Attention Deficit Hyperactivity Disorder

The association between oral language deficits and behavior disorders has been well-established in the recent literature in child psychopathology (Beitchman et al., 2001; Cantwell & Baker, 1991; Cohen, Barwick, Horodetsky, Vallance, & Im, 1998; Cohen, Davine, Horodetsky, Lipsett, & Jsaacson, 1993; Prizant, Audet, Burke, Hummel, Maher, & Theodore, 1990; Valiance, Im, & Cohen, 1999; Warr-Leeper, Wright, & Mack, 1994). There have been relatively few studies, however, specifically investigating language abilities of children with attention-deficit hyperactivity disorder (ADHD), the most commonly diagnosed psychiatric disorder in children. Despite reports of its frequent co-occurrence with language impairment (LI; Cohen et al., 1993; Cohen, Bariwick et al., 1998; Tirosh & Cohen, 1997), and considering the importance of language competence for successful social and academic functioning (Bashir & Scavullo, 1992), this lack of in-depth study of oral language abilities and related cognitive processes in ADHD children has created a critical gap in the research literature.

According to the DSM-IV (American Psychiatric Association, 1994), ADHD is characterized by developmentally inappropriate levels of hyperactivity/impulsivity and/ or inattention symptoms, which are pervasive across time and settings, and impairing to daily functioning. Although not considered directly in diagnosis, oral language deficits are implicated in both of these symptom clusters. For example, the DSM-IV suggests that in school-age children, hyperactive/impulsive symptoms may manifest as blurting-out answers in class, speaking out of turn, interrupting, and talking excessively, which also signal poor pragmatic language functioning. Alternately, inattentive symptoms such as appearing not to listen, frequent forgetting, and failure to follow through on instructions, implicate language comprehension abilities. In short, impairments in some aspects of language functioning may be integral to ADHD, rather than a correlate or a comorbid disorder.

The DSM-IV specifies that inattentive behavior must not be attributable to comprehension difficulties, however this is nearly impossible to verify through diagnostic interviews and behavior ratings. Furthermore, standardized checklists and behavior rating scales survey children's typical behavior without consideration of underlying cognitive processes or situational demands. Moreover, considering that everyday spoken language varies in complexity and processing demands across situations (e.g., school vs. home), one cannot completely rule out by interview and rating scales the possibility that subtle listening comprehension deficits might underlie some of the behavioral symptoms in ADHD children whose basic language skills appear normal.

Most language studies to date have focused on the expressive abilities of children with ADHD, reporting weaknesses such as poor sentence-formulation skills (Oram, Fine, & Tannock, 1997) and difficulties with coherence, organization, and self-monitoring of verbal production during the retelling of narratives (Purvis & Tannock, 1997; Tannock, Purvis, & Schachar, 1993; Zentall, 1988). These aspects of language expression that require planning and self-monitoring of performance have been associated with executive function deficits, which are believed to underlie the behavioral symptoms of ADHD (Douglas, 1999; Pennington & Ozonoff, 1996; Tannock & Schachar, 1996; Ylvisaker & DeBonis, 1998). However, it is also possible within the context of these considerations that some expressive discourse deficits in ADHD children might in part reflect weak comprehension abilities.

Only a few studies have examined receptive language ability in ADHD children, despite its functional importance in everyday social interactions and learning. One study by Tannock et al. (1993) found that despite having difficulty verbally retelling a complex story, ADHD children comprehended main ideas and facts from the story as well as normal controls. On the basis of a review of normal children's comprehension of story information, Lorch, Milick, and Sanchez (1998) suggested that ADHD children might be poorer than normal children in compreheading complex causal relationships in stories. In a subsequent study, Lorch et al. (2000) found that ADHD children comprehended factual details from televized stories as well as normal children, but were poorer in understanding and recalling some causal relationships among events in the stories. A third study by Brock and Knapp (1996) assessing reading comprehension for passages from science texts found ADHD children to have lower scores on comprehension questions than normal controls, despite having good word decoding skills. These preliminary findings suggest that ADHD children may comprehend surface details adequately, but may show deficits on tasks that require relatively higher degrees of vigilance, effort, and controlled processing. Beyond these few empirical findings, the area of language comprehension and ADHD has remained virtually unexplored.

Current conceptualizations of ADHD propose that a neurodevelopmental deficit in executive function limits development of self-regulation skills that guide behavior and cognitive functioning, thus leading to the symptoms and typical performance deficits associated with ADHD. Working memory is one aspect of executive function implicated in recent theories of ADHD (Barkley, 1997; Brown, 2000), and also central to current theories of language comprehension and discourse processing (Kintsch, 1998; Zwaan & Radvansky, 1998). Working memory, often described as a "mental workspace," refers to the ability to hold task-relevant information actively in mind during processing of information or problem solving, and is a key cognitive resource associated with individual differences in comprehension (Just & Carpenter, 1992). It has also been linked with a number of developmental outcomes associated with language functioning in children, such as reading comprehension ability (Nation, Adams, Bowyer-Crane, & Snowling, 1999; Swa nson, 1999), vocabulary acquisition (Baddeley, Gathercole, & Papagno, 1998), and early academic achievement (Gathercole & Pickering, 2000).

Most current models of working memory specify two functional components, one for short-term maintenance of information, and the other for manipulation of information during complex cognitive tasks (Miyake & Shah, 1999). One model further separates the maintenance component into separate units that maintain either verbal or visual-spatial information (Baddeley, 1986). Verbal information is maintained in a phonological loop, and visual information is held in a visuospatial sketchpad, each accessed as needed for processing by the central executive, where information is manipulated to serve complex cognitive functions. With regard to the role of working memory in comprehension, considerable evidence supports a link between verbal and visual-spatial working memory and language comprehension in adults (Just & Carpenter, 1992; Palladino, Comoldi, de Beni, & Pazzaglia, 2001; Zwaan & Radvansky, 1998), with some suggestion that verbal and visual-spatial working memory make unique contributions to different aspects of c omprehension (Friedman & Miyake, 2000). However, to date, these relationships have not been investigated in children's comprehension of discourse-level language.

Current theories of normal discourse processing propose that working memory is utilized to construct, maintain, and update detailed and coherent "mental representations" of both explicit (i.e., facts) and implied (i.e., inferential) information during listening and reading (Zwaan & Radwansky, 1998). More in-depth understanding and greater skill in recalling facts and making inferences are associated with more elaborate mental representations. Failure to make inferences may be one cause of comprehension deficits common to children with LI (Cain, Oakhill, & Bryant, 2000b).

Several recent studies have reported that children with ADHD have working memory deficits (Cohen et al., 2000; Karatekin & Asarnow, 1998;; Kempton et al., 1999; Martinussen & Mclnnes, 2000; Williams, Stott, Goodyer, & Sahakian, 2000), and a subset of these has attempted to address the potential link between working memory and basic language deficits in this population (Cohen et al., 2000; Williams et al., 2000). However, differences in the age range of participants, research designs, and working memory tasks, preclude clear interpretations of these preliminary findings. For example, findings from an important study by Cohen et al. (2000) that examined working memory and language abilities in children with ADHD and LI, suggested that ADHD children's working memory deficits were primarily related to their language abilities. However, the use of the sentence span task (Daneman & Carpenter, 1980) as one of their working memory tasks may have confounded their results, because it requires both sentence-level langua ge processing ability and working memory.

