JSLHR Research Article Clinical Outcomes Following LanguageSpecific Attention Treatment Versus Direct Attention Training for Aphasia: A Comparative Effectiveness Study Richard K. Peach,a Katherine M. Beck,a Michelle Gorman,a and Christine Fishera Purpose: This study was conducted to examine the comparative effectiveness of 2 different approaches, 1 domain-specific and the other domain-general, to language and attention rehabilitation in participants with strokeinduced aphasia. The domain-specific treatment consisted of language-specific attention treatment (L-SAT), and the domain-general treatment consisted of direct attention training (DAT) using the computerized exercises included in Attention Process Training-3 (Sohlberg & Mateer, 2010). Method: Four individuals with mild–moderate aphasia participated in this study. A randomized controlled cross-over single-subject design was used to assess the effectiveness of the 2 treatments administered in this study. Treatment outcomes were evaluated in terms of participants’ task performance for each program, standardized language and attention measures, tests of functional abilities, and patient-reported outcomes. Results: Visual comparisons demonstrated linear improvements following L-SAT and variable patterns following DAT. Omnibus effect sizes were statistically significant for 9 of the 13 L-SAT tasks. The weighted standardized effect sizes for posttreatment changes following L-SAT ranged from small to large, with the exception of 1 task. The average group gain following DAT was 5%. The Western Aphasia Battery–Revised Aphasia Quotients (Kertesz, 2007) demonstrated reliable improvements for 3 of the 4 participants following L-SAT, whereas only 1 of the participants improved reliably following DAT. The margins of improvements in functional language were substantially larger following L-SAT than DAT. Performance on the Test of Everyday Attention improved significantly for 2 participants following L-SAT and for 1 participant following DAT on selected Test of Everyday Attention (Robertson, Ward, Ridgeway, & Nimmo-Smith, 1994) subtests. Patient-reported outcomes for communication and attention following treatment favored L-SAT compared to DAT. Conclusions: The results support the view that attention is allocated in ways that are particular to specific tasks rather than as a general resource that is allocated equivalently to all processing tasks. Domain-specific treatment for language deficits due to attentional impairment appears to be a suitable, if not preferable, approach for aphasia rehabilitation. Supplemental Material: https://doi.org/10.23641/asha. 8986427 I Pabón, 1964; Wepman, Jones, Bock, & Van Pelt, 1960). However, Wepman’s interests in the effects of language disturbances on a patient’s thought processes led him to argue for an indirect, more contextualized approach to language treatment (Wepman, 1972), whereas Schuell, who conceptualized aphasia as a multimodal, unidimensional language disturbance based on a primary deficit of auditory language, advocated for direct, intensive auditory stimulation as the preferred approach to aphasia rehabilitation (Schuell & Jenkins, 1959; see also Coelho, Sinotte, & Duffy, 2008). These methods have constituted some of the most frequently used approaches to aphasia therapy (e.g., Peach, 1993, 2001) while their application has been extended subsequently to more conversational contexts (e.g., Davis, n aphasia rehabilitation, the theoretical underpinnings of the disorder guide the methods that have been developed and tested for the treatment of the condition. For example, two contemporaries in early clinical aphasiology, Joseph Wepman and Hildred Schuell, argued for direct language stimulation to ameliorate the language deficits of people with aphasia (Schuell, Jenkins, & Jiménez- a Department of Communication Disorders and Sciences, Rush University Medical Center, Chicago, IL Correspondence to Richard K. Peach: richard_k_peach@rush.edu Editor-in-Chief: Sean Redmond Editor: Sarah Wallace Received December 19, 2018 Revision received February 17, 2019 Accepted April 1, 2019 https://doi.org/10.1044/2019_JSLHR-L-18-0504 Disclosure: The authors have declared that no competing interests existed at the time of publication. Journal of Speech, Language, and Hearing Research • Vol. 62 • 2785–2811 • August 2019 • Copyright © 2019 American Speech-Language-Hearing Association Downloaded from: https://pubs.asha.org Richard Peach on 08/16/2019, Terms of Use: https://pubs.asha.org/pubs/rights_and_permissions 2785 2005; Difrancesco, Pulvermüller, & Mohr, 2012; Peach & Wong, 2004). Some treatment approaches have been based on representational accounts of language impairment following aphasia, that is, how aphasia impairs the patient’s ability to construct syntactic or semantic representations (see, e.g., Grodzinsky, 1986; Rogers, Patterson, Jefferies, & Ralph, 2015). The corresponding treatment approaches to improve syntactic (e.g., Treating Underlying Forms [Thompson & Shapiro, 2005]; message-level treatment [Peach & Wong, 2004]) or semantic abilities (e.g., semantic feature analysis [Boyle, 2004; Boyle & Coelho, 1995; Peach & Reuter, 2010]; semantic complexity training [Kiran, 2007; Kiran & Thompson, 2003]; Verb Network Strengthening Treatment [Edmonds, Nadeau, & Kiran, 2009]) exhibit strong evidence for their efficacy and thus have widespread usage. Alternatively, a number of processing-based approaches have been used for aphasia treatment. Unlike representational approaches, these approaches view the disorder as a problem in the patient’s ability to access his or her preserved linguistic knowledge during language processing. Some examples of these methods include syntactic mapping treatment (Jacobs & Thompson, 2000; Rochon, Laird, Bose, & Scofield, 2005), semantic and phonological cueing treatments (Conroy, Sage, & Lambon Ralph, 2009; Wambaugh, Cameron, Kalinyak-Fliszar, Nessler, & Wright, 2004), semantic and phonological processing treatments (Kendall, Oelke, Brookshire, & Nadeau, 2015; Kiran, Sandberg, & Abbot, 2009; Peach, 1996; Raymer & Ellsworth, 2002), and contextual repetition priming (Martin, Fink, Renvall, & Laine, 2006; Renvall, Laine, & Martin, 2005). Recently, a number of treatment approaches have been based on resource theory. Unlike the approaches above, however, these approaches assume that the language impairment in aphasia results from difficulty in using cognitive resources (e.g., attention, memory) for the comprehension or production of language (Code, 2018). Representative models emphasize the effects of deficient attentional processing (Hula & McNeil, 2008; McNeil et al., 2004; Peach, Newhoff, & Rubin, 1993; Peach, Rubin, & Newhoff, 1994), auditory–verbal short-term memory (Kalinyak-Fliszar, Kohen, & Martin, 2011; Minkina, Rosenberg, KalinyakFliszar, & Martin, 2017; Peach, 1987; Salis, 2012), and working memory (Henderson, Kim, Kintz, Frisco, & Wright, 2017; Martin, Kohen, Kalinyak-Fliszar, Soveri, & Laine, 2012; Salis, Hwang, Howard, & Lallini, 2017; Wright & Shisler, 2005) on language processing. Improving these deficiencies, therefore, is the goal of methods that target these resources in treatment. Attention Treatment for Aphasia Of the cognitive resources that have been targeted in aphasia rehabilitation, attention, perhaps, has received the most consideration (Connor, Albert, Helm-Estabrooks, & Obler, 2000). Helm-Estabrooks, Connor, and Albert (2000) treated two patients with mixed nonfluent aphasia (PWA) using nonlinguistic sustained attention tasks followed by 2786 selective and alternating attention tasks, including symbol cancellation, trail making, repeated graphomotor patterns, auditory continuous performance, and card sorting, to improve auditory comprehension deficits. Improvements of 6 and 13 percentile points on the Boston Diagnostic Aphasia Examination (Goodglass & Kaplan, 1983) were observed immediately after the treatment program, with slight declines at approximately six to seven months posttreatment. Kohnert (2004) attempted to improve response speed and accuracy in a patient with transcortical motor aphasia by administering a series of nonverbal attention activities, including card sorting, written single-digit computations, and visual letter and number searches. Tasks from a high-level, computer-based training program were also administered (e.g., discrimination between target and nontarget icons in alternating quadrants of the computer screen). Gains were observed across all treatment tasks followed by modest improvements in some language tasks. Ramsberger (2005) also reported the outcome of a treatment study using a computer-based program to improve attention/executive functioning, linguistic processing, and conversational success for a patient with borderline fluent aphasia. The patient showed improvement on all of the training tasks, and the treatment appeared to generalize to two attention tests as well. Importantly, the patient’s ability to convey main ideas in conversation improved dramatically. A number of studies have administered either Attention Process Training–II (APT-II; Sohlberg, Johnson, Paule, Raskin, & Mateer, 2001) or APT-3 (Sohlberg & Mateer, 2010) to individuals with aphasia to address their language impairments. Both programs are commercially available and designed to address attentional deficits in individuals with acquired brain injury. Coelho (2005) administered APT-II to a 50-year-old woman with chronic aphasia (10 months postonset of stroke) to improve reading comprehension and reading rate. Reading comprehensions scores, based on responses to comprehension questions regarding magazine articles, improved from approximately 40%–60% to 83% accuracy following treatment. Reading rate (words per minute) remained variable throughout treatment. Posttreatment gains on two reading batteries provided support for improved reading outcomes secondary to this program. Sinotte and Coelho (2007) replicated Coelho’s study with a 60-year-old woman with mild anomic aphasia 6 months postonset of a left frontal hemorrhagic stroke. Following 16 sessions of APT-II over a 5-week period, few changes in reading comprehension accuracy or reading rate were observed. Murray, Keeton, and Karcher (2006) administered APT-II to a 57-year-old man with chronic, mild conduction aphasia and deficits in repetition, high-level auditory comprehension, language, and working memory. Mildly impaired attention for timed tasks was also observed. After more than 50 hr of training, the patient demonstrated faster response latencies to a paragraph listening task, but no other improvements in auditory comprehension. Murray et al. concluded that structured attention programs such as APT-II may not provide a viable approach to treating attention problems in patients with aphasia. Journal of Speech, Language, and Hearing Research • Vol. 62 • 2785–2811 • August 2019 Downloaded from: https://pubs.asha.org Richard Peach on 08/16/2019, Terms of Use: https://pubs.asha.org/pubs/rights_and_permissions Lee and Sohlberg (2013) administered APT-3 to treat reading comprehension in a single-subject design study with four participants demonstrating mild–moderate chronic aphasia. APT-3 combines direct attention training (DAT; as in APT-II) with metacognitive training. Despite the modest results obtained previously with this approach, the authors hypothesized that the addition of metacognitive instruction would improve resource allocation and, therefore, the reading comprehension of their participants with aphasia. Participants completed a maze reading task in which they select a target word from three choices in parentheses to replace missing words within each passage. The task is reported to be sensitive to the working memory and attentional demands required for reading comprehension. Two of the four participants demonstrated small improvements in maze reading scores during the intervention phase, although a rising trend during baseline for one of the participants made the source of the improvement uncertain. Small gains on a standardized reading test were also observed for these two participants. Lee, Sohlberg, Harn, Horner, and Cherney (2018) replicated the Lee and Sohlberg (2013) study, but with six participants presenting with mild aphasia, impairments in attention and/or working memory (as suggested by standardized testing), and complaints of reading difficulty. Also, the metacognitive facilitation component of APT-3 was expanded to include individualized instruction to promote generalization of self-monitoring to reading. The results suggested an effect between APT-3 and maze reading for three of the six participants, as indicated by visual analysis of the data plots and Tau-U effect sizes. However, participants who adopted metacognitive strategies performed better following treatment than those who did not. The authors concluded that APT-3 has the potential to improve reading in participants with aphasia. Yet the design of this study leaves open to question which component of the treatment, DAT or metacognitive strategy instruction, was the more effective for achieving the observed outcomes. Domain-Specific Versus Domain-General Attention Training All of the treatment studies reported above utilized domain-general approaches for aphasia rehabilitation, that is, treatment that targeted a variety of cognitive processes to improve language. The alternative to such an approach is a domain-specific approach, that is, one that focuses on and exposes participants to highly practical tasks (e.g., language) to increase generalization (Pella, Kendra, Hill, & Gouvier, 2008). The relative contributions of domain-general or domain-specific processes for language processing are a topic of continuing discussion. Recent reviews raise questions about the interaction between domain-specific regions for language processing and large-scale domain-general systems (Fedorenko & Thompson-Schill, 2014; Hartwigsen, 2018). How these domain-general processes contribute to language processing and “whether they are even necessary for…natural, everyday language” are uncertain (Campbell & Tyler, 2018, p. 132). According to Campbell and Tyler (2018), “domain-general regions may be active during these tasks because they are contributing to language processing or, more likely, because they are contributing to general attention and memory demands of the task” (p. 132). Despite the modest improvements reported in the previous treatment studies, it has been suggested that domain-general attentional approaches to language intervention result in poor outcomes (Rohling, Faust, Beverly, & Demakis, 2009). Other studies discourage the use of decontextualized computer-based attentional tasks, including APT, that address domain-general processes because of a lack of demonstrated impact on everyday attentional functions (Loetscher & Lincoln, 2013; Ponsford et al., 2014; Zickefoose, Hux, Brown, & Wulf, 2013). Alternatively, some researchers (see, e.g., Simonet, von Roten, Spierer, & Barral, 2019) have argued that treatment for attention deficits should require specific training for the activities to which the training is expected to generalize. 