European Child & Adolescent Psychiatry 10:5866 (2001) Steinkopff Verlag 2001

ORIGINAL CONTRIBUTION

M. Noterdaeme H. Amorosa K. Mildenberger S. Sitter F. Minow

Evaluation of attention problems in children with autism and children with a specific language disorder

Accepted: 1 September 2000

M. Noterdaeme (&) H. Amorosa r Kinder- und Heckscher Klinik fu Jugendpsychiatrie r teilleistungs- und Abteilung fu rte Kinder verhaltensgesto Wolfratshauser Strasse 350 nchen, Germany 81479 Mu K. Mildenberger S. Sitter r Kinder- und Jugendpsychiatrie Institut fu t der Ludwig-Maximilians-Universita nchen, Germany Mu F. Minow diatrie r Soziale Pa Institut fu t der Ludwig-Maximilians-Universita nchen, Germany Mu

j Abstract Children with autism and children with a specic language disorder show additional attention decits. The literature on the neuropsychological investigation of attention in both groups of children suggests that the nature of their attention problems might be different. The purpose of this study is to examine the attention test proles in these two groups of children with developmental disorders. Nineteen children and adolescents with autism, 17 subjects with a specic language disorder and 19 control subjects participated in the study. Non-verbal intelligence was normal for all subjects. The ``Testbatterie zur fung'' was Aufmerksamkeitspru administered to all subjects. This instrument provides the possibility to examine a wide range of

attention functions and executive functions. The results showed that the autistic individuals had decits in executive functions, whereas the language impaired children had decits in auditory sustained attention, in auditory selective attention, and in the domain of executive functions. It is concluded that although both groups of developmentally impaired subjects showed attention problems, the decits are not the same in both groups. The different neuropsychological proles probably reect different mechanisms in the pathogenesis of the attention deficits in both types of developmental disorders. j Key words Autism language disorder attention neuropsychology

Introduction
Autism and specic developmental disorders of speech and language are both described in chapter F8 ``Disorders of psychological development'' of the international classication scheme. In the introduction to this chapter it is stated that there is an impairment or delay in the development of functions closely related to biological, maturational processes of the central nervous system (33). Attention is a higher cerebral function and both children with autism and children with a specic language disorder have been described as having clear attention problems (4, 5, 1719). Attention decits have a major impact on the

cognitive and the social development; hence, the accurate diagnosis of these problems is of immediate relevance for these children. The actual models on attention emphasise that it is a complex function and suggest that different attention components are represented in distinct, widely distributed but interconnected cortical and subcortical sites, which collectively constitute an integrated network. It is proposed that each cortical/subcortical site represents a more or less specic aspect of attention. Selective attention mechanisms have been attributed to the parietotemporal cortex, executive aspects of attention such as planning, exibility and working memory are considered to be frontal lobe functions, whereas

ECAP 0209

M. Noterdaeme et al. Attention in children with developmental disorders

59

quantitative dimensions of attention such as alertness and the ability to sustain attention over a long period of time are proposed to be functions of subcortical sites (15). Moreover, according to the working memory model of Baddeley, attention is closely related to memory processes through the mechanism of the central executive processor (1). The results of the studies focussing on attention decits in children with autism are rather heterogeneous and there is still no clear picture about the nature of their attention problems. Ornitz suggested that overselective attention mechanisms might be the core decits in children with autism (20). Burack also studied selective attention in autistic children and found that the performance of these children was more impaired by the presence of distracters than the performance of control children (6). Garretson and co-workers examined the ability to sustain attention in a group of autistic children and did not detect signicant differences between autistic and control children on this task (9). Minshew et al. tested children with autism with a broad battery of different neuropsychological attention tests and found no differences between these children and the controls (13, 14). Courchesne and co-workers found that autistic children had considerable problems in rapidly and accurately moving or shifting the focus of attention from one source of information (e.g. auditory) to another (e.g. visual) but were not affected in the ability to simply maintain attention on only a single source. They suggested that the cerebellum might play a part in this attention decit (8). Several studies reported that autistic children had clear problems on tasks tapping executive functions. Pennington and Ozonoff reviewed the literature on executive functions in autism and other developmental psychopathologies (26). The authors concluded that decits in executive functions were consistently found in autistic samples across many studies, using a wide variety of measures with subjects of all ages and functioning levels. Decits in executive functions, however, were also present in ADHD children. The severity and prole of the decits appeared to be different in both groups. There seems to be some evidence that the autistic children are specically impaired on planning and exibility aspects of the executive functions, whereas ADHD children have more inhibition problems (2125). Recently, Russell et al. established evidence that autistic children are especially impaired on executive tasks requiring the subjects to follow novel rules and to make a non-verbal response. Their hypothesis is that children with autism are particularly challenged by executive tasks because they are unlikely to encode rules in a verbal form (27). Studies on language impaired children concentrate on auditory perception and processing as well as on