The present research was designed to examine concurrently the constructs of language comprehension and working memory in children with ADHD. The specific question addressed was whether ADHD children are impaired on listening comprehension tasks that require higher levels of working memory, self-monitoring, and controlled processing, even when their basic language abilities are adequately developed. While earlier studies have focused on comprehension of narratives, our objective was to investigate ADHD children's comprehension of language presented in the expository genre, which is typically experienced during classroom instruction and in textbooks.

Successful comprehension of language in academic situations requires competence in comprehending factual detail, making inferences, and the ability to self-monitor comprehension (Westby, 1991). Comprehending instructional language can be challenging even for normal listeners because it relies heavily on the expository genre, which is the formal language of explanation, description, and logical argument. Expository texts typically present new information; therefore, the listener or reader cannot rely as much on context and prior knowledge. They are also less predictable in structure, have greater information density, and consequently make greater demands on working memory resources (Graesser, Singer, & Trabasso, 1994). As classroom performance problems and school failure are common in children with ADHD (Zentall, 1993), and there is emerging evidence that their working memory skills are weak (e.g., Cohen et al., 2000), they may be experiencing unexpected difficulty comprehending more complex information than n ormal children. Consequently, examination of their ability to comprehend spoken expository information appears warranted.

The objectives of this study were threefold. The main objective was to examine three aspects of listening comprehension in children with ADHD with respect to expository information: comprehension of facts, inferences, and comprehension monitoring ability. A second objective was to assess the maintenance and manipulation components of working memory ability in both verbal and spatial modalities, and determine their relationship to listening comprehension ability. A third objective was to examine relationships among listening comprehension performance, parent-teacher behavior symptom ratings, and working memory abilities. Considering the frequent comorbidity of ADHD and LI (Cantwell & Baker, 1991; Cohen et al., 1993), a four-group design was employed to allow comparisons among groups with ADHD, ADHD+LI, LI alone, and Normal controls. This design also facilitated examination of language performance in children with co-occurring ADHD and LI, a seldomstudied population in the clinical literature.

Hypotheses were, for listening comprehension, that (1) children with LI (LI, ADHD+LI) would be more impaired than non-LI children (ADHD, normals) in comprehending facts and inferences, and monitoring their comprehension; (2) children with ADHD+LI would be more impaired than children with LI or ADHD alone, because of potentially additive effects of the two disorders; and (3) children with ADHD (no LI) would show impairments in malting inferences and self-monitoring comprehension, despite showing normal basic oral language abilities and unimpaired comprehension of facts. Next, on the basis of recent findings regarding working memory deficits in ADHD and LI, it was hypothesized that (1) children with LI (LI, ADHD+LI) would show poorer verbal and spatial working memory abilities (i.e., both span and maintenance components of working memory) than would non-LI children, (2) children with ADHD+LI would be more impaired than children with LI, and (3) children with ADHD (no LI) would show poorer verbal and spatial wor king memory abilities than would normals. Lastly, it was predicted that parent and teacher ratings of inattention and hyperactivity/impulsivity symptoms would be associated with both working memory and listening comprehension abilities.

METHOD

Sample

A community sample of 77 boys aged 9-12 years participated in the study, with signed informed consent from parents and verbal assent from children. ADHD participants were recruited from a caseload of previously diagnosed ADHD patients of one pediatrician serving a large rural area in southwestern Ontario. LI participants were recruited through a large public school board in the same region, and were students with identified language impairments, or who had academic problems and were awaiting language assessments. Normal controls were teacher-referred children in Grade 4, 5, and 6 classes with average achievement in all subject areas.

Procedures

Children were tested over two 2-hr sessions at school or in a professional office. Stimulant medication usually taken by the child was withheld for 36-48 hr prior to testing. None of the ADHD children in this sample were being treated with nonstimulants or medication for anxiety or depression. All children passed a standard audiometric screening, using a portable audiometer. Standardized measures were administered according to standard procedures during the first session, and the listening comprehension and working memory measures were administered during the second session.

Group Classification Measures

Children were classified into three clinical groups (ADHD, ADHD+LI, and LI) and a normal control group based on previous diagnosis of ADHD, current verification of the disorder from Parent and Teacher-reports on the Conners Rating Scales--Revised (CRS-R, Long Form; Conners, 1997), and three standardized measures of language ability (Peabody Picture Vocabulary Test--III, Dunn & Dunn, 1997; Expressive Vocabulary Test, Williams, 1997; Clinical Evaluation of Language Fundamentals--3 Semel, Wiig, & Secord, 1995).

Results of the CRS-R were used to confirm that children with a previous diagnosis continued to meet criteria for ADHD, using both symptom counts and standardized scores on separate CRS-R indices. Children were classified in the ADHD groups if one informant (e.g., teacher) endorsed at least six symptoms on the DSM-IV Inattentive or Hyperactivity/Impulsivity subscales of the CRS-R, and if T-scores from the other informant's ratings (e.g., parent) were in the clinical range (T-Scores > 60) on the Inattention, Hyperactivity/Impulsivity, ADHD, and DSM-IV indices. Elevated scores on the ADHD Index of the CRS-R are considered to be strong evidence for ADHD diagnosis when used clinically. T-Scores on Inattention and Hyperactive/Impulsive indices can be used to determine clinical subtypes of ADHD. It should be noted that for a symptom to be present on the DSM-IV subscales, behaviors must be rated as a 3 on a 0-3 scale, with 3 indicating high frequency of the behavior ("very much true, very often"). Five referred child ren with previous ADHD diagnoses did not meet these classification criteria regarding symptom counts and were not included in the study.

Children were classified in the LI groups if their language scores fell 1 SD below the mean of the standardization sample on two language measures, or if one score was 1.5 SD below the mean, and if there was evidence that both receptive and expressive language deficits were present. None of the children classified as LI met criteria for ADHD or had previous concerns about ADHD. Children classified in the ADHD+LI group met criteria for both disorders. Children in the normal group did not meet criteria for either disorder. Seventy-seven of 101 children referred for the study met inclusionary criteria for one of the four groups. Of the remainder, 5 ADHD referrals did not meet the required symptom counts, 2 ADHD referrals had expressive-only language deficits, and 17 children were low achievers in reading but did not meet criteria for either ADHD or LI.

Two additional measures were administered: the Block Design subtest from the Wechsler Intelligence Scale for Children-III (Wechsler, 1991) was used as a nonverbal cognitive ability measure, using Canadian norms for this sample. Also, considering the frequent comorbidity of ADHD and Reading Disability (Tannock & Brown, 2000), the Word Attack subtest from the Woodcock Reading Mastery Test--Revised (Woodcock, 1987) was administered. Our aim was to evaluate phonological decoding ability, one important language subskill related to reading, in the decoding of nonwords.

