed word copying, spelling to dictation for regular, irregular, and non-words, and written picture description. They also completed tests of sustained attention (letter cancellation), visual attention (visual search), and language ability (Boston Naming Test). The percentages of phonemically implausible (PI) spelling errors for the AD participants as well as for a group of matched normal participants were calculated for the real words in the test battery. AD participants were found to produce more PI spelling errors than their normal counterparts. The two visual attention tests were found to be better predictors of these errors than the language test. The typical pattern of cognitive decline in AD is generally thought to be one of early episodic memory loss followed by combinations of attention-executive, language, and visuospatial impairment, although there is growing evidence of atypical focal cortical presentations of AD (Alladi, Xuereb, Bak, et al., 2007). Recently, Davidson, Irizarry, Bray, et al. (2009) analyzed the scores obtained from administration of the Mini-Mental Status Examination and the Mattis Dementia Rating Scale-2 to a large group of patients with mild/moderate AD to explore the existence of cognitive subgroups within this sample. Four subgroups were identified: a mild group and a severe group with fairly uniform impairment across cognitive domains that were distinguished by the severity of their impairments; a memory group with impairment of memory and orientation and relative sparing of attention, construction, and language; and an attention/construction group with impairments in attention and construction and relative sparing of memory and orientation. The attention/construction group was also characterized by language impairment at levels that were similar to those observed in the severe group. Separate groups with prominent deficits in language and visuospatial construction were not identified. Impairments to language (Coelho, 2007; Hagen, 1984; Levin, 1981; Sarno, Buonaguro, & Levita, 1986) and attention (Stierwalt & Murray, 2002; Willmott, Ponsford, Hocking, & Schonberger, 2009) are reported frequently following traumatic brain injury (TBI). In a study of 25 participants with severe closed head injuries, impairments in lexical-semantic and sentential semantic skills, verbal fluency, complex auditory comprehension, and attentional operations were found to comprise a set of “cardinal” cognitive-linguistic deficits following TBI (Hinchliffe, Murdoch, Chenery, et al., 1998). The co-occurrence of these deficits suggests that the problems these individuals have with language may be a result of difficulties in allocating attentional resources effectively for linguistic cognitive operations (Peach, 1992). That is, TBI, even in mild cases, may affect how much and how rapidly linguistic information can be processed (Whelan, Murdoch, & Bellamy, 2007). The profile of language deficits following TBI is generally referred to as confused or disorganized, which suggests that the disorder is due, at least in part, to problems in verbal planning (Alexander, 2002). Recent work has demonstrated that TBI patients have difficulty planning sentences in isolation (Ellis & Peach, 2009) and in discourse (Deschaine & Peach, 2008). Deficient planning for sentences in discourse have been found to be related to difficulties allocating attention for complex tasks (as indexed by the Trail Making Test, Part B) and suggest that the problem may be part of a more global planning impairment resulting from executive dysfunction. Such difficulties might be thought to result from impairment to the supervisory attentional system, a voluntary, top-down component of the executive system that facilitates the activation of mental schemas that are needed to interpret inputs and determine subsequent actions (Shallice, 1982). Sentential and discourse impairments therefore might be seen as a failure of executive control over cognitive and linguistic organizing processes (Ylvisaker, Szekeres, & Feeney, 2008). Patients who suffer TBI often have difficulties participating in conversations; these difficulties can be linked to their attentional impairments (Coelho, 2007; Stierwalt & Murray, 2002). They tend to demonstrate problems with maintaining or extending the topic of discussion, using reference, and integrating relevant information because of poor sustained or selective attention. Their communication may be incoherent due to difficulty attending to and maintaining a plan for discourse as well as the listener’s perspective. They also produce socially inappropriate output because of a failure to attend to social cues. The outcomes associated with attentional treatments for acquired language disorders have been weak (Rohling, Faust, Beverly, & Demakis, 2009) and are most likely due to the generalized or non-specific approach that these studies have taken with regard to attention intervention. For example, Helm-Estabrooks, Connor, and Albert (2000) used nonlinguistic tasks to treat sustained, selective, and alternating attention. Researchers at the University of Florida treat spatial attention by increasing their patients’ orientation to left hemi-space during picture naming so as to exploit right-hemisphere attention mechanisms (Crosson et al., 2007; Dotson et al., 2008; Peck et al., 2004). Still others (Coelho, 2005; Murray et al., 2006, Sinotte & Coelho, 2007; Youse & Coelho, 2009) have treated focused, alternating, selective, and divided attention using a variety of linguistic stimuli (numbers, letters, words) and tasks that are included in Attention Process Training II (Sohlberg, Johnson, Paule, Raskin, & Mateer, 2001), a program to treat attention impairments in patients with mild cognitive impairments. While all of these approaches assume that improved language will result from increased attention to linguistic stimuli, none of them are motivated by an analysis of the ways that specific linguistic processes recruit select attentional operations in the service of language. The need for specificity in attention treatment has been addressed previously. Sturm, Willmes, Orgass, and Hartje (1997) demonstrated that specific attention functions improve in patients with localized vascular lesions only