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. In a follow-up study, Sturm et al. (2003) administered the attention training programs to a participant group composed of individuals with both traumatic brain injury and vascular brain damage. A baseline phase was included to control for spontaneous recovery, and participants received treatment for only one of their impaired attention functions. The results demonstrated comparable training effects in both groups for specific training, not only for the intensity aspects of attention but also for divided attention. Specific attention training for alertness was also found to contribute to reorganization of the right-hemisphere functional network known to subserve the alertness domain in healthy subjects. Similar reorganization was not observed for patients with righthemisphere brain damage who received nonspecific (memory) training for alertness (Sturm et al., 2004). Park and colleagues (Park & Barbuto, 2005; Park & Ingles, 2001; Park, Proulx, & Towers, 1999) found no evidence to support APT for remediating impaired attention functions (e.g., sustained, selective, divided, and alternating attention) but did conclude that it results in learning of specific skills. This is consistent with recent results reported by Peers et al. (2018), who found that a group of 23 participants with brain damage improved significantly on a working memory outcome measure following working memory training but not on a series of selective attention tasks, whereas performance improved on attentional outcome measures following selective attention training but not following working memory training. Furthermore, Park and colleagues found that attention treatments that focus on learning or relearning of specific skills that are important Peach et al.: Language-Specific Attention Treatment for Aphasia Downloaded from: https://pubs.asha.org Richard Peach on 08/16/2019, Terms of Use: https://pubs.asha.org/pubs/rights_and_permissions 2787 to desired outcomes or behaviors having functional consequences resulted in significant improvement. Curran, Hussain, and Park (2001, as cited in Park & Barbuto, 2005) found that 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. It may also be, however, that the verbal descriptions directed the patients’ attention to the relevant environmental information and, in this way, focused attention to facilitate the development of mental representations for these actions. Many of these investigators have called for further exploration into the generalization of specific training effects to everyday functions relying on different aspects of attention. With regard to language training, this requires a consideration of the ways that specific linguistic processes recruit select attentional operations in the service of language. That is, treatment for language disorders due to attentional impairments might be better served by addressing the underlying attentional deficits that influence specific linguistic operations. That most previous treatment studies of language and attentional impairments have not done this might provide one explanation for the modest outcomes that have been observed. This requires an analysis of some of the ways that language operates as an attention director. For example, Crosson and colleagues (Crosson & Cohen, 2012; Crosson et al., 2007; Dotson et al., 2008) described two treatments manipulating attention and intention during confrontation naming that, unlike the previously reported studies, produced robust improvements in participants with moderate and severe word-finding deficits. Peach et al. (Peach, 2012; Peach, Schenk, Nathan, & Beck, 2018) developed the framework for a domain-specific program for the remediation of language disorders due to attentional impairment (language-specific attention treatment [L-SAT]). The approach is based on five principles: (a) train attentional focus and resource management for language, (b) increase attentional demands, (c) automatize attentional recruitment for language, (d) engage undamaged attentional mechanisms in the nondominant hemisphere, and (e) incorporate linguistic devices that require controlled attention. A preliminary analysis of the results obtained from four participants with aphasia for this treatment program provided evidence for improved language and attention but only for patients with aphasia having no greater than a moderate degree of attentional impairment (Peach, Nathan, & Beck, 2017). Based on these findings, the authors concluded that greater scrutiny was warranted regarding the effectiveness of L-SAT for stimulating attentional processing and language skills in people with aphasia. 2788 Purpose Given the varied opinions, approaches, and results that have been reported regarding attention rehabilitation, especially as it relates to attention-related language deficits following aphasia, this study was conducted to examine the comparative effectiveness of two diverging attentional approaches to language rehabilitation in participants with stroke-induced aphasia. One approach consisted of a domain-general approach, whereas the other utilized a domain-specific approach. We assessed the clinical outcomes associated with these treatments using multiple procedures at a variety of disability levels, including (a) statistical and visual analyses to measure the participants’ responses to each treatment, (b) standardized test scores to examine the effects of each treatment on the participants’ language and attention abilities, (c) functional tests of language and attention to assess the impact of each treatment on everyday abilities, and (d) patient-reported outcome scales to assess the participants’ own views of the impact of each treatment on their language and attentional abilities. Based on the modest results obtained previously using domain-general approaches to the treatment of language deficits following aphasia and the relatively stronger results in cognitive training observed following domain-specific methods, we hypothesized that the specific skills approach to improving attention for language would result in more desirable outcomes than a domain-general attentional approach for treating language deficits following aphasia. Method Participants This study was approved by the institutional review board of Rush University Medical Center (RUMC). A convenience sample consisting of four right-handed individuals (three men, one woman) with an average age of 62 years (SD = 17.0, range: 39–80 years) and an average education of 15 years (SD = 2.58, range: 12–18 years) participated in this study. They included two White and two Black participants who were recruited from RUMC and were paid for their participation in the study. All participants were monolingual speakers of English. All participants reported adequate hearing and had adequate near vision as determined by a brief vision screening (Schneider, 2002). No participant demonstrated a history of alcohol or substance abuse or psychiatric difficulties. All participants had a documented, previous diagnosis of aphasia from a single left-hemisphere stroke. Three of the participants had been enrolled previously in speechlanguage treatment; the treatment provided in this study was the first exposure to speech-language treatment for the remaining participant (P2). All participants had an Aphasia Quotient (AQ) below 93.8 on the Western Aphasia Battery–Revised (WAB-R; Kertesz, 2007), indicating the presence of aphasia. Aphasia severity was mild–moderate. Based on the profiles obtained from the WAB-R, all participants demonstrated fluent aphasia, although one (P3) Journal of Speech, Language, and Hearing Research • Vol. 62 • 2785–2811 • August 2019 Downloaded from: https://pubs.asha.org Richard Peach on 08/16/2019, Terms of Use: https://pubs.asha.org/pubs/rights_and_permissions Treatment All participants were exposed to two attention treatment programs: a domain-specific treatment (languagespecific treatment) and a domain-general treatment (DAT). Both programs followed written guidelines for their administration and were delivered largely by the same advanced graduate student in speech-language pathology in the same clinical setting at RUMC. The study was designed so that each participant received the same dosage and intensity of both treatment programs. All components of the study (initial assessment, Treatment Program A, post–Treatment Program A assessment, washout phase, Treatment Program B, post– Treatment Program B assessment, and final 1 month posttreatment assessment) were completed in approximately 25 weeks. Each treatment program consisted of 15 sessions (three 1-hr sessions per week for 5 weeks), for a total of 30 treatment sessions. Both treatment programs were discontinued after 15 sessions whether or not they had been fully completed. Domain-specific treatment. The domain-specific treatment consisted of L-SAT (Peach, 2012; Peach et al., 2017). In this program, the tasks and stimuli are language-based and impose increasing attentional demands on lexical and sentence processing. The goals of the treatment are to improve attentional focus and resource management for language and to automatize attentional recruitment for language. Administration of the program followed a written protocol that includes directions for task administration, the materials needed for each task, the task stimuli, the scoring procedures, the advancement criteria, and the task discontinuation procedures. The protocol is provided in Supplemental Material S1 for this article. The approach is based on a hierarchy of increasingly more complex language tasks (Peach et al., 2018). For the simplest task, picture-naming procedures are included to engage attentional mechanisms in the right cerebral hemisphere by moving the locus of stimulus presentation into left hemispace. Based on the evidence to date, this modification is appropriate for language-impaired individuals with left hemisphere brain damage (Coslett, 1999; Crosson et al., 2007; Dotson et al., 2008). A dual processing task (LaPointe & Erickson, 1991; see also Hula & McNeil, 2008; McNeil et al., 2004) is included to promote attention allocation for lexical processing. The sentence tasks exploit linguistic devices that are known to focus attention during language clearly was evolved from nonfluent aphasia and exhibited simplified sentence production and occasional agrammatism. The participants were, on average, 32 months postonset of their aphasia (SD = 35.7 months, range: 3–82 months). The participant characteristics, including their WAB-R aphasia classifications, are summarized in Table 1. Procedure Assessment A battery of tests and measurement scales were administered to the participants to characterize the nature of their acquired language and communication deficits, to establish and measure concomitant attention deficits, and to determine the outcomes of the two treatment programs investigated in this study. Assessment was performed at four time points: (a) prior to initiation of the first treatment program (Time 1), (b) after completion of the first treatment program and prior to initiation of a washout phase (Time 2), (c) after completion of the second treatment program (Time 3), and (d) 1 month after completion of the second treatment program (Time 4). The same battery of measures was given at each administration. Participants were not exposed to their previous assessment scores at any time. General language skills were assessed using the WAB-R (Kertesz, 2007) and the Object and Action Naming Battery (Druks & Masterson, 2000). Higher level language skills were assessed using the Discourse Comprehension Test (Brookshire & Nicholas, 1997). Functional language was assessed using the Communication Activities of Daily Living–Second Edition (CADL-2; Holland, Fratalli, & Fromm, 1999). Communicative participation was assessed using the ASHA Quality of Communication Life Scale (QCL; Paul et al., 2004). To assess attentional abilities, the Test of Everyday Attention (TEA; Robertson, Ward, Ridgeway, & NimmoSmith, 1994) was administered. Attention allo cation was assessed using the Stroop test (MacLeod, 1992). The Paced Auditory Serial Addition Test–3-Second Version (PASAT; Gronwall, 1977) was administered to assess processing speed. Finally, the Rating Scale of Attentional Behavior (RSAB; Ponsford & Kinsella, 1991) was used to assess ecological outcomes regarding attentional abilities following treatment. The participants’ language and attention standardized test scores at baseline are provided in Table 2. Table 1. Participant demographic and clinical data. Participant P1 P2 P3 P4 Age (years) Race Gender Education (years) Occupation MPOa WAB-R AQ Aphasia typeb 39 80 62 66 W B B W M F M M 18 14 16 12 Teacher Postal worker Environmental auditor Carpenter 25 3 82 10 71.7 86.6 86.0 66.0 Anomic Anomic Anomic Conduction Note. WAB-R AQ = Western Aphasia Battery–Revised Aphasia Quotient; W = White; M = male; B = Black; F = female. a Months postonset of aphasia. bWAB-R classification. Peach et al.: Language-Specific Attention Treatment for Aphasia Downloaded from: https://pubs.asha.org Richard Peach on 08/16/2019, Terms of Use: https://pubs.asha.org/pubs/rights_and_permissions 2789 Table 2. Pretreatment standardized test scores across participants. Test Western Aphasia Battery–Revised Aphasia Quotient (max. = 100) Information Fluency Auditory Verbal Comprehension Repetition Naming Object and Action Naming Battery (max. = 131) Discourse Comprehension Test (max. = 40) Communication Activities of Daily Living–Second Edition Test of Everyday Attention (Attention Factor) Map Search (Selective) Elevator Counting With Distraction (Selective) Visual Elevator (Switching/Mental Flexibility) Elevator Counting With Reversal (Working Memory) Telephone Search (Selective) Telephone Search While Counting (Sustained/Divided) Lottery (Sustained) Stroop Color–Word Score Paced Auditory Serial Addition Test–3-Second Version (max. = 60) P1 P2 P3 P4 71.1a 9 7 7.15 7 5.4 56a 29a 49b 86.6 9 9 8.7 9.8 6.8 111 35 78 86 9 6 9.5 9 8.5 110 30 94 66. 7. 7. 9.05 3.75 6.2 64. 27. 45. 5 26.6 2.4 37.2 2.4 0.4 2.4 20 0 37.2 16.2 9.5 16.2 74.5 9.5 0.4 18. 0. 