Table I presents results of MANOVAs regarding selected sample characteristics. The four groups (ADHD, ADHD+LI, LI, and Normals) did not differ in age. Both ADHD groups had parent and teacher ratings in the clinically significant range on the Oppositional and Anxiety indices of CRS-R. For sample description purposes, ADHD subtypes were identified by examining the DSM-IV subscales and T-scores for inattention and hyperactivity/impulsivity symptoms. Children who showed symptoms and elevated t scores (i.e., more than six symptoms, and T-scores >60) in both symptom clusters were considered to represent the Combined subtype, and children whose scores were elevated only for inattention symptoms and whose hyperactivity/impulsivity scores were well below clinical levels (i.e., <55) were considered to represent the primarily inattentive subtype. The composition of the two ADHD groups was roughly similar with respect to these two subtypes. There were no children who demonstrated a predominantly hyperactive/impulsive pro file on the CRS-R ratings. With respect to symptom counts on the DSM-IV subscales on the CRS-R, the two ADHD groups did not differ in mean number of inattention and hyperactive/impulsive symptoms in Parent and Teacher reports--means and standard deviations for symptom counts for Parent inattention: ADHD group 4.5 (2.4); ADHD +LI group 4.0 (2.7); Parent Hyperactive/Impulsive ADHD group 2.0(2.5); ADHD+LI group 1.8(2.4); Teacher Inattention: ADHD group 5.5 (2.2); ADHD+LI group 5.7 (2.4) Teacher Hyperactive/Impulsive ADHD group 3.0 (3.2); ADHD + LI group 1.9 (2.4).

Table II displays means, standard deviations, and results of group comparisons for the ADHD and LI classification measures (i.e., CRS-R; standardized language tests), and the Word Attack and Block Design subtests. The two ADHD groups (ADHD, ADHD+LI) did not differ in parent and teacher ratings of ADHD symptoms across indices used for classification, and the two LI groups (ADHD+LI, LI) did not differ on standard scores on the receptive and expressive language measures. These analyses provided assurance that group differences on the dependent measures could not be attributed to differences in severity of ADHD symptoms or language deficits across the clinical groups. Parent and teacher behavior ratings (CRS-R) for the LI group were elevated compared with the Normal group but were below clinically significant levels for ADHD diagnosis or classification. The LI group also had higher teacher ratings for anxiety and oppositional behavior than did the Normal group. None of the LI participants upon referral had teache r concerns regarding specific attention difficulties. The three clinical groups each had participants with poor Word Attack scores (e.g., standard scores <85), suggesting the presence of reading difficulties, with roughly similar proportions of lower scores in each group.

Dependent Measures and Procedures

Listening comprehension ability for expository passages was assessed using two novel expository passage comprehension tasks, which were initially piloted with normally achieving and LI children (see Appendix). The first task assessed comprehension of facts and inferences in four audiotaped expository passages controlled for length and vocabulary difficulty (Chall, 1995), each containing some novel information on familiar topics (e.g., animals, procedures to make something) that children might encounter during classroom instruction. Following a familiarization trial, children listened to the test passages, each one followed by 6 factual and 10 inference questions. Inference questions included five direct questions (i.e., question--answer) and five true--false questions that the child had to support with an explanation. As many of the children in the clinical groups had expressive language problems, specific prompts such as "tell me more" or "what do you mean by that" were used to promote optimal responses (i.e ., they were not penalized for poor expression). The true--false questions were designed to assess inferential comprehension with reduced demands on retrieval processes, which might be difficult for children with ADHD and LI. Responses to the comprehension questions were tape-recorded for scoring reliability procedures and further analyses. A criterion-based scoring guide was used to assign a full score of 2 or a partial score of 1 to responses, or 0 to completely incorrect responses, for a maximum possible score of 48 for facts across the four passages, and 40 for each type of inference question. Responses had to meet specific predetermined content criteria for assignment of scores. Scoring reliability was checked for 20% of the sample by a second independent rater who was trained to use the scoring procedures, and who was blind to group classification of each participant; interrater agreement regarding scoring for factual questions and both types of inference questions ranged from 88 to 93.2%.

A narrative comprehension task was also included in the protocol as a link with earlier studies of comprehension in ADHD children. The Listening to Paragraphs subtest from the Clinical Evaluation of Language Fundamentals--Revised (CELF-3) comprises two narrative passages that are read aloud to the child with no picture cues, each followed by five comprehension questions, which deal primarily with facts from the stories. The passages are roughly similar in length to the expository passages described earlier. The total possible score out of 10 is converted to a standard score. It should be noted that this task is a supplementary subtest of the CELF-3, and not used in calculating composite Receptive or Expressive Language scores. Therefore, it was not used in classifying participants into the four groups.

The second listening comprehension task assessed the ability to self-monitor comprehension by using an error detection paradigm (Markman, 1979). Two types of expository passages, instructions and descriptions, were developed, each comprising four passages, some of which contained errors that would cause the passage not to make sense. The instructions passage contained instructions for familiar tasks that required following a specific sequence, with three of four passages presenting instructions in the wrong order. The descriptions passages presented information on interesting topics (e.g., animals habitat), with three of four passages containing factual inconsistencies. Children were asked to listen carefully and act like a judge to determine if the passages made sense or if they had mistakes in them, and to explain any errors they heard. They listened to the passages twice before giving their final judgment, with their initial responses for the first hearing also recorded. Each correctly identified error acc ompanied by a plausible explanation was assigned a score of 2. For the foil passages (i.e., no error), a score of 1 was assigned for correctly stating that the passage made sense (no explanation was required), for a maximum score of 7 for each task. The two error detection tasks were presented in random order throughout the sample, and were not administered consecutively. Scores for both first and second hearings of the passages were used in the analyses.

Working memory was assessed in both verbal and spatial modalities by using two tasks involving sequential information; tasks selected for this study were considered to make minimal demands on language processing, and examined both the maintenance and processing functions of working memory separately. The Numbers subtest from the Children's Memory Scale (Cohen, 1997) assessed forward (verbal span) and backward (verbal working memory) repetition of digits. The Finger Windows subtest from the Wide Range Assessment of Memory and Learning (WRAML; Adams & Sheslow, 1990) assessed memory for sequences of spatial locations, using the standardized forward (spatial span) administration plus a novel backward (spatial working memory) administration that paralleled the format of the verbal task. In the standardized forward administration of this task, the examiner inserts a pencil through a series of randomly spaced holes ("windows") on an 8 x 11 inch card at a rate of one hole per second. The child, who is seated on the o pposite side of the card, then reproduces the same spatial sequence by putting his or her finger through each window. Items gradually increase in length from sets of two to sets of six windows, thereby tapping spatial memory span. In the backward administration, the child watches the sequence of windows shown by the examiner and then reproduces it in backward order. Raw scores for both verbal and spatial tasks were used for data analyses.

Statistical Analyses

Analyses of group differences on the listening comprehension and working memory tasks were conducted via MANOVAs and MANCOVAs with post hoc comparisons. Simple correlations were then examined to determine relationships among listening comprehension and working memory measures, and behavior ratings. The alpha level was set a priori at p < .05. The Bonferroni test was applied for post hoc multiple comparisons.

RESULTS

Listening Comprehension

Standardized Narrative and Expository Passage Comprehension Tasks

Table III summarizes results of multivariate analyses applied to the scores on the standardized and experimental listening comprehension tasks. For each dependent measure, Listening to Paragraphs (CELF-3), expository facts, direct inferences, true-false inferences, there was a significant main effect for group. For the Listening to Paragraphs subtests, post hoc comparisons found that the LI group's scores on this narrative task were significantly poorer than those of the Normal group and both ADHD groups, but that the other groups did not differ from each other. For the expository passages, both LI and ADHD+LI groups had significantly lower mean scores for factual questions (p < .01), although they did not differ from each other. As predicted, the ADHD group performed as well as the Normal group in factual comprehension. For comprehension of inferences (direct question-answer; true- false), the two LI groups were again significantly poorer than the ADHD and Normal groups. The ADHD group outperformed the ADHD+ LI and LI groups on factual and direct (i.e., question-answer) inference scores, but their true-false inference scores were as poor as those of the other two clinical groups. The ADHD group was marginally poorer than the Normal group (p < .06) in comprehending direct inferences, with their groups mean score approximately midway between the LI groups' and Normal group means. For inferential comprehension assessed via true-false questions, the difference between the ADHD and Normal groups performance (ADHD < Normal) reached statistical significance (p < .05).