when specific training is received for that function. In this study, computer based programs were used to train the intensity (alertness, vigilance) and selectivity (selective and divided attention) aspects of attention. Even when patients demonstrated deficits in both domains of intensity or selection, improvements were only noted for the single domain that received training. This was particularly evident for the intensity aspects of attention. Specific attention training for alertness has also been found to contribute to reorganization of the right-hemisphere functional network known to subserve the alertness domain in normal subjects. Similar reorganization was not observed for right-hemisphere brain-damaged patients who received non-specific (memory) training for alertness (Sturm, Longoini, Weis, et al., 2004). Park and colleagues (Park, Proulx, & Towers, 1999; Park & Ingles, 2001; Park & Barbuto, 2005) find no evidence to support approaches that incorporate direct training of distinct attentional components (e.g., sustained, selective, divided, and alternating attention). They have found, however, that attention treatments that focus on learning or relearning of specific skills that are important to desired outcomes or behaviors that have functional significance resulted in significant improvement. An interesting outcome, for the purposes of this chapter, regarding the need for specificity in attention training was reported by Curran, Hussain, and Park (2001, as cited in Park & Barbuto, 2005). In their study, patients with mild cognitive impairment following stroke learned novel naturalistic actions (goal-directed activities that require the production of several actions in a particular order that cannot be learned prior to instruction, e.g., preparing an unfamiliar recipe) more effectively when the trainer verbally described the action while demonstrating it than when no verbal description accompanied the demonstration. The authors hypothesized that the verbal descriptions facilitated learning of the actions by enabling the patients to develop a more accurate conceptual representation of the novel actions. However, in more severely impaired patients, the additional information may actually impair performance because of the demands integrating verbal and visuo-spatial information place on patients with more limited cognitive resources (Green, Rich, & Park, 2003; Park & Barbuto, 2005). Accounts of sentence processing that argue for modularity—that is, autonomy of lexical access and syntactic analysis—are not uncommon (e.g., Frazier & Fodor, 1978; Fodor & Frazier, 1980; Swinney, 1979). These accounts consider sentence processing to be an isolated process independent of the conversational and real-world contexts in which such sentences occur. An alternative approach suggests that the processor attends to the referents that are being described (Altmann, 1996). That is, sentence processing focuses attention on the relevant aspects of the real-world context. Language processing therefore cannot be separated from the real-world context onto which the language must be mapped. Such fundamental relationships between language and attention can be expressed in the related concepts of grounding and windowing. Grounding refers to a speaker’s use of linguistic elements to direct a hearer’s attention to a particular meaning within a discourse (Langacker, 2008; Taube-Schiff & Segalowitz, 2005). Windowing refers to the way in which languages use explicit mention to place a coherent referent situation into the foreground of attention; omission of any mention of other portions of the situation places that information into the background of attention (Talmy, 2003). Through nominal grounding (e.g., the, this, that, some, a, each, every, no, any), the speaker directs the hearer’s attention to the intended discourse referent, which may or may not correspond to an actual individual. Clausal grounding (e.g., -s, -ed, may, should, will) situates the profiled relationship with respect to the speaker’s current conception of reality. (Langacker, 2008, p. 259) For conceptions evoked as meanings of linguistic expressions, the speaker and the hearer are the primary conceptualizers whose interactions in producing and understanding an utterance form the ground. Individually and together, the speaker and hearer function as the subject of conception and figure, at least minimally, in the meaning of every utterance (Figure 12-2). An important aspect of the subject’s activity is the focusing of attention. Within the full scope of awareness for the content of an utterance, the subject attends to a certain region (Langacker’s “onstage” region) and further singles out some onstage linguistic or grammaticized element as the focus of attention. This is the object of conception, which can be either a thing or a relationship. As the focused object of conception, it is interpreted most clearly with respect to the context and the listener (see Figure 12-2). Taube-Schiff and Segalowitz (2005) provided evidence that such grammaticized elements require a listener to refocus attention when such attention-directing words are encountered in natural language. In their study, participants demonstrated greater demands for attentional control (operationalized as shift costs in an alternating runs experimental design) when judging spatial (above-below) or temporal (past-present) function words embedded in phrases. Greater shift costs were also observed in a grammatically dissimilar condition (participants required to shift between spatial and temporal function words) versus a grammatically similar condition (participants required to shift between spatial words describing a vertical dimension, i.e., above-below, or a proximal condition, i.e., near-far). The authors concluded that “language itself acts as an attention-focusing mechanism, shaping the creation of a mental construction by the recipient that corresponds to the sender’s meaning” (p. 516). Coventry, Lynott, Cangelosi, and colleagues (2010) examined how such spatial language (e.g., the bottle is over the glass) directs attention to a visual scene. Two views have been offered as to how this might occur. In the first, spatial language directs the hearer’s attention to a