16.2c 62.9 37.2 50 83.8 25.6 0.4 28d 20a 5 5 2.4 16.2 5 1.05 0.4 21 0 a Raw scores. bPercentile ranks. cMidpoint of range for percentile ranks. dT scores. processing. These include noncanonical sentence structure (Caplan & Waters, 1999; Myachykov & Posner, 2005; Shankweiler, Crain, Gorrell, & Tuller, 1989), focus structure (Cutler & Fodor, 1979), anaphoric reference (Myachykov & Posner, 2005), grounding elements (Langacker, 2008), and event windowing (Talmy, 2003). The tasks and phases that comprise the program, as well as descriptions of the treatment stimuli, cues, therapeutic operations, and advancement criteria, are provided in Appendix A. Additional stimuli for the Spatial Attention and Attention Allocation tasks are available from the first author. Recent work has established the validity of these tasks for directing attentional processing in healthy individuals and in persons with aphasia (Peach et al., 2018). The battery of language tasks that comprises the L-SAT program follows the expected pattern of engaging attentional processing (selective, sustained/divided, attentional switching), auditory–verbal working memory, and executive functioning (attentional control). The attentional correlates for each task for people with aphasia are provided in Table 3. Each task includes baseline and posttreatment probes to assess treatment generalization. The baseline stimuli for the first task, Spatial Attention, include 10 items from each of the three treatment phases and 10 untreated items. All of the baseline stimuli for the remaining tasks consist of untreated exemplars of the same forms requiring the same target responses as those required in the treatment tasks with the exception of the fourth task, Topicalization. Since the focus structure of these sentences provides the attentiondirecting element, the baseline stimuli for this task assess comprehension of embedded topics in nontopicalized sentences. The posttreatment probe tasks are identical to the baseline tasks. Discontinuation criteria were established for the baseline probes. A baseline score of 80% accuracy (or 2790 approximately 80% accuracy for probes with odd numbers of items) on the initial administration of the probe resulted in discontinuation of the task. In these instances, the next task was introduced immediately. When the participant did not meet the discontinuation criterion, three baseline probes were completed for each task. Completion criteria were also developed for each task to expedite the participants’ progression through the program and to achieve the goal of completing the program in 15 sessions. Criterion-level performance for each phase of all tasks was set at 80% accuracy over two consecutive trials. Once this level of performance was achieved on the last phase of a task, the posttreatment probe was administered, and the next task was introduced. If, however, a participant did not reach criterion on a phase of a task, the task was discontinued after three sessions. In these instances, the posttreatment probe for that task was administered, and the next task was introduced. Domain-general treatment. The computer exercises included in APT-3 (Sohlberg & Mateer, 2010; DAT) were used for the domain-general treatment. The treatment adhered to the instructions for administering these exercises provided in the program manual. The exercises are organized to address basic sustained attention and four domains of executive control: working memory, selective attention, suppression, and alternating attention. The program includes over 400 tasks; many have multiple versions with different stimuli. Generally, task selection is based on a patient’s accuracy and level of effort for any specific task. Tasks are designed to be hierarchical and can be modified by manipulating the speed and number of presented stimuli. Guidelines for task selection, as well as strategy selection and generalization activities, are provided with the program. Because there are no standardized rules for task selection, sampling may be required to identify the most Journal of Speech, Language, and Hearing Research • Vol. 62 • 2785–2811 • August 2019 Downloaded from: https://pubs.asha.org Richard Peach on 08/16/2019, Terms of Use: https://pubs.asha.org/pubs/rights_and_permissions Table 3. Language-specific attention tasks and attentional correlates for participants with aphasia. Language task Spatial Attention Attention Allocation Object Manipulation Topicalization Anaphora Nominal Grounding Clausal Grounding Windowing Attention Memory Sustained/divided Sustained/divided Selective Attentional switching Sustained/divided Attentional switching Selective Attentional switching Sustained/divided Selective Sustained/divided Sustained/divided Sustained/divided Central executive Auditory working Executive functioning (attentional control) Auditory working Executive functioning (attentional control) Auditory working Executive functioning (attentional control) Note. This table is based on results reported by Peach et al. (2018). appropriate level for a patient (Sohlberg & Mateer, 2010, p. 29). In order to increase the reliability of the sampling process for this study, we developed baseline procedures for identifying the most appropriate entry-level tasks for each participant. The procedures incorporated the program guideline suggesting that 80% accuracy with a medium to high level of effort indicates an appropriate task level (Sohlberg & Mateer, 2010, p. 29). The program addresses five components of attention: (a) auditory and visual sustained attention, (b) auditory and visual selective attention, (c) auditory working memory, (d) auditory and visual suppression, and (e) auditory and visual alternating attention. A hierarchy of the program tasks (from easiest to hardest) for each attention component was developed for this study to identify increasing levels of difficulty. The sampling began at the highest level of difficulty for the task that was at the approximate midpoint in the task hierarchy for a specific component. An iterating procedure was used until tasks were identified where the participants were performing at 60%–90% accuracy, with an effort rating of 5–10. Those tasks were used as the entry point for treatment within the appropriate attentional components. Once an entry point was determined, treatment began immediately within the same session on the selected tasks. If an entry-level task was not identified in the first session, baseline probing continued in a subsequent session until an appropriate, entry-level task was identified. The complete baseline procedures are described in Appendix B. Once entry-level tasks were identified for the five attentional components, treatment generally addressed all components in each session. The criterion for advancement to the next most difficult task for each component was 80% accuracy over two consecutive trials using different versions of the task where appropriate. Effort ratings were not considered for advancement. If a participant performed with 90% or greater accuracy on an ensuing task, that task was discontinued, and treatment advanced to the next most difficult task. This procedure was applied continuously to identify tasks at increasing levels of difficulty within the target range of performance. Treatment for a specific component was discontinued if all tasks for that component were completed. The treatment program was discontinued at the end of the 15th session. Design A randomized controlled cross-over single-subject design (Hart & Bagiella, 2012; Maxwell & Satake, 1997; Piantadosi, 2005), also known as an alternating treatments design (Barlow & Hayes, 1979; Barlow, Nock, & Hersen, 2009), was used to assess the effectiveness of the two treatments administered in this study. Each patient underwent both treatments. The participants were randomly assigned to each treatment order based on the sequence of their entry into the study. Two participants received the treatments in the order AB (language-specific treatment followed by DAT), while the other two participants received the treatments in the reverse BA order. The participants who were randomized to the AB order were designated Participants 1 and 2 for ease of tracking. The participants who were randomized to the BA order were designated Participants 3 and 4. At the end of the first treatment phase, all participants underwent a 1-month period of no treatment (washout period) to minimize any carryover effects. Analyses The following procedures were undertaken to analyze the data. For the language-specific treatment, effect sizes were calculated for the differences between the baseline phases and the intervention phases for each phase of each task as well as for the differences between the baseline and posttreatment phases of the study. Visual analyses were performed to assist in the interpretation of the statistical analyses. For the DAT, effect sizes could not be calculated as the administration procedures for the APT-3 computer exercises do not include collection of pretreatment or posttreatment probe data. Therefore, the participants’ performances were analyzed descriptively and visually. Trend analyses were performed to complement the visual assessment of the data. Peach et al.: Language-Specific Attention Treatment for Aphasia Downloaded from: https://pubs.asha.org Richard Peach on 08/16/2019, Terms of Use: https://pubs.asha.org/pubs/rights_and_permissions 2791 Pretreatment and posttreatment standardized test scores were analyzed and compared to assess the outcomes of the two treatment programs. Finally, the results obtained from patient-reported outcome scales were examined to determine the extent of any changes in ecological communication and attention outcomes. Reliability The L-SAT tasks have been shown to demonstrate good to excellent reliability in participants with aphasia (Peach et al., 2018). Test–retest reliability ranges from r = .84 to .98 with one exception, Attention Allocation, when using the scoring procedure for this task adopted in this study (see Peach et al., 2018, for discussion of this finding). The interclass correlation coefficients for inter- and intrarater reliability are .97 and 1.0, respectively. None of the administration procedures for the APT-3 exercises have been evaluated “empirically” (Sohlberg & Mateer, 2010, p. 31). The program is computer based, and guidelines for delivering the program are provided in the program manual. The score sheets that accompany each task allow for tracking of important performance indicators, for example, accuracy/time, error patterns, and level of effort. A case example is provided in the manual to increase adherence to the program procedures. Results Interventions L-SAT While all participants received 15 sessions of treatment, the baseline and completion criteria resulted in the participants progressing through the program at different rates. This resulted in their completing varying numbers of task trials and different numbers of phases and/or tasks. The participants completed between 49 and 76 (M = 62, SD = 11.9) task trials (including baseline and posttreatment probes) during the treatment period. The participants’ performances across all tasks are provided in Figures 1–4. Participants 2 and 3 met the discontinuation criteria for Spatial Attention on initial exposure at baseline. Therefore, only Participants 1 and 4 completed this task. Participants 1 and 3 met the discontinuation criteria for Attention Allocation on initial exposure, whereas Participants 2 and 4 met the completion criteria for this task during baseline testing. None of the participants, therefore, were treated with the Attention Allocation task (see Figures 1–4). Participant 2 also met the completion criteria for Object Manipulation during baseline testing. Participant 4 did not advance to Clausal Grounding, and no participant advanced sufficiently to be exposed to the last task, Windowing, although Participant 3 did complete baseline testing for this task. Otherwise, all participants were exposed to at least four sentence-processing tasks. Over all, Participant 1 was exposed to six tasks (one word and five sentence level), Participant 2 was exposed to four tasks (all 2792 sentence level), Participant 3 was exposed to five tasks (all sentence level), and Participant 4 was exposed to five tasks (one word and four sentence level). Some participants did not complete all phases for all of the tasks to which they were exposed. Visual analyses of the plots for the participants’ performance across tasks demonstrated post–baseline improvements during the intervention phases for these tasks. By and large, the data were characterized by linear trends suggesting steady increases in task performances until the later, more complex sentence-processing tasks. Performances on these tasks were more variable. To determine the effect sizes associated with languagespecific treatment, Tau-U was calculated (Brossart, Laird, & Armstrong, 2018; Lee & Cherney, 2018; Parker, Vannest, Davis, & Sauber, 2011) for each phase of all tasks undergoing treatment using the online calculator available for this procedure (Vannest, Parker, Gonen, & Adiguzel, 2011). Tau-U measures both baseline and intervention-phase trends. Baselines phases with trends can be corrected. The procedure also provides significance testing and confidence intervals to assist in interpreting the outcomes. Data sets with tau ≥ .40 during both the baseline and intervention phases trending in the same direction (Parker et al., 2011) underwent baseline correction. Using these criteria, the baselines for Participants 2 and 4 on Topicalization and for all participants on Clausal Grounding were corrected; none of the remaining tasks required baseline correction. Omnibus effect sizes (weighted tau) combined across all participants were calculated for the intervention phases of each task to assess the significance of the resulting changes in performance (see Table 4). The effect sizes were statistically significant ( p < .05) for all phases of all tasks, with the exception of four: Phase C of Anaphora ( p = .56), Phases A ( p = .44) and B ( p = 1.0) of Nominal Grounding (no participant advanced to Phase C of this task), and Phase A ( p = .18) of Clausal Grounding. These findings indicate that the participants made significant improvements within the time frames of this study in 69% of the 13 program phases in which they participated until reaching the highest and most advanced levels of the treatment program. Standardized effect sizes (Beeson & Robey, 2006; Busk & Serlin, 1992; see also Lee & Cherney, 2018) were calculated where possible for the tasks exposed to treatment to determine the differences between the baseline and posttreatment phases of L-SAT intervention (see Table 5). Effect sizes were not calculated for tasks with no/minimal variability in the pretreatment baseline scores. To aid in the interpretation of these results and in the absence of previous benchmarks for this treatment program, the observed effect sizes were assembled into four nonoverlapping groupings (based on the distribution of the observed scores) and categorized as follows: large effects (7.44 or greater), medium effects (2.77–3.18), small effects (0.66–1.44), and no effect (< 0.66). Of the 18 individual effect sizes that were calculated, three (56%) indicated large, positive treatment effects; Journal of Speech, Language, and Hearing Research • Vol. 62 • 2785–2811 • August 2019 Downloaded from: https://pubs.asha.org Richard Peach on 08/16/2019, Terms of Use: https://pubs.asha.org/pubs/rights_and_permissions Figure 1. Performance scores for Participant 1 for language-specific attention treatment across phases and tasks. Baseline probe data are plotted using triangles, intervention phase data are plotted using circles, and posttreatment probes are plotted using squares. Attn. = Attention; Obj. = Object; Nom. = Nominal. five (28%) indicated medium, positive effects; three (17%) indicated small, positive effects; and six (33%) indicated no treatment effect. In summary, positive treatment effects were observed for 67% of the L-SAT tasks. Participants 1 and 3 showed medium to large treatment gains on 4/6 and 2/3 tasks, respectively, and Participants 2 and 4 showed small to medium gains on 3/5 and 2/4 tasks, respectively. Thus, all four of the individual participants demonstrated some degree of treatment gain in response to L-SAT exposure. Weighted standardized effect scores (Beeson & Robey, 2006) were also calculated to assess group performance patterns for each task (see Table 5). Fitting the weighted scores to the effect size groupings described above, positive treatment effects were observed in response to five of the six L-SAT tasks. Anaphora generated a large effect, Spatial Attention and Object Manipulation produced medium effects, and Topicalization and Clausal Grounding yielded small effects. Only one task, Nominal Grounding, produced no effect. DAT As in L-SAT, the number of APT-3 computer tasks that were performed and the number of sessions to complete each task varied across participants (see Table 6). The mean number of tasks performed by each participant per cognitive domain was 2.3 (SD = 1.58, range: 1.3–3.8), and the mean number of sessions per task to completion was 3.49 (SD = 1.07, range: 2.63–5.25). The mean level of accuracy across tasks at entry into the program was 61% (SD = 0.19, range: 28%–79%), and the mean change across tasks following treatment was 5% (SD = 0.21, range: −4% to 11%). Participants 1 and 4 demonstrated mean increases in performance across all tasks. Participant 2 declined from baseline levels of performance on 89% of the tasks posttreatment, and Participant 3 declined on 67% of the tasks. The participants’ performances for each APT-3 computer task within the five cognitive domains are demonstrated in Figures 5–9. In lieu of effect sizes (because of the absence of baseline phases), trend lines were fit using Microsoft Excel to assist in the visual analyses of the data. Single data points were not included in these analyses. Trend lines were fit to achieve an R2 as close to 1 as