To address the possibility that differences in inferential comprehension scores between the ADHD and Normal groups might be due to differences in language sub skills, such as phonological skills associated with decoding, or sentence formulation ability, these analyses were repeated as separate MANCOVA's using scores from the Word Attack subtest of the WRMT-R and the Formulated Sentences subtest from the CELF-3 as covariates. When Word Attack scores were covaried, the pattern of significantly poorer inferential comprehension in the ADHD versus Normal groups was retained, although the findings were at a trend level for the true-false inferences (p < .07). When Formulated Sentences scores were covaried with the listening comprehension scores, group differences were eliminated for the direct inference scores, but maintained at the trend level (p < .06) for true-false inferences.

A third set of analyses was conducted for the listening comprehension measures but included only children whose Word Attack scores were within 1 SD from the mean (i.e., standard scores >85). This analysis was intended to further ensure that comprehension deficits were independent of phonologically based reading difficulties. Although this approach reduced group sizes somewhat with consequent loss of power, general maintenance of the above pattern of results was suggested. For the Listening to Paragraphs subtest, only the LI group's performance was significantly poorer than that of the Normal group. For the expository passages, ADHD children comprehended facts as well as normal children, however they were poorer at comprehending both direct (p < .09) and true-false inferences (p < .03). Also consistent with initial analyses, ADHD children appeared to outperform ADHD+LI children in answering direct inference questions (p < .09) and true-false inference questions (p < .01). The ADHD+LI and LI groups' performance did not differ from one another on any of the measures.

Error Detection Tasks

Table IV presents results of separate repeated measures analyses applied to scores on the two error detection tasks, with Group as the between-participants factor and Time (i.e., first hearing -- 2nd hearing) as the within-participants factor. For the instructions task, there were significant effects of Time (p < .0001) and Group (p < .0001); despite group differences, all groups improved their scores after the second hearing of the passages. Post hoc tests indicated that all three clinical groups (ADHD, ADHD+LI, and LI) were significantly poorer than the Normal group in detecting the sequence errors in the instructions, but did not differ from each other. On the descriptions task, there were again significant effects for Time and Group (p < .000 1). Inspection of means indicated that there were no group differences in the first hearing scores. A significant Group x Time interaction suggested that the Normal group showed relatively greater improvement in their second hearing scores than did the other groups. Post hoc comparisons indicated that the ADHD+LI group was significantly poorer than the Normal group in detecting factual inconsistencies, however their performance did not differ statistically from that of either the ADHD or LI group.

Verbal and Spatial Span and Working Memory

Table V displays multivariate analyses applied to the verbal and spatial span and working memory measures. For verbal span (Numbers subtest forward), the LI and ADHD+LI groups were significantly poorer than the ADHD and Normal groups, which did not differ in performance. For verbal working memory (Numbers subtest backward), all three clinical groups were significantly poorer than the Normal group, and did not differ from each other in their mean scores. The same pattern was found for performance on the spatial span (Finger Windows forward) and spatial working memory tasks (Finger Windows backward); again the three clinical groups did not differ from each other statistically in their obtained scores on these measures. These analyses were repeated using Word Attack scores as a covariate, given the relationship between phonologically based decoding and working memory (Wagner, Torgesen, & Rashotte, 1994). This approach eliminated group differences in verbal span, but did not alter findings for verbal working memo ry. However, for spatial memory, covarying Word Attack scores eliminated group differences only for the LI group, whereas the two ADHD groups continued to show significantly poorer spatial span and working memory (p < .01) than did the Normal group.

Relationships Among Listening Comprehension and Working Memory Scores and Parent and Teacher Behavior Ratings

To address the question of whether listening comprehension performance was associated with parent-teacher ratings of inattention and hyperactivity/impulsivity symptoms, correlations are reported in Table VI for scores among these measures and the verbal and spatial working memory measures. Several significant moderate correlations were evident among the passage comprehension and error detection scores, parent and teacher symptom ratings, and verbal and spatial working memory scores. Comprehension of facts was associated with teacher inattention ratings and verbal and spatial memory scores. Both types of inference scores were significantly associated with teacher inattention ratings, with somewhat weaker associations evident with parent inattention ratings and teacher hyperactivity/impulsivity ratings. Inference scores were also significantly associated with both span and working memory components of verbal and visual-spatial memory. Error detection scores for the instructions task were associated with parent and teacher inattention ratings, spatial span, and verbal and spatial working memory. For the descriptions task, error detection scores were associated with teacher inattention ratings, and parent and teacher hyperactivity/impulsivity ratings. Correlations were also significant between the error detection scores on descriptions, and both memory measures. In both error detection tasks, stronger associations were evident between with the working memory (i.e., digits backward, Finger Windows backward) compared with the span (i.e., forward) measures.

DISCUSSION

A main finding of this study was that a community sample of children with ADHD, with adequate language abilities for their age as measured by commonly used standardized language tests, showed evidence of comprehension deficits when listening to spoken expository passages. In support of our hypothesis and consistent with earlier findings of Tannock et al. (1993) regarding narrative comprehension, ADHD children comprehended factual details from standardized narrative passages and expository passages as well as normal children. However, they had significantly more difficulty than did normal children with subtle aspects of comprehension such as making inferences from the expository information, and monitoring their comprehension of instructions. Furthermore, the general pattern of findings for factual versus inferential comprehension was maintained even when word decoding and sentence formulation abilities were statistically controlled.

As expected, children with LI and ADHD+LI had significantly more difficulty than the ADHD and Normal groups in comprehending facts and inferences from the expository passages, but contrary to our prediction, the ADHD+LI group was not poorer than the LI group across the measures. This finding suggests that the presence of ADHD, in addition to a language impairment, does not further compromise listening comprehension ability beyond the effect of the language impairment alone. However, because this study did not include sufficient numbers of the three ADHD subtypes (i.e., inattentive, hyperactive/impulsive, and combined) to examine whether this finding would hold across subtypes or whether degree of comprehension impairment varies with subtype, further investigation of this issue is needed.

A second important finding was that although the ADHD children's verbal memory span (i.e., maintenance) was comparable to that of the Normal group, they showed poorer verbal working memory (i.e., manipulation) skills, consistent with findings of several recent studies and adding to the growing evidence that working memory deficits are a consistent cognitive correlate of ADHD (e.g., Cohen et al., 2000; Kempton et al., 1999; Martinussen & McInnes, 2001; William et al., 2000; see also Barkley, 1997; Brown, 2000). Furthermore, the ADHD group was significantly poorer than the Normal group on both the spatial span and spatial working memory measures, and their performance was as poor as that of the other clinical groups. These findings suggest that, overall, ADHD children's spatial memory skills are more impaired than their verbal memory skills. However, it should be noted that our working memory tasks focused on verbal and spatial sequences. Therefore, future investigations should examine whether similar patterns of deficits (e.g., span vs. maintenance) are evident in these clinical groups using other types of working memory tasks (e.g., static matrices tasks, self-ordered pointing tasks; Petrides & Milner, 1982), or subtests from the Cambridge Neurological Test Automated Battery (CANTAB; Sahakian & Owen, 1992). As with the listening comprehension measures, the prediction of greater impairment in the ADHD+LI group than in the LI group on verbal and spatial working memory tasks was not supported by these data. The two groups were each significantly poorer than the Normal group was on all measures but showed similar levels of impairment.