reference object in the array being described and then specifies how attention should be switched to the object to be located. A second view takes into account how objects are experienced and used in the world. So, according to the first view, the sentence above would activate a minimal representation of a bottle oriented in the standard position higher than a glass. In the second view, the sentence would activate knowledge that includes the placement of the objects according to the way objects typically interact. Following two experiments designed to discriminate how spatial language drives visual attention, Coventry and colleagues concluded that spatial language comprehension is associated with a situational representation of how objects usually function. Spatial language therefore is thought to summon a range of perceptual simulations (i.e., dynamic routines) of the typical interactions among objects, including motion processing where attention is directed to objects not mentioned in the heard sentence (see path windowing below). Talmy (2003) describes several types of event frames and the types of windowing that operate on these events (Table 12-1). The first is a path event frame, which gives rise to path windowing. A path event frame refers to the entirety of a path of motion. There are three categories of paths: open, closed, and fictive. Open paths concern the paths that are described by objects that are physically in motion during a period of time and have beginning and ending points that are in different locations in space. For example, the sentence in (1) illustrates an open path event with maximal windowing over the whole of the conceived path while (1a) presents examples of gapping over one portion of the path and (1b) shows windowing over a single portion of the path. Table 12-1 Event Frames and Associated Types of Windowing* *See Text for Examples of Sentences Displaying Each Event Frame. Adapted from Talmy, L. (2003). The windowing of attention in language. In Toward a cognitive semantics volume I: Concept structuring systems (pp. 258–309). Cambridge, MA: The MIT Press. (1)(a) Maximal windowing: The crate that was in the aircraft’s cargo bay fell out of the plane through the air into the ocean. (b) Medial gapping: The crate that was in the aircraft’s cargo bay fell out of the plane into the ocean. (c) Initial gapping: The crate that was in the aircraft’s cargo bay fell through the air into the ocean. (d) Final gapping: The crate that was in the aircraft’s cargo bay fell out of the plane through the air. (e) Initial windowing: The crate that was in the aircraft’s cargo bay fell out of the plane. (f) Medial windowing: The crate that was in the aircraft’s cargo bay fell through the air. (g) Final windowing: The crate that was in the aircraft’s cargo bay fell into the ocean. Fictive motion sentences attribute figurative motion to static objects in space. In fictive motion sentences, a motion verb is applied to a subject that is not literally capable of physical movement (e.g., the path swings along the cliff, the tattoo runs along his spine) (Ramscar, Matlock, & Boroditsky, 2009; Ramscar, Matlock, & Dye, 2009). According to Talmy (2003), a spatial configuration that is otherwise conceived as static can be alternatively conceptualized as being sequentialized and include a path of fictive motion. One such fictive path is the trajectory of a person’s focus of attention shifting over a conceived scene. When a sentence can direct the hearer’s attention along such a path, it is amenable to the same types of windowing patterns as is possible with paths involving physical motion. One example would be sentences such as those in (3) containing the “across from” construction in which the focus of attention is directed along a path that traverses two reference points. A causal-chain event consists of a sequence of linked event or subevents where causality is associated with the boundaries between each subevent and its linked successor (Talmy, 2003). Causal-chain events exhibit causal-chain windowing of attention. A characteristic of these constructions is gapping of the entire medial portion of the sequence as in (4). Talmy (2003) suggests that the medial gapping of causal sequences reflects a cognitive structuring in which a particular state or event and its occurrence are conceptualized together in the foreground of attention while little or no attention is given to the intervening mediating stages. (5)(a) Departure phase windowing: (c) Departure phase plus return phase windowing:
Management of acquired language disorders associated with attentional impairment
Attention deficits in populations with acquired language disorders
Stroke
Aphasia
Right-hemisphere brain damage
Alzheimer disease
Traumatic brain injury
Specificity of attention interventions for language disorders
Language and attention
Grounding
Windowing
Event Frame
Definition
Type of Windowing
Path
Entirety of a path in motion
Path windowing
Causal chain
Sequence of linked events or sub-events where causality is associated with boundaries between each sub-event and linked successor
Causal-chain windowing
Cycle
Used to direct strongest attention to particular phase of iterating cycle; overall event is sequential but may have no clear beginning, middle, or end
Phase windowing
Participant interaction
Situational complex consisting of two parts: (1) a primary circumstance and (2) some individual interacting with that circumstance on two or more occasions; heightened attention placed on one or the other of the interactions fixes point for locating temporal perspective
Participant interaction windowing
Interrelationship
Conceptual complex comprised of parts that are not autonomous in themselves but intrinsically relative with respect to each other, i.e., the presence of one such part necessarily entails the presence of the other parts
Related posts:
The cognition of language and communication
The effects of aging on language and communication
Treating communication problems in individuals with disordered language
The role of memory in language and communication
Language and communication disorders associated with attentional deficits
Attention: architecture and process
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