possible. All data were found to conform to either a linear or polynomial trend. R2 for linear trends ranged between .9346 and 1. R2 for polynomial trends are shown in each graph. The data for all tasks, with the exception of those addressing Working Memory, were characterized by both linear and polynomial trends. The data for Working Memory were characterized exclusively by polynomial trends. In all domains, excluding visual selective attention, the data conformed to polynomial trends a majority of the time (see Table 6). In the case of visual selective attention, the data demonstrated a polynomial trend just under half (42%) of the time. These findings indicate that the participants’ Peach et al.: Language-Specific Attention Treatment for Aphasia Downloaded from: https://pubs.asha.org Richard Peach on 08/16/2019, Terms of Use: https://pubs.asha.org/pubs/rights_and_permissions 2793 Figure 2. Performance scores for Participant 2 for language-specific attention treatment across phases and tasks. Baseline probe data are plotted using triangles, intervention phase data are plotted using circles, and posttreatment probes are plotted using squares. Attn. = Attention; Obj. = Object; Nom. = Nominal. performance on the APT-3 computer tasks largely fluctuated from trial to trial and suggest a wide degree of variability in performance across trials on these tasks. Standardized Testing The difference scores comparing pretreatment and posttreatment scores for both intervention programs are provided in Table 7. Difference scores were calculated by subtracting the Time 1 test scores from the Time 2 test scores following the first intervention and by subtracting the Time 2 test scores from the Time 3 test scores following the second intervention. Language To assess the outcomes for each intervention program on standardized language testing, summary scores were calculated using the raw difference scores for the WAB-R, the Object and Action Naming Battery, and the Discourse Comprehension Test. The summary scores therefore indexed a wide array of language abilities, including basic skills, lexical retrieval for both nouns and verbs, and higher order auditory comprehension skills. Comparing these scores across participants revealed that Participant 1 made substantial improvements following both L-SAT and DAT, Participant 2 made greater gains following L-SAT than DAT, Participant 4 made greater gains following DAT than L-SAT, and Participant 3 made essentially no gains following either 2794 program. Changes in functional language, however, were notable following L-SAT, where the participants improved an average of 17.8 percentile points on the CADL-2, compared to an average decline of 1.25 percentile points on this test following DAT. Next, we investigated whether the WAB-R scores captured posttreatment changes in language abilities. Previously, increases in the WAB-R AQ greater than the estimated standard error of measurement for the WAB-R (i.e., 5 points) have been used to suggest clinically significant improvement (Elman & Bernstein-Ellis, 1999; Katz & Wertz, 1997). A problem with this approach concerns the assumption of equal measurement error across individuals, regardless of the severity of their aphasia (Hula, Donovan, Kendall, & Gonzales-Rothi, 2010). Instead, we analyzed the pretreatment–posttreatment changes on the WAB-R for each participant using the method reported by Hula et al. (2010). In this approach, WAB-R scores are fit to a Rasch model to create a more normal distribution of item scores and standard error scores that match the ability of each participant. According to Hula et al., the approach is a more valid method for indexing aphasia severity and change. Thus, the Rasch-based WAB-R scores for each participant were compared to determine whether the posttreatment results for each treatment approach were reliably different from one another. The WAB-R scores for Times 1 and 2 were compared using the Microsoft Excel WAB Rasch score sheet developed Journal of Speech, Language, and Hearing Research • Vol. 62 • 2785–2811 • August 2019 Downloaded from: https://pubs.asha.org Richard Peach on 08/16/2019, Terms of Use: https://pubs.asha.org/pubs/rights_and_permissions Figure 3. Performance scores for Participant 3 for language-specific attention treatment across phases and tasks. Baseline probe data are plotted using triangles, intervention phase data are plotted using circles, and posttreatment probes are plotted using squares. Attn. = Attention; Obj. = Object; Nom. = Nominal. for this analysis (W. Hula, personal communication, May 25, 2018) to assess changes after the first intervention and for Times 2 and 3 to assess changes after the second intervention. The procedure uses one-tailed probabilities that assume improvements in WAB-R scores following treatment. Therefore, only those change scores that demonstrated increases in performance were considered for this analysis. To assess the significance of the individual changes in WAB-R scores in this small sample, α was set at .15 (Crawford, Sommerville, & Robertson, 1997; Wechsler, 1981). Significantly improved WAB-R scores were found for Participant 1 ( p = .09), Participant 2 ( p = .08), and Participant 4 (p = .09) following L-SAT. Only Participant 1 showed significant improvement on the WAB-R following DAT ( p = .01). Using the percentile ranks obtained from the Hula et al. (2010) sample, Participant 1 improved from the 26th to the 32nd percentile, Participant 2 improved from the 57th to the 71st percentile, and Participant 4 improved from the 29th to the 34th percentile following L-SAT. Following DAT, Participant 1 improved from the 32nd to the 46th percentile. Finally, maintenance of posttreatment scores 1 month after the conclusion of treatment was assessed by comparing the WAB-R scores obtained at Times 3 and 4. No significant changes were observed for Participants 1 and 2 following the completion of DAT or Participant 4 following L-SAT. However, significant improvement was observed for Participant 3 following L-SAT (p = .07). Interestingly, it should be noted that Participant 3 was the only individual who did not show significant posttreatment improvements to either treatment at the termination of each program. Attention Difference scores were calculated between the pretreatment and posttreatment administrations of the TEA, the Stroop test, and the PASAT following L-SAT and DAT (see Table 7). The mean percentile differences for the seven subtests of the TEA were determined for each participant using the midpoint of the percentile ranges for each scaled score provided by Robertson et al. (1994). Overall, the only improvement on the TEA was observed for Participant 1 following L-SAT. Otherwise, the average change for the other three participants was 1 percentile point or less. When inspecting the changes for individual subtests by participant, the largest changes (greater than 5 percentile points) occurred following L-SAT for Map Search (Participant 1), Elevator Counting With Distraction (Participants 1, 3, and 4), Visual Elevator (Participant 1), and Elevator Counting With Reversal (Participant 1). Large changes on Telephone Search (Participants 2 and 4) and Telephone Search While Counting (Participants 1 and 3) were observed following both L-SAT and DAT. To investigate these individual differences further, we analyzed each participant’s TEA scaled scores using Peach et al.: Language-Specific Attention Treatment for Aphasia Downloaded from: https://pubs.asha.org Richard Peach on 08/16/2019, Terms of Use: https://pubs.asha.org/pubs/rights_and_permissions 2795 Figure 4. Performance scores for Participant 4 for language-specific attention treatment across phases and tasks. Baseline probe data are plotted using triangles, intervention phase data are plotted using circles, and posttreatment probes are plotted using squares. Attn. = Attention; Obj. = Object; Nom. = Nominal. the method recommended by Crawford et al. (1997). In this approach, participants’ performances are assessed with regard to the pattern of individual strengths and weaknesses across the seven subtests of the battery to identify reliable differences among subtests. To achieve this, discrepancy scores estimating specific levels of significance are calculated using particular subtest scores and the mean subtest score for the battery. A computer program is available to perform these calculations (Crawford, 2018). We applied this procedure to assess the significance of changes across TEA administrations following both treatment programs. Our interest was in identifying those subtests Table 4. Omnibus effect sizes calculated across participants for language-specific attention treatment tasks by intervention phases. Task Spatial Attention Attention Allocation Object Manipulation Topicalization Anaphora Nominal Grounding Clausal Grounding Windowing Phasea Participantsb Tauc z p Confidence intervald A B C A B A B — A B C A B C A B A–C 2 1 1 0 0 3 3 4 4 4 3 4 3 0 4 2 0 0.84 1.0 1.0 — — 1.0 −0.88 0.59 1.0 0.56 0.21 0.18 0 — 0.39 0.76 — 2.83 2.24 1.96 — — 3.49 −2.88 2.57 3.62 2.34 0.58 0.77 0 — 1.69 2.15 — 0.005 0.025 0.049 — — 0.001 0.004 0.010 0.000 0.019 0.561 0.439 1.0 — 0.089 0.032 — [0.2564, 1] [0.264, 1] [0.162, 1] — — [0.4377, 1] [−1, −0.2813] [0.1406, 1] [0.4583, 1] [0.0913, 1] [−0.4926, 0.9075] [−0.2761, 0.6363] [−0.6160, 0.6160] — [−0.0600, 0.8346] [0.0658, 1] — Note. Em dashes indicate there are no data for these phases. a Phase of intervention for tasks with multiple phases (see Appendix A). bNumber of participants completing designated phase. cOmnibus effect size for participants completing phase (in cases with one participant, tau equals effect size for contrast between participant’s baseline and intervention phases). dConfidence intervals reported at 95% level except for Spatial Attention B and C, which is reported at 90% level. 2796 Journal of Speech, Language, and Hearing Research • Vol. 62 • 2785–2811 • August 2019 Downloaded from: https://pubs.asha.org Richard Peach on 08/16/2019, Terms of Use: https://pubs.asha.org/pubs/rights_and_permissions Table 5. Standardized effect sizes (d ) for pretreatment–posttreatment differences in language-specific attention treatment performance by tasks and participants as well as across participants. Task P1 P2 P3 P4 Spatial Attention 2.77 NA NA 3.00 Attention Allocation NA NA NA NA Object Manipulation 7.44 −1.20 * 0.80 Topicalization 2.89 3.18 * −2.02 Anaphora 15.17 −0.50 12.86 * Nominal Grounding 0.10 0.66 −2.92 −0.59 Clausal Grounding −0.43 1.44 2.28 ** Counting With Reversal ( p < .05), and Participant 3 improved significantly on Elevator Counting With Distraction (p < .01). Following DAT, the probability of reliable change for Participant 1 increased for both Visual Elevator (p < .05) and Elevator Counting With Reversal (p < .01). No other significant changes in TEA subtest performance were observed. The pattern of performance for Participants 1 and 4 on the Stroop test was characterized by improvement after administration of either treatment initially followed by additional but smaller improvement after exposure to the second treatment. For Participant 1, the change score following L-SAT resulted in his improving from the impaired range to the borderline range (Psychometric Conversion Table, n.d.). Additional increases following DAT improved his performance into the low average range of performance. For Participant 4, exposure to DAT improved his performance from the profound range to the impaired range, and subsequent exposure to L-SAT improved his performance to the borderline range. Participant 2 did not improve on this test in response to L-SAT but showed slight improvement to the severe range following DAT. Participant 3 did not improve on this test in response to either treatment. Finally, Participant 1 demonstrated a large change on the PASAT in response to L-SAT, whereas Participants 3 and 4 demonstrated large changes on this test in response to DAT. The change observed in Participant 1 following L-SAT was sufficient to boost his score into the low normal range for this test (Rao, Leo, Bernardin, & Unverzagt, 1991). As was the case with the Stroop test, Participants 1 and 4 showed smaller, additional improvements when exposed to DAT and L-SAT, respectively. Participant 3 showed no further improvement following L-SAT. Participant 2 demonstrated no changes in response to either treatment. Weighted d 2.89 — 4.12 1.35 9.18 −0.72 1.10 Note. Em dash indicates there is no effect size. NA = participant met task criteria at baseline; * = unable to calculate effect size due to no or minimal variability in baseline; ** = program completed, no final posttreatment probe. that improved significantly, relative to the participants’ individual abilities, following treatment. To do this, we first identified subtests on which a participant performed significantly better than average at baseline and within the range of normal performance. Participants 1 and 4 performed significantly above average on Telephone Search ( p < .01), and Participant 3 performed significantly above average on Elevator Counting With Reversal ( p < .10). As these subtests represented individual strengths, they were eliminated from further consideration when assessing changes following treatment. There were no significant discrepancies between any of the subtests performed by Participant 2. Next, we examined whether participants, given their own levels of ability, improved significantly on any of the remaining subtests for the two posttreatment administrations that followed. Following L-SAT, Participant 1 improved significantly on Visual Elevator ( p < .15) and Elevator Table 6. Grouped performance data for direct attention training. Cognitive domain Input modality No. of tasks performed Sessions/ task Initial percent accuracy Mean percent change Sustained Attention Auditory 3.3 (0.96) 2.75 (1.50) Visual 1.8 (1.50) 5.25 (0.50) Auditory 3.8 (2.22) 3.99 (0.37) Visual 3.3 (3.86) 2.88 (1.65) Working Memory Auditory 1.8 (1.50) 4.38 (1.49) Suppression Auditory 1.5 (1.00) 3.63 (1.49) Visual 2.8 (2.22) 2.79 (0.98) Auditory 1.3 (0.50) 3.13 (1.18) Visual 1.3 (0.50) 2.63 (0.48) 67 (0.19) (39–79) 65 (0.22) (37–83) 68 (0.12) (55–81) 79 (0.18) (61–96) 28 (0.21) (0–50) 47 (0.30) (13–82) 72 (0.14) (62–92) 50 (0.27) (11–72) 76 (0.10) (63–86) 11 (0.31) (−17 to 52) 8 (0.18) (−18 to 26) 1 (0.15) (−20 to 16) −4 (0.22) (−36 to 16) 11 (0.21) (−12 to 30) 6 (0.10) (−7 to 14) 10 (0.22) (−5 to 43) 6 (0.25) (−16 to 33) 0 (0.20) (−24 to 23) Selective Attention Alternating Attention Trend (%) Linear Polynomial 10 90. 20 80. 25 75. 57.1 42.9 0 100. 20 80. 37.5 62.5 20 80. 