Comprehending complex and lengthy spoken information without specific visual cues requires coordination among several interdependent language and cognitive resources. Working memory is a key cognitive resource in comprehension, as it facilitates the ability to hold task-relevant information on-line while comprehension processes take place. Current theories of language comprehension based on adult studies propose that mental representations of both explicit and implied information are formed during listening, which are maintained in working memory and continually modified as new information is processed (Kintsch, 1998; Zwaan & Radvansky, 1998). Mental representations of text-based information may involve both verbal and visual working memory, and different genres of verbal information may make differential demands on available working memory resources. Considering that there is relatively little empirical information regarding normal development of text-level comprehension and associated cognitive processes in children, existing adult models must be relied upon for clinical studies involving children. The expository passage comprehension task in this study was designed to be challenging with respect to working memory demands, focusing on ADHD children's ability to comprehend novel and complex information, such as that encountered in instructional language at school.

A potential explanation for the ADHD group's poorer comprehension of inferences may be contextualized within Baddeley's model of working memory (Baddeley, 1986), the normal discourse comprehension processes described earlier, and the particular task demands of the expository comprehension task in this study. Comprehension of inferences while listening to these passages would require both on-line formation of mental representations of information presented (i.e., during listening) and the ability to retrieve and connect two or more mental representations relevant to each inference question being asked. Answering the comprehension questions "off-line" would require both the short-term maintenance (i.e., the phonological loop) and manipulation functions (i.e., the central executive) of verbal working memory while information was being retrieved and considered for the answer. If the child showed an intact verbal span but weak ability to simultaneously recall and manipulate information from the passages, comprehen sion of inferences might therefore be limited.

For example, in the Making Mummies passage (see Appendix), one inference question asks: What did the mummy smell like when it was put into the tomb? To answer this question correctly, the child would have to retrieve mental representations of the following explicit text information: "When the body was all wrapped, sweet smelling oils were poured all over it and allowed to soak in for several weeks," and "The mummy was put into a coffin and lowered deep into the pyramid through a secret passageway." These representations would then be maintained in working memory while the child made the implied connections between them: "like sweet smelling oils" or "it smelled good from the oils." Without available visual cues (e.g., pictures, diagrams), one would assume that mental representations formed during listening would be verbally based, which was supported by significant correlations between verbal span and verbal working memory and passage comprehension. However, spatial span and spatial working memory were also s ignificantly correlated with these measures, raising the speculation that both verbal and visual-spatial skills are implicated in language comprehension even with auditory-only input.

The finding of the ADHD group's significant relative deficit in spatial span compared with verbal span suggests that ADHD children may have significantly more difficulty maintaining visual-spatial than verbal information in working memory during performance of complex tasks. Furthermore, correlations between spatial span scores and listening comprehension scores were as strong as those for verbal span, suggesting that children's visual-spatial memory capacity is associated with their language comprehension, possibly by facilitating formation and use of visual imagery during listening. Spatial span deficits might affect the ability to invoke mental imagery needed for processing procedural explanations or task instructions that involve spatial or sequential information, which could in turn affect both on-line and off-line comprehension processes. Clearly, more study is needed to further explore these relationships.

The ADHD group was also significantly poorer than the Normal group in detecting errors in a sequence of instructions (see Appendix). This difference occurred despite the fact that the instructions passages were shorter and linguistically easier than the other comprehension measures, and the information involved familiar routines as opposed to novel information. Therefore, children could rely on existing procedural schema or familiarity with the routine to determine if the instructions made sense or not. Scores on the instructions task were significantly correlated with both verbal and spatial working memory scores, but correlations with spatial working memory were higher, suggesting that monitoring ability may be more spatially than verbally mediated, or that when verbal-plus spatial working memory resources are available, complex task performance is enhanced. Individuals who are able to visualize and hold on-line an entire sequence of steps while listening to task instructions may be better able to complete complex tasks successfully and efficiently. ADHD children's poorer performance in comprehending and monitoring their listening for errors on the instructions task may have been associated with their spatial span and working memory deficits, which would limit their ability to visualize and hold enough information to determine which steps were out of order. This possibility has important implications for ADHD children because task completion and failure to follow through on instructions are both within the list of inattentive symptoms used for diagnosis.

STUDY LIMITATIONS

These findings should be replicated using a larger sample, given the relatively small size of each clinical group. Furthermore, even though covarying Word Attack scores did not alter the general pattern of findings on listening comprehension scores, these results should be replicated with a larger group of ADHD children without reading difficulties and without LI. Also, as this study was restricted to the study of boys with ADHD, future investigations should extend this methodology to include girls with ADHD. Preliminary evidence from normative studies suggests that females may exhibit stronger spatial working memory skills than may males (Duff & Hampson, 2001), although it is not known whether gender differences in this aspect of working memory are also evident in ADHD children.

Clinical and Educational Implications

Our findings have several general clinical implications. First, some academic problems that are typically attributed to compliance or task-completion problems in children with ADHD may be an outcome of weak comprehension skills for longer complex information such as is presented in classroom instructions, lessons, and textbooks. This possibility has particular relevance in higher grades where requirements for accurate and efficient comprehension of new information increase markedly. Second, working memory deficits have been linked to academic underachievement (Gathercole & Pickering, 2000), and may be an important factor underlying the academic problems of children with ADHD. Third, given that potential consequences of comprehension failure include inappropriate and/or off-task responses, frustration, performance errors, and misinterpretation, of social cues, all characteristic of ADHD, these findings suggest that some behaviors used to diagnose ADHD may not be solely attributable to current constructs of hyp eractivity/impulsivity and inattention. While speculative, this issue has implications for reevaluating the role of linguistic (e.g., language comprehension) and cognitive factors (e.g., working memory) that may underlie behavior symptoms used to diagnose ADHD. Moreover, as these results were obtained under ideal listening conditions and children were given optimal support to express their responses, their functional comprehension abilities in more complex and less than ideal listening situations (e.g., the classroom) may be further compromised.

The findings of this study suggest that the issue of LI co-occurring with ADHD goes beyond formal diagnostic categories, and presents some unique clinical challenges. Although our clinical groups were carefully classified according to rigourous criteria, the data indicate that in general, the ADHD group had lower scores than did the Normal group on some standardized receptive and expressive language tests although none in this group would come close to meeting clinical criteria for diagnosis of LI or qualify for language intervention services at school. Yet, the impact of a combination of low-average basic language abilities, weak inferential comprehension in listening, and significant deficits in verbal and spatial working memory, would have significant implications for an ADHD child's academic performance in the classroom and on standardized achievement tests. Second, posthoc comparisons indicated that the ADHD group was superior to the ADHD+LI group on scores for facts and direct inferences but were as imp aired on the true--false inference questions and comprehension monitoring for instructions. There may be a threshold of task difficulty at which ADHD children who appear to have normal language abilities begin to function like children with language deficits.