40 60. Peach et al.: Language-Specific Attention Treatment for Aphasia Downloaded from: https://pubs.asha.org Richard Peach on 08/16/2019, Terms of Use: https://pubs.asha.org/pubs/rights_and_permissions 2797 Figure 5. Performance scores across participants for Attention Process Training–3 alternating attention tasks. Auditory tasks are plotted using circles, and visual tasks are plotted using triangles. Patient-Reported Outcomes In order to assess the ecological outcomes following these treatments, participants completed the QCL (Paul et al., 2004) and the RSAB (Ponsford & Kinsella, 1991; see Table 8). The QCL consists of 17 statements scored 1–5 (5 = highest), assessing the impact of the participant’s aphasia on his or her communication and participation in life situations. Higher scores indicate less communicative disruption. The RSAB describes 12 everyday behaviors that are associated with attentional impairments and the frequency with which they affect the participant. Items are scored on a 5-point scale (0 = not a problem, 5 = is a problem all the time). Lower scores indicate less impairment. The mean QCL difference scores were compared by participant following administration of the two treatment programs (see Table 8). Participant 3 responded at ceiling for all items (despite his significant aphasia) on the posttreatment administrations for both programs. Because of the lack of variability and questionable reliability of his responses, he was removed from this analysis. The group QCL difference score following L-SAT (M = 0.58, SD = 0.05) exceeded that following DAT (M = 0.24, SD = 0.78), although this difference was not statistically significant (z = −0.54, p = .59, Wilcoxon signed-ranks test). Individually, two of the three participants rated their communication higher following L-SAT than DAT. 2798 The total RSAB scores following each treatment were also compared. Again, Participant 3 rated himself as having virtually no attentional problems on the first two administrations of the RSAB but followed these ratings with one indicating substantial attentional problems on the last administration. As with the QCL above, this pattern suggested unreliable responses, and he was excluded from this analysis. The mean decline in attentional problems following L-SAT was 9.3 points (SD = 17.0). The mean decline following DAT was 2.7 points (SD = 14.7). The probability of this difference representing a reliable change was 89% (z = −1.60, p = .11, Wilcoxon signed-ranks test). All three participants rated their attentional abilities more favorably following L-SAT than DAT. Discussion In this study, we compared the effectiveness of L-SAT, a domain-specific approach, to DAT, a domain-general approach, for treating the language and attentional deficits associated with aphasia. Based on recent reports and recommendations in the extant literature, we hypothesized that the domain-specific approach to improving attention for language would result in more desirable outcomes than a domain-general approach to treating attention for language in people with aphasia. Although both treatment programs produced variable improvements in both Journal of Speech, Language, and Hearing Research • Vol. 62 • 2785–2811 • August 2019 Downloaded from: https://pubs.asha.org Richard Peach on 08/16/2019, Terms of Use: https://pubs.asha.org/pubs/rights_and_permissions Figure 6. Performance scores for Participants 1, 2, 3, and 4 for Attention Process Training–3 selective attention tasks. Auditory tasks are plotted using circles, and visual tasks are plotted using triangles. language and attention across participants, the preponderance of the findings in this study provides support for our hypothesis. We investigated the outcomes for each approach with regard to these participants’ language impairment, their functional language, and their personal reports regarding their language and attentional abilities before and after treatment. Visual comparisons of the participants’ performances on the treatment tasks in each program demonstrated steadier, linear improvements following L-SAT versus more variable and unpredictable patterns of performance following DAT. This latter finding is consistent with the increased between-sessions and within-session variability that has been observed in people with aphasia in response to domain-general attention training (Villard & Kiran, 2018). Tau-U effect sizes were statistically significant for nine of the 13 L-SAT task phases that were administered suggesting reliable improvements. The three task phases for which improvement was not demonstrated were among the most complex tasks in the program. Also, the weighted effect sizes for posttreatment changes following L-SAT ranged conservatively from small to large with the exception of one task (Nominal Grounding). Using the widespread metric Peach et al.: Language-Specific Attention Treatment for Aphasia Downloaded from: https://pubs.asha.org Richard Peach on 08/16/2019, Terms of Use: https://pubs.asha.org/pubs/rights_and_permissions 2799 Figure 7. Performance scores across participants for Attention Process Training–3 attention suppression tasks. Auditory tasks are plotted using circles, and visual tasks are plotted using triangles. suggested by Cohen (1988), all of these effect sizes would be considered large. Alternatively, the group average gain following DAT was 5%. Individually, while two of the participants showed mean increases overall, one declined from baseline levels on two thirds of the tasks, and another declined from baseline on almost 90% of the tasks. The results of standardized testing also favor the domain-specific approach. Simple sums of change scores for the three standardized language tests suggested that one or the other program (or both) may result in language improvements. However, the analysis of the WAB-R AQs, taking person ability into consideration, demonstrated reliable improvements for three of the four participants following L-SAT, while only one participant improved reliably on the WAB-R following DAT. Also, the margins of improvements in functional language were substantially larger following L-SAT than DAT. These results provide further evidence that either program may produce some language improvements in people with aphasia but that a more robust outcome may be achieved from the domain-specific approach both in terms of reducing language impairment and improving functional language. Some may assert that a language-based approach to treatment (i.e., L-SAT) would be expected to produce better outcomes in people with aphasia than a non–language-based approach (i.e., DAT) and that the improvements described 2800 above are simply an outcome of this difference. However, the test of this concern would be whether a domain-general attentional approach results in improvements to attentional processing that is not observed following a domain-specific language-based approach. While both L-SAT and DAT are known to engage attentional processing, the posttreatment results suggested that L-SAT resulted in attentional improvements, even in purported nonlanguage tasks, that exceeded those of the purely domain-general approach. These results provide further support for the attentional requirements of the L-SAT tasks (Peach et al., 2018). Following treatment, only one participant showed an overall net improvement in performance on the TEA, and this was in response to L-SAT. In addition, the individual patterns demonstrated that large improvements for four of the six TEA subtests occurred following L-SAT, while improvements on the remaining two tasks occurred following both L-SAT and DAT. When assessing TEA performance relative to the participants’ own strengths and weaknesses, one participant improved significantly on two TEA subtests following L-SAT and one on one TEA subtest. The only significant individual improvement following DAT occurred for one participant on two TEA subtests that were exposed previously to L-SAT. Equivocal results were demonstrated following these treatments with regard to the Stroop test (attention Journal of Speech, Language, and Hearing Research • Vol. 62 • 2785–2811 • August 2019 Downloaded from: https://pubs.asha.org Richard Peach on 08/16/2019, Terms of Use: https://pubs.asha.org/pubs/rights_and_permissions Figure 8. Performance scores across participants for Attention Process Training–3 sustained attention tasks. Auditory tasks are plotted using circles, and visual tasks are plotted using triangles. allocation) and the PASAT (speed of processing). For the Stroop test, two participants showed initial improvement following the first treatment (L-SAT in the first instance, DAT in the second) and continued improvement following the alternate treatment. Another showed slight improvement on this test following DAT, and the fourth participant showed no improvement to either treatment. For the PASAT, two participants improved substantially following DAT and one following L-SAT. Continued smaller improvements were noted following exposure to the alternate treatment in two cases. One of the participants who improved following DAT showed no further improvement to L-SAT, and one participant showed no improvements to either treatment. Greater improvements in speed of processing might be expected following DAT as some of the tasks in this program include speeded presentations while none of the L-SAT tasks address this component of attention. Finally, the patterns for communication and attentional improvements in functional language and patient perceptions following treatment favored L-SAT when compared to DAT. Thus, L-SAT appears to provide a more ecologically valid approach to treating language and attention given its emphasis on specific skills. Apparently, this emphasis is perceived as translating to everyday language and attentional behaviors more effectively than domain-general tasks. In this way, L-SAT provides (a) a working framework for individual training that is tailored to the participant’s particular needs and expectations and (b) a more ecological perspective to theoretical models of cognition and their application (Moreau & Conway, 2014). Given the promising results obtained with L-SAT, despite none of the participants finishing the program as constructed, a question arises whether even more robust outcomes could be achieved when PWA are exposed to the complete program. The treatment schedule used with these participants was constrained by the parameters of this study. Nonetheless, the schedule emulated a somewhat typical, nonintensive course of contemporary aphasia treatment (Cherney, Patterson, & Raymer, 2011). As a result, no participant was exposed to the most complex tasks in the L-SAT program. It is intriguing then, given the beneficial effects associated with higher complexity and language processing (Kiran & Thompson, 2003; Rubin, Newhoff, Peach, & Shapiro, 1996; Thompson, Ballard, & Shapiro, 1998; Thompson & Shapiro, 2007; Thompson, Shapiro, Kiran, & Sobecks, 2003), to speculate whether an extended treatment schedule allowing for exposure to these tasks might have resulted in further improvements. This will be an important issue for future investigations regarding L-SAT effectiveness. Peach et al.: Language-Specific Attention Treatment for Aphasia Downloaded from: https://pubs.asha.org Richard Peach on 08/16/2019, Terms of Use: https://pubs.asha.org/pubs/rights_and_permissions 2801 Figure 9. Performance scores across participants for Attention Process Training–3 working memory tasks. & Zimba, 2000; McNeil, Hula, & Sung, 2011; McNeil, Odell, & Tseng, 1991; Tseng, McNeil, & Milenkovic, 1993). Descriptions that attempt to separate attention from language processing (e.g., language and cognition; American SpeechLanguage-Hearing Association, 2016; Villard & Kiran, 2017) or that refer to the attentional contributions to language These results demonstrate how attention and language bind together in the service of word and sentence processing to form a cohesive mechanism for language processing. Thus, resource perspectives regarding aphasia are well positioned to explain how attentional impairments in aphasia account for language deficits (Granier, Robin, Shapiro, Peach, Table 7. Standardized test difference scores posttreatment for language-specific attention treatment (L-SAT) and direct attention training (DAT). P1 Test L-SAT Western Aphasia Battery–Revised Aphasia Quotient Object and Action Naming Battery Discourse Comprehension Test Summary score Communication Activities of Daily Living–Second Edition Test of Everyday Attention (Attention Factor) Map Search (Selective) Elevator Counting With Distraction (Selective) Visual Elevator (Switching/Mental Flexibility) Elevator Counting With Reversal (Working Memory) Telephone Search (Selective) Telephone Search While Counting (Sustained/Divided) Lottery (Sustained) Mean difference Stroop Color–Word Score Paced Auditory Serial Addition Test–3-Second Version P2 P3 P4 DAT L-SAT DAT DAT L-SAT DAT L-SAT 5.6 25a −5a 25.6 30b 5.7 7 14 26.7 11 4.2 −1 2 5.2 8 0.4 −4 −1 −4.6 −8 1.3 0 −1 0.3 −4 −5.4 6 −1 −0.4 −12 −3.6 22 4 22.4 −4 10.3 2. 2. 14.3 45. 9.4b 20.9 53.4 45.0 −9.3 49 0.0 24.0 6c 17a 0 −20.9 0 2.6 −37.4 −49.0 0.65 −14.85 5 8 0 0 0 0 5 0 0 0.7 1 0 4.5 0 0 0 −7.05 8.4 0.7 0.92 3 0 0 −9.4 2.6 −11.6 0 25.2 −2.0 0.7 −1 20 0 33.8 −4.0 −20.6 −1.4 −25.2 0 −2.5 −4 −11 −11.6 −6.8 0 0 16.1 0 0 −0.3 9 10 0. 6.8 0. 0. 0. 0. 0. 1.0 2. 3. a a Raw score difference. bPercentile rank difference. cT-score difference. 2802 Journal of Speech, Language, and Hearing Research • Vol. 62 • 2785–2811 • August 2019 Downloaded from: https://pubs.asha.org Richard Peach on 08/16/2019, Terms of Use: https://pubs.asha.org/pubs/rights_and_permissions Table 8. Participant responses on patient-reported outcome measures by treatment program and assessment time (T). P1 Measure T1 ASHA Quality of Communication 3.6 Life Scale (maximum = 5.0) Rating Scale of Attentional Behavior 22 (0 = highest, 48 = lowest) P2 T2 L-SAT T3 DAT 4.1 14 3.9 24 T1 P4 T2 L-SAT T3 DAT T1 T2 DAT T3 L-SAT 3.7 40 P3 4.3 13 4.7 21 5.0 5.0 0 1 5.0 37 T1 3.4 17 T2 DAT T3 L-SAT 2.8 26 3.4 24 Note. L-SAT = language-specific attention treatment; DAT = direct attention training. processing as somehow being “extralinguistic” (e.g., Murray & Mayer, 2017) miss these integral relationships. The results of this and previous work provide evidence for how attention is recruited to process language and for how attentional processing can be manipulated and strengthened through the performance of language tasks that require increasingly complex attentional abilities. The results also support the view that attention is allocated in ways that are particular to specific tasks rather than as a general resource that is allocated equivalently to all processing tasks (Park & Ingles, 2001; Sturm et al., 1997, 2003). Limitations While the randomized controlled cross-over design offers a valid design for comparing the effectiveness of two different treatments (see, e.g., Cherney, Kaye, & van Vuuren, 2014; Crosson et al., 2007; Wilson, Emslie, Quirk, & Evans, 2001), an important limitation concerns possible confounds from carryover effects between treatments (also known as multiple treatment interference). That is, treatment effects observed from a first treatment may continue to influence the outcomes of a second treatment (Barlow & Hayes, 1979; Hart & Bagiella, 2012; Piantadosi, 2005). However, some recommended methods to address these problems (Barlow & Hayes, 1979) were used in this study. For example, the treatment periods in this study were widely spaced because of the washout period so carryover effects, if any, would be expected to be transient. Also, the counterbalanced order of treatments are thought to minimize carryover effects across the treatment conditions. Finally, only one treatment was administered per session in each treatment phase. Carryover effects have been thought to be most pronounced when the treatments are alternated rapidly and within a short time. Nonetheless, despite the fact that the study followed the typical structure of a randomized controlled cross-over design, testing prior to the beginning of the second treatment phase might have provided information to assess the extent of carryover effects, if any. The convenience sample for this study included four PWA with varied profiles. Three of the participants presented with fluent aphasia from the onset of their condition, while one evolved to a profile of fluent aphasia from nonfluent aphasia. Three presented with chronic aphasia, while one (Participant 2) was in the subacute stage of her recovery. While the time postonset for this latter participant might raise concerns regarding spontaneous recovery, the pattern of substantially greater improvement following her first treatment when compared with the second should diminish this concern. Two presented with mild aphasia, while two presented with moderate aphasia. The attentional profiles across participants also varied. While the data for the outcome measures as a whole appear to indicate that the group of participants responded more favorably to L-SAT than to DAT, the individual profiles were mixed. For example, the participants completed different numbers of L-SAT tasks, and the levels of improvement across tasks varied from small to large. Also, the results of standardized language and attention testing for each participant were equivocal, with different participants demonstrating different outcomes regarding the effectiveness of treatment and the program that produced the superior results. Of note, Participant 3 demonstrated the poorest language outcomes to either treatment as suggested by the summary score for his language testing and his CADL-2 score. As this individual was the participant who evolved from a nonfluent to a fluent aphasia, these results may suggest that a participant with nonfluent aphasia (or those who evolve from nonfluent aphasia) may not be as good a candidate for L-SAT as a participant with fluent aphasia. Caution is advised, however, in drawing such a conclusion from a single individual. While the overall outcomes regarding L-SAT are encouraging, further study ideally employing a stronger design with a larger pool of participants would provide additional data to test the comparative effectiveness of the two approaches to intervention and to determine the profile(s) of participants with aphasia who enjoy the greatest benefit from this type of treatment. Also, none of the participants in the study completed the full L-SAT program. This was a result of the constraints on the schedule for L-SAT that arose from the design of the study (i.e., pre- and posttreatment testing, two interventions, a washout period between interventions, and posttreatment follow-up). In a typical clinical environment, such constraints on L-SAT delivery might be loosened so that all of the tasks could be administered. Inasmuch as this includes the most complex tasks of the program, participants might derive even greater benefit from the program than that which was observed here. However, since no data relating to the participants’ performance on these tasks was available, it is unknown what contributions these tasks might make, if any, to the outcomes associated with this program. For these reasons, it will be important that the Peach et al.: Language-Specific Attention Treatment for Aphasia Downloaded from: https://pubs.asha.org Richard Peach on 08/16/2019, Terms of Use: https://pubs.asha.org/pubs/rights_and_permissions 2803 full L-SAT program be administered in any future studies of L-SAT to fully gauge the effectiveness of this program as written or whether modifications may be in order. Finally, all of the participants met criterion for the Attention Allocation task after initial exposure or during baseline testing. Thus, the treatment portion of this task was not administered to any of the participants. Recall that the task requires participants to identify a target word from a list of monosyllabic words presented auditorily while completing a card sorting task simultaneously. The score for the task is the number of target words identified correctly (LaPointe & Erickson, 1991). But given that the directions for this task as administered emphasize the identification of the target words while sorting, it is very likely that the participants allocated their attention primarily to the word lists while focusing attention secondarily to the card sorting task. This then could result in a high rate of accuracy for target word identification with perhaps a high degree of inaccuracy for the sorting task. This may have contributed to the scores observed for these participants on this task. In order for this task to index attention allocation more reliably, scoring should include both the number of target words identified correctly and the number of card sorts performed correctly. Going forward, we would recommend that the scoring for this task include performance on both of these components. Based on the data reported by LaPointe and Erickson (1991) for their PWA, we would recommend using concurrent scores of 80% accuracy on the word identification task and 50% accuracy on the card sorting task twice consecutively as the criteria for advancement. This change has now been incorporated into the program protocol. Conclusion L-SAT provides a domain-specific approach to attention-related language rehabilitation following aphasia that results in relatively stronger clinical outcomes than those observed with a domain-general approach. The program is based on known principles of learning, is consistent with current practice guidelines, and appears to be ecologically valid in that the treatment tasks are largely equivalent to the skills to which they are expected to generalize. While further investigation of this program is warranted, the results suggest that L-SAT is an effective tool for managing language disability due to aphasia. Acknowledgment This work was supported by faculty research grants from the College of Health Sciences, Rush University, Chicago, Illinois. References American Speech-Language-Hearing Association. (2016). Scope of practice in speech-language pathology [Scope of practice]. Retrieved from http://www.asha.org/policy/ Barlow, D. H., & Hayes, S. C. (1979). Alternating treatments design: One strategy for comparing the effects of two treatments 2804 in a single subject. Journal of Applied Behavior Analysis, 12(2), 199–210. Barlow, D. H., Nock, M., & Hersen, M. (2009). Single case experimental designs: Strategies for studying behavior for change (3rd ed.). Boston, MA: Pearson. Beeson, P. M., & Robey, R. R. (2006). Evaluating single-subject treatment research: Lessons learned from the aphasia literature. Neuropsychology Review, 16(4), 161–169. Boyle, M. (2004). Semantic feature analysis treatment for anomia in two fluent aphasia syndromes. American Journal of SpeechLanguage Pathology, 13, 236–249. Boyle, M., & Coelho, C. A. (1995). Application of semantic feature analysis as a treatment for aphasic dysnomia. American Journal of Speech-Language Pathology, 4(4), 94–98. Brookshire, R. H., & Nicholas, L. E. (1997). Discourse Comprehension Test: Test manual. Minneapolis, MN: BRK Publishers. Brossart, D. F., Laird, V. C., & Armstrong, T. W. (2018). Interpreting Kendall’s tau and Tau-U for single-case experimental designs. Cogent Psychology, 5(1), Article ID 1518687. Busk, P. L., & Serlin, R. C. (1992). Meta-analysis for single case research. In T. R. Kratochwill & J. R. Levin (Eds.), Singlecase research design and analysis: New directions for psychology and education (pp. 187–212). Hillsdale, NJ: Erlbaum. Campbell, K. L., & Tyler, L. K. (2018). Language-related domainspecific and domain-general systems in the human brain. Current Opinion in Behavioral Sciences, 21, 132–137. Caplan, D., & Waters, G. S. (1999). Verbal working memory and sentence comprehension. Behavioral and Brain Sciences, 22(1), 77–94. Cherney, L. R., Kaye, R. C., & van Vuuren, S. (2014). Acquisition and maintenance of scripts in aphasia: A comparison of two cuing conditions. American Journal of Speech-Language Pathology, 23(2), S343–S360. Cherney, L. R., Patterson, J. P., & Raymer, A. M. (2011). Intensity of aphasia: Evidence and efficacy. Current Neurology and Neuroscience Reports, 11, 560–569. Code, C. (2018). Aphasia is not a disorder of language: Discuss [Abstract]. Aphasiology, 32(Supp. 1), 52–53. Coelho, C. (2005). Direct attention training as a treatment for reading impairment in mild aphasia. Aphasiology, 19(3), 275–283. Coelho, C., Sinotte, M., & Duffy, J. (2008). Schuell’s stimulation approach to rehabilitation. In R. Chapey (Ed.), Language intervention strategies in aphasia and related neurogenic communication disorders (5th ed., pp. 403–449). Philadelphia, PA: Lippincott Williams & Wilkins. Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Erlbaum. Connor, L. T., Albert, M. L., Helm-Estabrooks, N., & Obler, L. K. (2000). Attentional modulation of language performance. Brain and Language, 71(1), 52–55. Conroy, P., Sage, K., & Lambon Ralph, M. A. (2009). The effects of decreasing and increasing cue therapy on improving naming speed and accuracy for verbs and nouns in aphasia. Aphasiology, 23(6), 707–730. Coslett, H. B. (1999). Spatial influences on motor and language function. Neuropsychologia, 37(6), 695–706. Crawford, J. R. (2018). Computer programs for research and practice in neuropsychology/clinical psychology. Retrieved from https:// homepages.abdn.ac.uk/j.crawford/pages/dept/psychom.htm Crawford, J. R., Sommerville, J., & Robertson, I. H. (1997). Assessing the reliability and abnormality of subtest differences on the Test of Everyday Attention. British Journal of Clinical Psychology, 36(4), 609–617. Crosson, B., & Cohen, M. L. (2012). Language and communication disorders associated with attentional deficits. In R. K. Journal of Speech, Language, and Hearing Research • Vol. 62 • 2785–2811 • August 2019 Downloaded from: https://pubs.asha.org Richard Peach on 08/16/2019, Terms of Use: https://pubs.asha.org/pubs/rights_and_permissions Peach & L. P. Shapiro (Eds.), Cognition and acquired language disorders: An information processing approach (pp. 167–182). St. Louis, MO: Elsevier Mosby. Crosson, B., Fabrizio, K. S., Singletary, F., Cato, M. A., Wierenga, C. E., Parkinson, R. B., . . . Rothi, L. J. (2007). Treatment of naming in nonfluent aphasia through manipulation of intention and attention: A phase 1 comparison of two novel treatments. Journal of the International Neuropsychological Society, 13, 582–594. Cutler, A., & Fodor, J. A. (1979). Semantic focus and sentence comprehension. Cognition, 7(1), 49–59. Davis, G. (2005). PACE revisited. Aphasiology, 19(1), 21–38. Difrancesco, S., Pulvermüller, F., & Mohr, B. (2012). Intensive Language–Action Therapy (ILAT): The methods. Aphasiology, 26(11), 1317–1351. Dotson, V. M., Singletary, F., Fuller, R., Koehler, S., Moore, A. B., Gonzalez Rothi, L. J., & Crosson, B. (2008). Treatment of wordfinding deficits in fluent aphasia through the manipulation of spatial attention: Preliminary finding. Aphasiology, 22(1), 103–113. Druks, J., & Masterson, J. (2000). An Object and Action Naming Battery. London, United Kingdom: Psychology Press. Edmonds, L. A., Nadeau, S. E., & Kiran, S. (2009). Effect of Verb Network Strengthening Treatment (VNeST) on lexical retrieval of content words in sentences in persons with aphasia. Aphasiology, 23(3), 402–424. Elman, R. J., & Bernstein-Ellis, E. (1999). The efficacy of group communication treatment in adults with chronic aphasia. Journal of Speech, Language, and Hearing Research, 42(2), 411–419. Fedorenko, E., & Thompson-Schill, S. L. (2014). Reworking the language network. Trends in Cognitive Sciences, 18(3), 120–126. Goodglass, H., & Kaplan, E. (1983). The assessment of aphasia and related disorders (2nd ed.). Philadelphia, PA: Lea & Febiger. Granier, J. P., Robin, D. A., Shapiro, L. P., Peach, R. K., & Zimba, L. D. (2000). Measuring processing load during sentence comprehension: Visuomotor tracking. Aphasiology, 14(5), 501–513. Grodzinsky, Y. (1986). Language deficits and the theory of syntax. Brain and Language, 27(1), 135–159. Gronwall, D. M. (1977). Paced auditory serial-addition task: A measure of recovery from concussion. Perceptual and Motor Skills, 44(2), 367–373. Hart, T., & Bagiella, E. (2012). Design and implementation of clinical trials in rehabilitation research. Archives of Physical Medicine and Rehabilitation, 93(8), S117–S126. Hartwigsen, G. (2018). Flexible redistribution in cognitive networks. Trends in Cognitive Sciences, 22(8), 687–698. Helm-Estabrooks, N., Connor, L. T., & Albert, M. L. (2000). Treating attention to improve auditory comprehension in aphasia. Brain and Language, 74, 469–472. Henderson, A., Kim, H., Kintz, S., Frisco, N., & Wright, H. H. (2017). Working memory in aphasia: Considering discourse processing and treatment implications. Seminars in Speech and Language, 38(1), 40–51. Holland, A. L., Frattali, C., & Fromm, D. (1999). Communication Activities of Daily Living–Second Edition: CADL-2. Austin, TX: Pro-Ed. Hula, W. D., Donovan, N. J., Kendall, D. L., & Gonzalez-Rothi, L. J. (2010). Item response theory analysis of the Western Aphasia Battery. Aphasiology, 24(11), 1326–1341. Hula, W. D., & McNeil, M. R. (2008). Models of attention and dual-task performance as explanatory constructs in aphasia. Seminars in Speech and Language, 29(3), 169–187. Jacobs, B. J., & Thompson, C. K. (2000). Cross-modal generalization effects of training noncanonical sentence comprehension and production in agrammatic aphasia. Journal of Speech, Language, and Hearing Research, 43(1), 5–20. Kalinyak-Fliszar, M., Kohen, F., & Martin, N. (2011). Remediation of language processing in aphasia: Improving activation and maintenance of linguistic representations in (verbal) shortterm memory. Aphasiology, 25(10), 1095–1131. Katz, R. C., & Wertz, R. T. (1997). The efficacy of computerprovided reading treatment for chronic aphasic adults. Journal of Speech, Language, and Hearing Research, 40(3), 493–507. Kendall, D. L., Oelke, M., Brookshire, C. E., & Nadeau, S. E. (2015). The influence of phonomotor treatment on word retrieval abilities in 26 individuals with chronic aphasia: An open trial. Journal of Speech, Language, and Hearing Research, 58(3), 798–812. Kertesz, A. (2007). Western Aphasia Battery–Revised. San Antonio, TX: The Psychological Corporation. Kiran, S. (2007). Complexity in the treatment of naming deficits. American Journal of Speech-Language Pathology, 16(1), 18–29. Kiran, S., Sandberg, C., & Abbott, K. (2009). Treatment for lexical retrieval using abstract and concrete words in persons with aphasia: Effect of complexity. Aphasiology, 23(7–8), 835–853. Kiran, S., & Thompson, C. (2003). The role of semantic complexity in treatment of naming deficits: Training semantic categories in fluent aphasia by controlling exemplar typicality. Journal of Speech, Language, and Hearing Research, 46, 773–787. Kohnert, K. (2004). Cognitive and cognate-based treatments for bilingual aphasia: A case study. Brain and Language, 91(3), 294–302. https://doi.org/10.1016/j.bandl.2004.04.001 Langacker, R. L. (2008). Complex sentences. Cognitive grammar: A basic introduction (pp. 406–453). New York, NY: Oxford University Press. LaPointe, L. L., & Erickson, R. J. (1991). Auditory vigilance during divided task attention in aphasic individuals. Aphasiology, 5(6), 511–520. Lee, J. B., & Cherney, L. R. (2018). Tau-U: A quantitative approach for analysis of single-case experimental data in aphasia. American Journal of Speech-Language Pathology, 27(1S), 495–503. Lee, J. B., & Sohlberg, M. M. (2013). Evaluation of attention training and metacognitive facilitation to improve reading comprehension in aphasia. American Journal of Speech-Language Pathology, 22(2), S318–S333. Lee, J. B., Sohlberg, M. M., Harn, B., Horner, R., & Cherney, L. R. (2018). Attention Process Training–3 to improve reading comprehension in mild aphasia: A single-case experimental design study. Neuropsychological Rehabilitation. https://doi.org/10. 