A third issue concerns the LI children in this sample, whose teachers rated them higher on inattention symptoms than did teachers of the Normal group. Although these children did not come close to meeting classification criteria for the ADHD+LI group, their slightly higher inattention ratings, though not in the clinical range, reinforce the notion that some inattention symptoms may be associated with underlying language deficits.

The above issues are also relevant to assessment and management practices regarding ADHD co-occurring with other disorders. Clinically, behavior rating scales and parent/teacher interviews are typically used to diagnose ADHD, however presence of frequently co-occurring disorders such as LI are difficult to ascertain using these approaches. For example, despite evidence that LI is present in as many as 50% of ADHD cases, it is identified only upon formal assessment, and tends to be obscured by parental and teacher focus on disruptive behavior (Cohen et al., 1993; Cohen, Barwick et al., 1998; Cohen, Mevna et al., 1998). In this community sample, it is noteworthy that only 4 of 18 children in the ADHD+LI group had been previously assessed for possible language deficits, despite having chronic academic difficulties. Clinicians should be aware that behavior and learning difficulties reported by parents and teachers may be outcomes of co-occurring basic language deficits. Second, given that presence of co-occurring language deficits is not addressed in the DSM-IV diagnostic criteria, and comprehension difficulties may be ruled out or overlooked superficially during routine clinical assessments, unidentified LI could limit performance on other related assessments (e.g., intellectual ability, academic achievement). Undetected LI could also influence the potential success of common behavioral treatment programs for ADHD such as social skills groups or counseling that rely on oral communication skills (i.e., talking, listening, and comprehending).

Our findings also have educational implications for ADHD children who are not LI. The ADHD group's normal scores on the standardized narrative measure suggest that narratives are relatively easy for them to comprehend. Furthermore, they comprehended factual detail from expository passages as well as normal children. These apparently normal comprehension skills may mask more subtle comprehension deficits in ADHD children, which may not surface until comprehension demands increase in higher grades. For example, teachers may report young children being assessed for ADHD as having normal comprehension skills in early grades where teaching focuses on basic skills, and most literacy instruction involves the familiar and predictable genre of narratives. Furthermore, clinical assessments of comprehension at this level tend to also focus on story comprehension skills, often emphasizing factual details over story inferences. However, when instruction becomes more content-focused in higher grades, classroom language ten ds to shift to the less predictable expository genre, and children must develop metacognitive strategies and critical thinking skills to process complex and novel information. Subtle difficulties with inferential comprehension and comprehension monitoring not evident in earlier grades may begin to affect ADHD children's later learning and achievement. Findings of this study suggest that teachers may need to fine-tune their observations of ADHD students to ensure that they have understood information in explanations and descriptions presented verbally or in print, regardless of how well-developed their oral language skills appear on the surface.

Our findings also have implication for teaching strategies used with ADHD children. Although verbal span and working memory were associated with factual and inferential comprehension, so was spatial span and working memory. Teaching children to actively visualize information during listening and reading is a common strategy to develop and improve comprehension and memory (Harvey & Goudvis, 2000). It is possible that spatial memory skills may at times be needed to visualize aspects of spoken or written information in order to connect ideas for later retrieval. Therefore, ADHD children will continue to need visual supports (e.g., diagrams, illustrations, and graphic organizers) when processing new or complex information at school to reduce the load on both verbal and visual-spatial working memory.

Lastly, there is a need for more clinical techniques and instruments to assess language and cognitive functioning in children with ADHD. For example, existing standardized tests do not adequately address comprehension of complex language. The novel listening tasks developed for this study were modeled on tasks typically found in reading comprehension tests, but were also designed to model real-world learning situations that children face every day, and were effective in differentiating good from poor comprehenders in this sample. No comparable well-controlled standardized instruments are available commercially that would tap the skills investigated in this study. There is also a dearth of working memory tasks that have been clinically validated and that clinicians can use to predict functional outcomes in ADHD children. Furthermore, the associations found among parent and teacher ratings of inattention symptoms, listening comprehension skills, and working memory suggest the need for diagnostic measures that a ddress possible language and cognitive deficits in ADHD children, while determining presence or absence of symptoms. One recent clinical approach to this issue involves supplementing behavior rating scales with semistructured interviews (e.g., Tannock, Hum, Masellis, Humphries, & Schachar, 2000) that probe behavior symptom reports to clarify symptom descriptions. Given the complexity of ADHD, approaches such as this are needed to validate informant reports of behavior symptoms so that interventions can be based on comprehensive assessment of all areas of deficit.

ADHD is associated with chronic impairment in academic and social functioning, and places a considerable economic burden on health and education resources. Oral language deficits limit one's ability to participate fully in situations requiring accurate and efficient language processing, and by themselves can affect children's adaptive behavior, learning, social skills, and mental health (Johnson et al., 1999; Prizant et al., 1990). Both basic and higher-level receptive and expressive language abilities have been underresearched in ADHD children. Our findings raise the possibility that listening comprehension deficits co-occurring with ADHD, both at basic and higher levels of language functioning, underlie some behavior symptoms associated with ADHD. This issue has critical potential importance to the field if weak listening comprehension skills also extend into the domain of complex discourse in social interactions, an issue that should be addressed in future investigations.

APPENDIX

Listening Comprehension Passage (Example)

Making Mummies (Chall-Dale Level: Grades 5-6)

Do you know how mummies were made? In ancient Egypt, kings believed they could live forever, after they died, by turning their bodies into mummies that would not rot or decompose. When a king died, priests from the temple went to work to turn his body into a mummy. It took about 2 months to get the body ready to be buried in one of the greatest pyramids. First, the blood was drained out of the body, and then the priests took out the brain, heart, liver, and stomach. Each of these was kept in a special jar and buried with the mummy later on. Next, the body was dried with salt, and then special oils were rubbed into the skin to keep it soft. After it was all dried out, the body was wrapped with long strips of thin linen cloth. The priests put the king's jewelry and precious stones in between the layers of cloth. When the body was all wrapped, sweet smelling oils were poured all over it and allowed to soak in over several weeks. Then, the mummy was put into a coffin and lowered deep into the pyramid through a se cret passageway. Larger valuables like gold vases and statues went down into the king's tomb with the coffin. Finally, the entrance to the passageway was sealed up forever so that no one could disturb the king in his afterlife.

Factual questions (examples)

1. Who turned the king's body into a mummy?

priests

2. Why did the king want his body made into a mummy?

to last forever

3. What was used to dry the king's body?

salt

Inferential questions

1. What did the mummy smell like when it was put in the tomb?

sweet smelling oils

2. What did the king think he would need in his afterlife?

jewelry, gold

3. Who knew where the mummy's tomb was located?

priests

T/F inferential questions with explanation for correct responses

1. The king was afraid of living in his next life.

F

2. The king was generous to others with his gold and riches.

F

3. People could go and pray for the king in his tomb.

F

Error Detection Passages (Examples)

Instructions: Changing a fiat tire (Chall-Dale Level: Grade 4)

It is a good idea to know the steps in changing a flat tire because sooner or later it will probably happen to you or your family. If your tire goes flat while you are driving, the first thing to do is slow down and pull over to the side of the road. Make sure to put on your emergency brake so that your car does not move while you change the tire. Then you have get the spare tire, tools, and jack out of the trunk. Next, you loosen the nuts on the wheel that has the fiat tire, take out the bolts, and take the wheel off After that, you jack up the car so it is up off the road. Then you put the spare tire on and tighten up the nuts. Finally, lower the jack so the wheel is back down on the road. Put away the tools and jack, and now you can get on your way again.