1080/09602011.2018.1477683 Loetscher, T., & Lincoln, N. B. (2013). Cognitive rehabilitation for attention deficits following stroke. Cochrane Database of Systematic Reviews, 31(5). https://doi.org//10.1002/14651858. CD002842.pub2 MacLeod, C. M. (1992). The Stroop task: The “gold standard” of attentional measures. Journal of Experimental Psychology: General, 121(1), 12. Martin, N., Fink, R. B., Renvall, K., & Laine, M. (2006). Effectiveness of contextual repetition priming treatments for anomia depends on intact access to semantics. Journal of the International Neuropsychological Society, 12(6), 853–866. Martin, N., Kohen, F., Kalinyak-Fliszar, M., Soveri, A., & Laine, M. (2012). Effects of working memory load on processing of Peach et al.: Language-Specific Attention Treatment for Aphasia Downloaded from: https://pubs.asha.org Richard Peach on 08/16/2019, Terms of Use: https://pubs.asha.org/pubs/rights_and_permissions 2805 sounds and meanings of words in aphasia. Aphasiology, 26(3–4), 462–493. Maxwell, D. L., & Satake, E. (1997). Research and statistical methods in communication disorders. Baltimore, MD: Williams & Wilkins. McNeil, M. R., Doyle, P., Hula, W., Rubinsky, H., Fossett, T., & Matthews, C. (2004). Using resource allocation theory and dual-task methods to increase the sensitivity of assessment in aphasia. Aphasiology, 18(5), 521–542. McNeil, M. R., Hula, W., & Sung, J. E. (2011). The role of memory and attention in aphasic language performance. In J. Guendouzi, F. Loncke, & M. J. Williams (Eds.), The handbook of psycholinguistic and cognitive processes (pp. 551–577). New York, NY: Psychology Press. McNeil, M. R., Odell, K., & Tseng, C. H. (1991). Toward the integration of resource allocation into a general theory of aphasia. Clinical Aphasiology, 20, 21–39. Minkina, I., Rosenberg, S., Kalinyak-Fliszar, M., & Martin, N. (2017). Short-term memory and aphasia: From theory to treatment. Seminars in Speech and Language, 38(1), 17–28. Moreau, D., & Conway, A. R. A. (2014). The case for an ecological approach to cognitive training. Trends in Cognitive Sciences, 18(7), 334–336. https://doi.org/10.1016/j.tics.2014.03.009 Murray, L., Keeton, R. J., & Karcher, L. (2006). Treating attention in mild aphasia: Evaluation of Attention Process Training–II. Journal of Communication Disorders, 39, 37–61. Murray, L., & Mayer, J. (2017). Aphasia and related neurogenic communication disorders. In I. Papathanasiou & P. Coppens (Eds.), Extralinguistic cognitive consideration in aphasia management (2nd ed., pp. 129–149). Burlington, MA: Jones & Bartlett Learning. Myachykov, A., & Posner, M. I. (2005). Attention in language. In L. Itti, G. Rees, & J. Tsotsos (Eds.), Neurobiology of attention (pp. 324–329). Waltham, MA: Elsevier. Park, N. W., & Barbuto, E. (2005). Treating attention impairments: Review with a particular focus on naturalistic action rehabilitation. In P. W. Halligan & D. T. Wade (Eds.), The effectiveness of rehabilitation for cognitive deficits. (pp. 81–90). New York, NY: Oxford University Press. Park, N. W., & Ingles, J. L. (2001). Effectiveness of attention rehabilitation after an acquired brain injury: A meta-analysis. Neuropsychology, 15(2), 199. Park, N. W., Proulx, G., & Towers, W. M. (1999). Evaluation of the attention process training programme. Neuropsychological Rehabilitation, 9(2), 135–154. Parker, R. I., Vannest, K. J., Davis, J. L., & Sauber, S. B. (2011). Combining nonoverlap and trend for single-case research: Tau-U. Behavior Therapy, 42(2), 284–299. Paul, D., Frattali, C., Holland, A., Thompson, C., Caperton, C., & Slater, S. (2004). ASHA Quality of Communication Life Scale (QCL): Manual. Rockville, MD: American Speech-LanguageHearing Association. Peach, R. K. (1987). A short-term memory treatment approach to the repetition deficit in conduction aphasia. Clinical Aphasiology, 17, 35–45. Peach, R. K. (1993). Clinical intervention for aphasia in the United States of America. In A. L. Holland & M. M. Forbes (Eds.), Aphasia treatment: World perspectives (pp. 335–369). San Diego, CA: Singular. Peach, R. K. (1996). Treatment for aphasic phonological output planning deficits. Clinical Aphasiology, 24, 109–120. Peach, R. K. (2001). Further thoughts regarding management of acute aphasia following stroke. American Journal of SpeechLanguage Pathology, 10, 29–36. 2806 Peach, R. K. (2012). Management of acquired language disorders associated with attentional impairment. In R. K. Peach & L. P. Shapiro (Eds.), Cognition and acquired language disorders: An information processing approach (pp. 241–274). St. Louis, MO: Elsevier Mosby. Peach, R. K., Nathan, M. R., & Beck, K. M. (2017). Languagespecific attention treatment for aphasia: Description and preliminary findings. Seminars in Speech and Language, 38(1), 5–16. https://doi.org/10.1055/s-0036-1597260 Peach, R. K., Newhoff, M., & Rubin, S. S. (1993). Attention in aphasia as revealed by event-related potentials: A preliminary investigation. Clinical Aphasiology, 21, 323–333. Peach, R. K., & Reuter, K. A. (2010). A discourse-based approach to semantic feature analysis for the treatment of aphasic word retrieval failures. Aphasiology, 24(9), 971–990. Peach, R. K., Rubin, S. S., & Newhoff, M. (1994). A topographic event-related potential analysis of the attention deficit for auditory processing in aphasia. Clinical Aphasiology, 22, 81–96. Peach, R. K., Schenk, K. A., Nathan, M. R., & Beck, K. M. (2018). Construct validity, external validity, and reliability for a battery of language-specific attention tasks. Aphasiology, 32(6), 618–645. Peach, R. K., & Wong, P. C. M. (2004). Integrating the message level into treatment for agrammatism using story retelling. Aphasiology, 18(5/6/7), 429–441. Peers, P. V., Astle, D. E., Duncan, J., Murphy, F. C., Hampshire, A., Das, T., & Manly, T. (2018). Dissociable effects of attention vs working memory training on cognitive performance and everyday functioning following fronto-parietal strokes. Neuropsychological Rehabilitation. Advance online publication. https://doi. org/10.1080/09602011.2018.1554534 Pella, R. D., Kendra, K., Hill, B. D., & Gouvier, W. D. (2008). Cognitive rehabilitation with brain-damaged patients. In A. M. Horton & D. Wedding (Eds.), The neuropsychology handbook (3rd ed., pp. 419–467). New York, NY: Springer Publishing Company. Piantadosi, S. (2005). Clinical trials: A methodologic perspective. Crossover designs (2nd ed., pp. 515–527). Hoboken, NJ: WileyInterscience. Ponsford, J., Bayley, M., Wiseman-Hakes, C., Togher, L., Velikonja, D., McIntyre, A., . . . INCOG Expert Panel. (2014). INCOG recommendations for management of cognition following traumatic brain injury, Part II: Attention and information processing speed. Journal of Head Trauma Rehabilitation, 29(4), 321–337. Ponsford, J., & Kinsella, G. (1991). The use of a rating scale of attentional behaviour. Neuropsychological Rehabilitation, 1(4), 241–257. Psychometric Conversion Table. (n.d.). Retrieved from https:// www.ritenour.k12.mo.us/cms/lib011/MO01910124/Centricity/ Domain/69/Psychometric_Conversion_Table.pdf Ramsberger, G. (2005). Achieving conversational success in aphasia by focusing on non-linguistic cognitive skills: A potentially promising new approach. Aphasiology, 19(10–11), 1066–1073. Rao, S. M., Leo, G. J., Bernardin, L., & Unverzagt, F. (1991). Cognitive dysfunction in multiple sclerosis. I. Frequency, patterns, and prediction. Neurology, 41(5), 685–691. Raymer, A., & Ellsworth, T. (2002). Response to contrasting verb retrieval treatments: A case study. Aphasiology, 16(10/11), 1031–1045. Renvall, K., Laine, M., & Martin, N. (2005). Contextual priming in semantic anomia: A case study. Brain and Language, 95(2), 327–341. Journal of Speech, Language, and Hearing Research • Vol. 62 • 2785–2811 • August 2019 Downloaded from: https://pubs.asha.org Richard Peach on 08/16/2019, Terms of Use: https://pubs.asha.org/pubs/rights_and_permissions Robertson, I. H., Ward, T., Ridgeway, V., & Nimmo-Smith, I. (1994). The Test of Everyday Attention. Suffolk, United Kingdom: Thames Valley Test Company. Rochon, E., Laird, L., Bose, A., & Scofield, J. (2005). Mapping therapy for sentence production impairments in nonfluent aphasia. Neuropsychological Rehabilitation, 15(1), 1–36. Rogers, T. T., Patterson, K., Jefferies, E., & Ralph, M. A. L. (2015). Disorders of representation and control in semantic cognition: Effects of familiarity, typicality, and specificity. Neuropsychologia, 76, 220–239. Rohling, M. L., Faust, M. E., Beverly, B., & Demakis, G. (2009). Effectiveness of cognitive rehabilitation following acquired brain injury: A meta-analytic re-examination of Cicerone et al.’s (2000, 2005) systematic reviews. Neuropsychology, 23(1), 20–39. Rubin, S. S., Newhoff, M., Peach, R. K., & Shapiro, L. P. (1996). Electrophysiological indices of lexical processing: The effects of verb complexity and age. Journal of Speech and Hearing Research, 39(5), 1071–1080. Salis, C. (2012). Short-term memory treatment: Patterns of learning and generalisation to sentence comprehension in a person with aphasia. Neuropsychological Rehabilitation, 22(3), 428–448. Salis, C., Hwang, F., Howard, D., & Lallini, N. (2017). Short-term and working memory treatments for improving sentence comprehension in aphasia: A review and a replication study. Seminars in Speech and Language, 38(1), 29–39. Schneider, J. (2002, May). Block letter eye chart. Retrieved from http://eyetestnow.com/images/snellen_block_letter_eye_chart_ eye_exam_test.pdf Schuell, H., & Jenkins, J. J. (1959). The nature of language deficit in aphasia. Psychological Review, 66(1), 45–67. Schuell, H., Jenkins, J. J., & Jimenez-Pabon, E. (1964). Aphasia in adults. New York, NY: Harper & Row. Shankweiler, D., Crain, S., Gorrell, P., & Tuller, B. (1989). Reception of language in Broca’s aphasia. Language and Cognitive Processes, 4(1), 1–33. Simonet, M., von Roten, F. C., Spierer, L., & Barral, J. (2019). Executive control training does not generalize, even when associated with plastic changes in domain-general prefrontal areas. NeuroImage, 197, 457–469. Sinotte, M. P., & Coelho, C. A. (2007). Attention training for reading impairment in mild aphasia: A follow-up study. Neurorehabilitation, 22, 303–310. Sohlberg, M. M., Johnson, L., Paule, L., Raskin, S. A., & Mateer, C. A. (2001). Attention Process Training–II: A program to address attentional deficits for persons with mild cognitive dysfunction (2nd ed.). Wake Forest, NC: Lash & Associates. Sohlberg, M. M., & Mateer, C. A. (2010). Attention Process Training APT-3: A direct attention training program for persons with acquired brain injury. Wake Forest, NC: Lash & Associates. Sturm, W., Fimm, B., Cantagallo, A., Cremel, N., North, P., Passadori, A., . . . Leclercq, M. (2003). Specific computerized attention training in stroke and traumatic brain-injured patients. Zeitschrift Für Neuropsychologie, 14(4), 283–292. Sturm, W., Longoni, F., Weis, S., Specht, K., Herzog, H., Vohn, R., . . . Willmes, K. (2004). Functional reorganisation in patients with right hemisphere stroke after training of alertness: A longitudinal PET and fMRI study in eight cases. Neuropsychologia, 42, 434–450. https://doi.org/10.1016/j.neuropsychologia.2003.09. 001 Sturm, W., Willmes, K., Orgass, B., & Hartje, W. (1997). Do specific attention deficits need specific training? Neuropsychological Rehabilitation, 7(2), 81–103. Talmy, L. (2003). Toward a cognitive semantics, Volume I: Concept structuring systems. The windowing of attention in language (pp. 258–309). Cambridge, MA: MIT Press. Thompson, C. K., Ballard, K. J., & Shapiro, L. P. (1998). The role of syntactic complexity in training wh-movement structures in agrammatic aphasia: Optimal order for promoting generalization. Journal of the International Neuropsychological Society, 4(6), 661–674. Thompson, C. K., & Shapiro, L. P. (2005). Treating agrammatic aphasia within a linguistic framework: Treatment of underlying forms. Aphasiology, 19(10–11), 1021–1036. Thompson, C. K., & Shapiro, L. P. (2007). Complexity in treatment of syntactic deficits. American Journal of Speech-Language Pathology, 16(1), 30–42. Thompson, C. K., Shapiro, L. P., Kiran, S., & Sobecks, J. (2003). The role of syntactic complexity in treatment of sentence deficits in agrammatic aphasia: The Complexity Account of Treatment Efficacy (CATE). Journal of Speech, Language, and Hearing Research, 46, 591–607. Tseng, C., McNeil, M. R., & Milenkovic, P. (1993). An investigation of attention allocation deficits in aphasia. Brain and Language, 45(2), 276–296. Vannest, K., Parker, R., Gonen, O. & Adiguzel, T. (2011). Single case research: Web based calculators for SCR analysis. Retrieved from http://www.singlecaseresearch.org Villard, S., & Kiran, S. (2017). To what extent does attention underlie language in aphasia? Aphasiology, 31(10), 1226–1245. Villard, S., & Kiran, S. (2018). Between-session and within-session intra-individual variability in attention in aphasia. Neuropsychologia, 109, 95–106. Wambaugh, J., Cameron, R., Kalinyak-Fliszar, M., Nessler, C., & Wright, S. (2004). Retrieval of action names in aphasia: Effects of two cueing treatments. Aphasiology, 18(11), 979–1004. Wechsler, D. (1981). Manual for the Wechsler Adult Intelligence Scale–Revised. New York, NY: The Psychological Corporation. Wepman, J. M. (1972). Aphasia therapy: A new look. Journal of Speech and Hearing Disorders, 37(2), 203–214. Wepman, J. M., Jones, L. V., Bock, R. D., & Van Pelt, D. (1960). Studies in aphasia: Background and theoretical formulations. Journal of Speech and Hearing Disorders, 25(4), 323–332. Wilson, B. A., Emslie, H. C., Quirk, K., & Evans, J. J. (2001). Reducing everyday memory and planning problems by means of a paging system: A randomised control crossover study. Journal of Neurology, Neurosurgery & Psychiatry, 70(4), 477–482. Wright, H. H., & Shisler, R. J. (2005). Working memory in aphasia: Theory, measures, and clinical implications. American Journal of Speech-Language Pathology, 14, 107–118. Zickefoose, S., Hux, K., Brown, J., & Wulf, K. (2013). Let the games begin: A preliminary study using Attention Process Training–3 and Lumosity™ brain games to remediate attention deficits following traumatic brain injury. Brain Injury, 27(6), 707–716. https://doi.org/10.3109/02699052.2013.775484 Peach et al.: Language-Specific Attention Treatment for Aphasia Downloaded from: https://pubs.asha.org Richard Peach on 08/16/2019, Terms of Use: https://pubs.asha.org/pubs/rights_and_permissions 2807 2808 Appendix A ( p. 1 of 3) L-SAT Treatment Program Journal of Speech, Language, and Hearing Research • Vol. 62 • 2785–2811 • August 2019 Attention manipulation Lexical processing Spatial Attention Attention Allocation Sentence processing Object Manipulation Topicalization Phase No. of items A 40 B 40 C 40 A 3 Trials, 25 targets, 75 foils/trial B 3 Trials, 25 targets, 75 foils/trial A 20 B 14 20 Sentences, 40 questions Treatment stimuli, tasks, cues, therapeutic operations, and advancement criteria Present simultaneous visual (fireworks) and auditory (4-s tone) warning signals and line drawings alternately on a computer monitor placed 45° to the left of the patient’s midline. Lexical items randomized for varying levels of frequency of occurrence (high, medium, low). Patient names each item on confrontation (20 s maximum). If patient provides incorrect name, correct response is modeled and patient repeats it. No more than three models per item. Criterion for advancement to next phase: 32 or more correct responses over two consecutive trials. Present auditory (4-s tone) warning signal and line drawings alternately on a computer monitor placed 45° to the left of the patient’s midline. Lexical items randomized for varying levels of frequency of occurrence (high, medium, low). Patient names each item on confrontation (20 s maximum). If patient provides incorrect name, correct response is modeled and patient repeats it. No more than three models per item. Criterion for advancement to next phase: 32 or more correct responses over two consecutive trials. Present auditory (0.5-s tone) warning signal and line drawings alternately on a computer monitor placed 45° to the left of the patient’s midline. Lexical items randomized for varying levels of frequency of occurrence (high, medium, low). Patient names each item on confrontation (20 s maximum). If patient provides incorrect name, correct response is modeled and patient repeats it. No more than three models per item. Criterion for completion of task: 32 or more correct responses over two consecutive trials. Patient listens to monosyllabic word lists and raises hand when target word is heard (auditory attention). If patient demonstrates 20 or more correct responses on any trial with no cueing, advance to Phase B. If patient demonstrates fewer than 20 correct responses, repeat trial with cueing and then proceed to next trial. Cues consist of auditory attention strategies (i.e., self-monitor, rehearse, listen and anticipate, repeat) alone or in combination with written cues (printed target word). As accuracy increases in combined cueing condition, visual cue is withdrawn systematically (100%, 75%, 50%, or 25% of trials). Patient listens to monosyllabic word lists and raises hand when target word is heard while simultaneously completing card sorting task (dual task attention). If patient demonstrates 20 or more correct responses on two consecutive trials, discontinue task. If patient demonstrates fewer than 20 correct responses, repeat trial with cueing and then proceed to next trial. Cues consist of auditory attention strategies (i.e., self-monitor, rehearse, listen and anticipate, repeat) alone or in combination with written cues (printed target word). As accuracy increases in combined cueing condition, visual cue is withdrawn systematically (100%, 75%, 50%, or 25% of trials). Patient provided cues for syntactic subjects of sentences and then manipulates objects to demonstrate correct actions in semantically reversible sentences. Ten active sentences and 10 passive sentences with two-place verb argument structure presented in random order. For incorrect responses, sentence is repeated, clinician models the correct relationship, and then patient performs the correct action. Criterion for advancement to next phase: 16 or more accurate responses over two consecutive trials. Patient provided cues for syntactic subjects of sentences and then manipulates objects to demonstrate correct actions in semantically reversible sentences. Seven active sentences and seven passive sentences with three-place verb argument structure presented in random order. For incorrect responses, sentence is repeated, clinician models the correct relationship, and then patient performs the correct action. Criterion for completion of task: 12 or more accurate responses over two consecutive trials. Twenty topicalized sentences (e.g., “Blue, the hat was, that the man on the corner was wearing”) presented orally to patient followed by two comprehension questions for each sentence. Question A targets sentence topic (e.g., “Which hat was the man wearing? The blue hat.”). Question B targets remaining information (e.g., “Which man was wearing the hat? The man on the corner.”). If patient does not answer Question A correctly, auditory cues provided to promote attention strategies (self-monitoring, rehearsal, listening and anticipating, repetition). May be combined with visual cues (printed sentences). Cues reduced systematically as response accuracy increases. If patient does not answer Question B correctly, feedback provided and correct answer modeled. No further cueing. Criterion for task completion: 32 or more correct responses over two consecutive trials. (table continues) Downloaded from: https://pubs.asha.org Richard Peach on 08/16/2019, Terms of Use: https://pubs.asha.org/pubs/rights_and_permissions Appendix A ( p. 2 of 3) .L-SAT Treatment Program Attention manipulation Anaphora Peach et al.: Language-Specific Attention Treatment for Aphasia Nominal Grounding Phase No. of items A 20 Sentences, 20 questions B 20 Sentences, 20 questions C 20 Sentences, 20 questions A 20 Sentences, 20 questions B 20 Sentences, 20 questions C 20 Sentences, 20 questions Treatment stimuli, tasks, cues, therapeutic operations, and advancement criteria Twenty sentences presented orally with close syntactic distance between referents and anaphoric pronouns and low load on working memory. Patient responds to a comprehension question for each sentence regarding referent of anaphoric pronoun (e.g., “Jasmine stayed after she finished the meal. Who finished the meal?”). If patient does not answer question correctly, auditory cues provided to promote attention strategies (self-monitoring, rehearsal, listening and anticipating, repetition). May be combined with visual cues (printed target). Cues reduced systematically as response accuracy increases. Criterion for advancement to Phase B: 16 or more correct responses over two consecutive trials. Twenty sentences presented orally with increased syntactic distance between referents and anaphoric pronouns and medium load on working memory. Patient responds to a comprehension question for each sentence regarding referent of anaphoric pronoun (e.g., “Tim argued with Kelly during the movie because she forgot to buy popcorn. Who forgot to buy popcorn?”). If patient does not answer question correctly, auditory cues provided to promote attention strategies (self-monitoring, rehearsal, listening and anticipating, repetition). May be combined with visual cues (printed target). Cues reduced systematically as response accuracy increases. Criterion for advancement to Phase C: 16 or more correct responses over two consecutive trials. Twenty sentences presented orally with long syntactic distance between referents and anaphoric pronouns and high load on working memory. Patient responds to a comprehension question for each sentence regarding referent of anaphoric pronoun (e.g., “The manager who questioned the teller from the local bank branch told the policeman how he suspected her of fraud. Who did the manager suspect of fraud?”). If patient does not answer question correctly, auditory cues provided to promote attention strategies (self-monitoring, rehearsal, listening and anticipating, repetition). May be combined with visual cues (printed target). Cues reduced systematically as response accuracy increases. Criterion for completion of task: 16 or more correct responses over two consecutive trials. Twenty sentences presented orally varying by definite (the) or indefinite (a) articles. Patient responds to yes/no comprehension questions for each sentence probing whether particular nouns are specific or nonspecific (e.g., “The girl in the class likes the boy. Do we know which boy the girl likes?” Correct answer = yes; “Tom found a home for the boy. Do we know which home Tom found?” Correct answer = no). Ten sentences require positive response, 10 sentences require negative response. If patient does not answer question correctly, auditory cues provided to promote attention strategies (self-monitoring, rehearsal, listening and anticipating, repetition). May be combined with visual cues (printed target). Cues reduced systematically as response accuracy increases. Criterion for advancement to Phase B: 16 or more correct responses over two consecutive trials. Twenty sentences presented orally varying by demonstrative pronouns (this, that, these, those). Patient responds to yes/no comprehension questions for each sentence probing whether particular nouns are in proximity (e.g., “This evidence should satisfy those detectives. Are the detectives close at hand?” Correct answer = no; “This book is more interesting than that movie. Is the book close at hand?” Correct answer = yes). Ten sentences require positive response, 10 sentences require negative response. If patient does not answer question correctly, auditory cues provided to promote attention strategies (self-monitoring, rehearsal, listening and anticipating, repetition). May be combined with visual cues (printed target). Cues reduced systematically as response accuracy increases. Criterion for advancement to Phase C: 16 or more correct responses over two consecutive trials. Twenty sentences presented orally varying by noun qualifiers (every, all, some, most, each, any). Patient responds to yes/no comprehension questions for each sentence probing whether particular nouns are enhanced or limited (e.g., “All of the guests enjoyed the party, but some guests left early. Did any of the guests enjoy the party?” Correct answer = yes; “Every player is part of the team but some receive more playing time. Do all players receive equal amounts of playing time?” Correct answer = no). Ten sentences require positive response, 10 sentences require negative response. If patient does not answer question correctly, auditory cues provided to promote attention strategies (self-monitoring, rehearsal, listening and anticipating, repetition). May be combined with visual cues (printed target). Cues reduced systematically as response accuracy increases. Criterion for task completion: 16 or more correct responses over two consecutive trials. 2809 (table continues) Downloaded from: https://pubs.asha.org Richard Peach on 08/16/2019, Terms of Use: https://pubs.asha.org/pubs/rights_and_permissions 2810 Journal of Speech, Language, and Hearing Research • Vol. 62 • 2785–2811 • August 2019 Appendix A ( p. 3 of 3) .L-SAT Treatment Program Attention manipulation Phase No. of items Clausal Grounding A 16 Sentences, 16 questions B 16 Sentences, 16 questions A 20 Sentence, 60 questions B 20 Sentences, 40 questions C 20 Sentences, 20 questions Windowing Treatment stimuli, tasks, cues, therapeutic operations, and advancement criteria Sixteen sentences presented orally with varying verb tenses (present, past). Patient responds to yes/no comprehension questions for each sentence probing specific meanings with respect to time (e.g., “The teacher announces that she was tired. Is the teacher tired now?” Correct answer = no; “Sara exclaims that she is excited. Is Sara exclaiming that she is excited now?” Correct answer = yes). Eight sentences require positive response, eight sentences require negative response. If patient does not answer question correctly, auditory cues provided to promote attention strategies (self-monitoring, rehearsal, listening and anticipating, repetition). May be combined with visual cues (printed target). Cues reduced systematically as response accuracy increases. Criterion for advancement to Phase B: 12 or more correct responses over two consecutive trials. Sixteen sentences presented orally with varying modal verbs (will, can, may, must). Patient responds to yes/no comprehension questions for each sentence probing specific meanings with respect to the possibility, probability, or certainty of an event (e.g., “The editor can veto the book proposal. Is there a chance the editor won’t veto the book proposal?” Correct answer = yes; “The student must pass his final exams. Is there a chance the student won’t have to pass his final exams?” Correct answer = no). Eight sentences require positive response, eight sentences require negative response. If patient does not answer question correctly, auditory cues provided to promote attention strategies (self-monitoring, rehearsal, listening and anticipating, repetition). May be combined with visual cues (printed target). Cues reduced systematically as response accuracy increases. Criterion for task completion: 12 or more correct responses over two consecutive trials. Twenty sentences presented orally describing a path event frame (the entirety of a path of motion). Patient responds to wh-questions (what, where, how) that window (focus attention on) the paths taken 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 (open paths). Questions elicit open path windowing for initial, medial, and final events in varying orders (e.g., “The ball kicked by the defender flew like a missile on a straight shot to the goal. How was the ball kicked? What kind of kick was it? Where did the ball go?”). If patient does not answer question correctly, auditory cues provided to promote attention strategies (self-monitoring, rehearsal, listening and anticipating, repetition). May be combined with visual cues (printed target). Cues reduced systematically as response accuracy increases. Criterion for advancement to Phase B: 48 or more correct responses over two consecutive trials. Twenty sentences presented orally describing a cycle event frame (an iterating cycle that is sequential but with no clear beginning, middle, or end). Patient responds to questions that direct attention to the departure and return phases of the event (phase windowing; e.g., “When the meal would get cold, I would keep reheating it. What did the meal do? What did I do?”). If patient does not answer question correctly, auditory cues provided to promote attention strategies (self-monitoring, rehearsal, listening and anticipating, repetition). May be combined with visual cues (printed target). Cues reduced systematically as response accuracy increases. Criterion for advancement to Phase C: 32 or more correct responses over two consecutive trials. Twenty sentences presented orally describing a factuality event frame (establish comparison of alternative conceptualizations for the occurrence/nonoccurrence of some referent). Patient responds to questions that heighten attention (factuality windowing) for an unrealized positive or negative event, an event at the opposite end of a continuum of certainty, or an overtly counterfactual event (e.g., “The salesman would have celebrated if he had exceeded his yearly quota. What happened because the salesman did not exceed his yearly quota?”). If patient does not answer question correctly, auditory cues provided to promote attention strategies (self-monitoring, rehearsal, listening and anticipating, repetition). May be combined with visual cues (printed target). Cues reduced systematically as response accuracy increases. Criterion for completion of task: 16 or more correct responses over two consecutive trials. Downloaded from: https://pubs.asha.org Richard Peach on 08/16/2019, Terms of Use: https://pubs.asha.org/pubs/rights_and_permissions Appendix B Direct Attention Training Baseline Procedure 1. Begin with the task listed for each component of attention. Complete only one trial of each task until baseline is determined. Task hierarchy increases from easiest to hardest within each component. See program for complete list of tasks. Sustained Attention–Auditory Sustained Attention–Visual Selective Attention–Auditory Selective Attention–Visual Working Memory Suppression–Auditory Suppression–Visual Alternating Attention–Auditory Alternating Attention–Visual Listening for Two Letters Fast (Level 4) Watching for Number Comparisons Hard–Fast (Level 10) Listening for 2 Letters Fast–Environmental Noise (Level 16) Watching for Number Comparisons (Hard)–Fast–White Noise (Level 34) 4-Number Sequences (0–100) Subtract 2 (Level 17) Happy–Sad Fast–Button Response (Level 2) Word Shapes Fast–Verbal Response (Level 8) Happy-Sad Fast–Button Response (Level 2) Word Shapes Fast–Button Response (Level 4) 2. Move to next hardest level within the same task if performance is greater or equal to 90% accurate with a false-positive (FP) rate of less than or equal to 10%; if performance is 100% accurate with an FP rate of less than 5%, move down to next hardest task in hierarchy. 3. Move up to next easiest task if performance is less than or equal to 60% accurate and/or if the FP rate is greater than 10%. a. If performance on a single task is less than or equal to 30% or the FP rate is greater than 50%, move up to next easiest task. b. If performance for two consecutive tasks is less than 50% and/or the FP rate is greater than 25%, move up to next easiest task. 4. If performance on a task is greater than 60% and less than 90% accurate and the FP rate is less than 10%, use as the starting point for treatment. 5. If patient does not meet starting criteria for easiest task (Level 1) in hierarchy, begin here with goal of increasing accuracy and reducing FP rate. Peach et al.: Language-Specific Attention Treatment for Aphasia Downloaded from: https://pubs.asha.org Richard Peach on 08/16/2019, Terms of Use: https://pubs.asha.org/pubs/rights_and_permissions 2811