Descriptions: The Newfoundland Pony (Chall-Dale Level: Grades 5-6)

Have you ever heard of a Newfoundland pony? It was a wild pony that used to roam the high rocky cliffs of Newfoundland about 200 years ago. Early settlers captured many of them to use as work animals because they had strong backs for pulling heavy loads. The ponies were valuable to the farmers because they could do a lot of labor and were strong and healthy. They were also quite gentle in nature and could be ridden safely by young children. There was only one problem the farmers had to be concerned about. The ponies had to be kept in their own special stables because they had bad tempers, and could become quite dangerous. Their main job was the hard work of pulling sleds, ploughing fields, and moving loads of rock off land that was being cleared for farms. Their thick coats helped them to stay warm and work through the bitter winds of winter. The ponies also had strong teeth and could even eat bark or twigs in winter, when there was not much vegetation. Nowadays, there are only a few Newfoundland ponies left, but they are protected in special government parks.

ACKNOWLEDGMENTS

This research is based on the first author's doctoral dissertation. A. McInnes is a postdoctoral fellow in the Brain and Behavior Research Program, supported by the Reva Gerstein Fellowship in Pediatric Psychology and the Research Training Institute of The Hospital for Sick Children. R. Tannock was supported by a Medical Research Council of Canada Scientist Award. The authors also thank Dr E. Porter and the Bluewater District School Board for their assistance in recruiting participants for this research.

Received February 22, 2002: revision received September 17, 2002; accepted September 28, 2002

REFERENCES

Adams, W., & Sheslow, D. (1990). Wide range assessment of memory and Ilearning. Wilmington, Delaware: Wide Rang.

American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders, (4th ed.). Washington, DC: Author.

Baddeley, A. D. (1986). Working memory. New York: Oxford University Press.

Baddeley, A. D., Gathercole, S., & Papagno, C. (1998). The phonological loop as a language learning device. Psychological Review, 105, 158-173.

Barkley, R. A. (1997). Behavioral inhibition, sustained attention, and executive functions: Constructing a unified theory of ADHD. Psychological Bulletin, 121, 65-94.

Bashir, A., & Scavuzzo, A. (1992). Children with language disorders: Natural history and academic success. Journal of Learning Disabilities, 25, 53-65.

Beitchman, J. H., Wilson, B., Johnson, C. J., Atkinson, L., Young, A., Adlaf, E., et al. (2001). Fourteen year follow-up of speech/language impaired and control children: Psychiatric outcome. Journal of the American Academy of Child and Adolescent Psychiatry, 40, 75-82.

Brock, S. E., & Knapp, P. K. (1996). Reading comprehension abilities of children with attention-deficit hyperactivity disorder. Journal of Attention Disorders, 1, 173-185.

Brown, T. E. (Ed.). (2000). Attention deficit disorders and comorbidities in children, adolescents, and adults, Washington, DC: American Psychiatric Press.

Cain, K., Oakhill, J., & Bryant, P. (2000). Investigating the causes of reading comprehension failure: The comprehension--age match design. Reading and Writing: An Interdisciplinary Journal, 12, 31-40.

Cantwell, D. P., & Baker, L. (1991). Psychiatric and developmental disorders in children with communication disorder. Washington, DC: American Psychiatric Press.

Chall, J. (1995). Dale Chall Readability--Revisited. The Dale Chall Readability Formula. New York: Brookline Books.

Cohen, M. J. (1997). Children's Memory Scale. San Antonio IX: The Psychological Corporation.

Cohen, N. J., Menna, R., Vallance, D., Barwick, M., Im, N., & Horodetsky, N. (1998). Language, social cognitive processing, and behavioral characteristics of psychiatrically disturbed children with previously identified and unsuspected language impairments. Journal of Child Psychology and Psychiatry, 39(6), 853-864.

Cohen, N., Davine, M., Horodetsky, N., Lipsett, L. & Isaacson, L. (1993). Unsuspected language impairment in psychiatrically disturbed children: Prevalence and language and behavioral characteristics. Journal of the American Academy of Child and Adolescent Psychiatry, 32, 595-603.

Cohen, N.J., Barwick, M., Horodetsky, N., Vallance, D., Im, N. (1998). Language, achievement, and cognitive processing in psychiatrically disturbed children with previously identified and unidentified language impairments. Journal of Child Psychology and Child Psychiatry, 39 865-877.

Cohen, N. J., Vallance, D. D., Barwick, M., Im, N., Menna, R., and Horodezky, N., et al. (2000). The interface between ADHD and language impairment: An examination of language, achievement, and cognitive processing. Journal of Child Psychology and Psychiatry, 43, 353-362.

Conners, C. K. (1997). Canners Rating Scales--Revised. Technical manual. Toronto: Multi-Health Systems.

Daneman, M., & Carpenter, P. (1980). Individual differences in working memory and reading. Journal of Verbal Learning and Verbal Behavior, 19, 450-466.

Douglas, V. I. (1999). Cognitive control processes in attention-deficit hyperactivity disorder. In H. C. Quay & A. E. Hogan (Eds.), Handbook of disruptive behavior disorders (pp. 105-138).New York: Kluwer Academic/Plenum.

Duff, S. J., & Hampson, E. (2001). A sex difference on a novel spatial working memory task in humans. Brain and Cognition, 47, 470-493.

Dunn, L. M., & Dunn, L. M. (1997). The Peabody Picture Vocabulary Test--III (3rd ed.). Circle Pines, MN: American Guidance Service.

Friedman, N. P., & Miyake, A. (2000). Differential roles for visuospatial and verbal working memory in situation model construction. Journal of Experimental Psychology: General, 129, 61-83.

Gathercole, S. E., & Pickering, S. J. (2000). Working memory deficits in children with low achievements in the national curriculum at 7 years of age.British Journal of Educational Psychology, 70, 177-194.

Graesser, A., Singer, M., & Trabasso, T. (1994). Constructing inferences during narrative text comprehension. Psychological Review, 101, 371-395.

Harvey, S., & Goudvis, A. (2000). Strategies that work Teaching comprehension to enhance understanding. York, Maine: Stenhouse.

Johnson, C. J., Beitchman, J. H., Young, A., Escobar, M., Atkinson, L., Wilson, B., et al. (1999). Fourteen year follow-up of children with and without speech/language impairments: Speech--language stability and outcomes. Journal of Speech-Language Hearing Research, 42, 744-760.

Just, M. A., & Carpenter, P. A. (1992). A capacity theory of comprehension: individual difference in working memory. Psychological Review, 99, 122-149.

Karatekin, C., & Asarnow, R. F. (1998). Working memory in childhood onset schizo-phrenia and attention-deficit hyperactivity disorder. Psychiatry Research, 80, 165-176.

Kempton, S., Vance, A., Maruff, E., Luk, E., Costin, C., & Pantelis, C. (1999). Executive function and attention deficit hyperactivity disorder: stimulant medication and better executive function performance in children. Psychological Medicine, 29, 527-538.

Kintsch, W. (1998). Comprehension. A paradigm for cognition. Cambridge, UK: Cambridge University Press.

Lorch, E. P., Milich, R., & Sanchez, R. P. (1998). Story comprehension in children with ADHD. Clinical Child and Family Psychology Review, 1, 163-178.

Lorch, E. P., Milich, R., Sanchez, R. P., van den Brock, P., Baer, S., Hooks, K., et al. (2000). Comprehension of televised stories in boys with attention deficit hyperactivity disorder. Journal of Abnormal Psychology, 109, 321-330.

Markman, E. M. (1979). Realizing that you don't understand. Elementary school children's awareness of inconsistencies. Child Development, 50, 643-655.

Martinussen, R., & McInnes, A. (2001). Visual spatial memory in attention-deficit/hyperactivity disorder: Subtype differences and stimulant effects. Poster presented at the Society for Research in Child Development Biennial Meeting, Minneapolis, Minnesota.

Miyake, A., & Shah, P. (1999). Toward unified theories of working memory. In A. Miyake & P. Shah (Eds.), Models of working memory: Mechanisms of active maintenance and executive control (pp. 442-473). New York: Cambridge University Press.

Nation, K., Adams, J. W., Bowyer-Crane, C. A., & Snowling, M. (1999). Working memory deficits in poor comprehenders reflect underlying language impairments. Journal of Experimental Child Psychology, 73, 139-158.

Oram, J., Fine, J., & Tannock, R. (1999). Assessing the language of children with attention deficit hyperactivity disorder. American Journal of Speech-Language Pathology and Audiology, 8, 72-80.

Palladino, P., Cornoldi, C., de Beni, R., & Pazzaglia, F. (2001). Working memory and updating processes in reading comprehension. Memory and Cognition, 29, 344-354.

Pennington, B. F., & Ozonoff, S. (1996). Executive functions and developmental psychopathology. Journal of Child Psychology and Psychiatry, 37, 51-87.

Petrides, M., & Milner, B. (1982). Deficits in subject-ordered tasks after frontal and temporal lobe lesions in man. Neuropsychologia, 20, 249-262.

Prizant, B. M., Audet, L. R., Burke, G. M., Hummel, L. J., Maher, S. R., & Theadore, G. (1990) Communication disorders and emotional/behavioral disorders in children and adolescents. Journal of Speech Language Hearing Research, 55, 179-192.

Purvis, K., & Tannock, R. (1997). Language abilities in children with attention deficit hyperactivity disorder, reading disabilities, and normal controls. Journal of Abnormal Child Psychology, 25, 133-144.

Sahakian, B. J., & Owen, A. M. (1992). Computerized assessment in neuropsychiatry using CANTAB. Journal of the Royal Society of Medicine, 85, 399-402.

Semel, E. M., Wiig, E. H., & Secord, W. (1995). Clinical evaluation of language fundamentals--III, San Antonio, Texas: The Psychological Corporation, Harcourt, Brace, Jovanovich.

Swanson, H. L. (1999). Reading comprehension and working memory in learning disabled readers: Is the phonological loop more important than the executive system? Journal of Experimental Child Psychology, 72, 1-31.

Tannock, R., & Brown, T. E. (2000). Attention deficit disorders with learning disorders in children and adolescents. In T. E. Brown (Ed.), Attention deficit disorders and comorbidities in children, adolescents, and adults, Washington, DC: American Psychiatric Press.

Tannock, R., Hum, M., Masellis, M., Humphries, T., & Schachar, R. (2000). Children 's Academic Performance. Attention, Behavior, and Learning Evaluation: Teacher Telephone Interview (CAPABLETTI). Presented at the annual conference of American Academy of Child and Adolescent Psychopathology, New York City.

Tannock, R., & Schachar, R. (1996). Executive dysfunction as an underlying mechanism of behavior and language problems in attention deficit hyperactivity disorder. In J. Beitchman, N. Cohen, M. M. Konstantearas, and R. Tannock. (Eds.), Language, learning, and behavour disorders (pp. 128-155). Cambridge: University Press.

Tannock, R., Purvis, K. L., & Schachar, R. (1993). Narrative abilities in children with attention deficit hyperactivity disorder and normal peers. Journal of Abnormal Child Psychology, 21, 103-117.

Tirosh, E., & Cohen, A. (1998). Language deficit with attention-deficit disorder: A prevalent comorbidity. Journal of Child Neurology, 13, 493-497.

Vallance, D. D., Im, N., & Cohen, N.J. (1999). Discourse deficits associated with psychiatric disorders and with language impairments in children. Journal of Child Psychology and Psychiatry. 40, 693-704.

Wagner, R. K., Torgesen, J., & Rashotte, C. A. (1994). Longitudinal studies of phonological processing and reading. Journal of Learning Disabilities, 27, 276-286.

Warr-Leeper, G., Wright, N. A., & Mack, A. (1994). Language disabilities of antisocial boys in residential treatment. Behavior Disorders, 19, 159-169.

Wechsler, D. (1991). The Wechsler Intelligence Scale for Children-- Third edition, San Antonio, TX: The Psychological Corporation.

Westby, C. (1991). Assessing and remediating text comprehension problems. In A. G. Kahmi & H. W. Catts (Eds.), reading disabilities, A developmental language perspective (pp. 199-260). Needham Heights, MA: Allyn and Bacon.

Williams, D., Stott, C. M., Goodyer, I. M., & Sahakian, B. J. (2000). Specific language development with or without hyperactivity: Neuropsychological evidence for frontostriatal dysfunction. Developmental Medicine and Child Neurology, 42, 368-375.

Williams, K. T. (1997). Expressive Vocabulary Test. Circle Pines, MN: American Guidance Service.

Woodcock, R. W. (1987).Woodcock Reading Mastery Tests--Revised, Austin, TX: DLM Teaching Resources.

Ylvisaker, M., & DeBonis, D. (2000). Executive function impairment in adolescence: TBI and ADHD. Topics in Language Disorders, 20, 29-57.

Zentall, S. S. (1988). Production deficiencies in elicited language but not in the spontaneous verbalizations of hyperactive children. Journal of Abnormal Child Psychology, 16, 657-673.

Zentall, S. S. (1993). Research on the educational implications of attention deficit hyperactivity disorder. Exceptional Children, 60, 143-153.

Zwaan, R., & Radvansky, G. A. (1998). Situation models in language comprehension and memory. Psychological Bulletin, 123, 162-185.

Alison Mclnnes, (1) Tom Humphries, (2) Sheilah Hogg-Johnson, (3) Rosemary Tannock (1, 4)

(1.) Brain and Behaviour Research Program, The Hospital for Sick Children, Toronto, Ontario, Canada.

(2.) Child Development Clinic, The Hospital for Sick Children, Toronto, Ontario, Canada.

(3.) Department of Biostatistics, The Institute for Work and Health, Toronto, Ontario, Canada.

(4.) Address all Correspondence to Rosemary Tannock, PhD, Brain and Behaviour Research Program, The Hospital for Sick Children, 555 University Avenue, Toronto, Ontario, Canada, M5S 108; e-mail: tannock@sickkids.on.ca

COPYRIGHT 2003 Plenum Publishing Corporation
COPYRIGHT 2003 Gale Group