auditory short-term memory. The studies of Tallal et al. proposed that these children were impaired in perceiving rapid auditory stimuli and that this affected their speech perception. The authors trained language impaired children to associate two easily distinguishable tones with a specic keypress response. Once this association had been learned, sequences of two tones were presented and the subjects had to press the corresponding keys in the correct sequence. The results of these studies point to a clear impairment of these abilities in language impaired children and also make clear that there is an overlap between processing/memory function and attention (2, 12, 28, 29). Few studies report specifically on a more detailed analysis of attention decits. Nicolay examined a group of school-aged language impaired children and age-matched controls with the Matching Familiar Figure Test and a vigilance-type task and found that the language impaired children had clearly more attention problems than the normal control children (16). Campbell stressed the importance of sustained auditory attention as a major factor in the development of receptive language skills (7). Kail reviewed several studies on language impaired children and analysed the reaction time of these children on a variety of verbal as well as on non-verbal tasks. His analysis demonstrated that language impaired children were consistently slower than the unimpaired children on all types of tasks. It was therefore postulated that the decits of these children could not be attributed to a language-specic system, but reect some general component of cognitive processing (10). Weyandt and Willis studied executive function tasks in hyperactive children and language impaired children and found that both groups were equally impaired on these tasks (32). The literature on the neuropsychological investigation of attention decits in children with autism and children with specic language disorders shows that it is not clear which attention functions might be decient in both groups of children. While there is a reasonable number of studies on the neuropsychological assessment of autistic children (6, 8, 9, 13, 14, 2027), there are only very few papers reporting on language impaired children (7, 10, 12, 16, 32). Moreover, most studies use quite different paradigms to investigate a broad range of attention functions in different clinical groups, making it difcult to compare results and to decide whether the nature of the attention decits is the same in both groups of developmentally disordered children. The purpose of this study is to analyse and compare the attention decits in children with autism and children with a specic language disorder on a broad range of attention tasks.

60

European Child & Adolescent Psychiatry, Vol. 10, No. 1 (2001) Steinkopff Verlag 2001 Table 1 Characteristics of the subjects Subjects with autism n = 19 Mean age Min.max. Nonverbal IQ Min.max. Gender boys:girls 13 (SD = 5.0) 721 101 (SD = 13) 85119 16:3 Subjects with specific language disorder n = 17 12.7 (SD = 5.5) 720 104 (SD = 10) 85121 16:1 Comparison group n = 19 13 (SD = 4.5) 720.6 105 (SD = 7) 93123 16:3

Methods
j Subjects
Nineteen children with infantile autism, 17 children with a specic developmental language disorder and 19 comparison subjects participated in the study. All subjects had been diagnosed individually by two child psychiatrists with long-standing experience in the eld of pervasive and specic developmental disorders. The language impaired children and the autistic children were identied through the special unit for children with developmental disorders at the Heckscher Clinic in Munich. Most of the subjects with autism and with a language disorder had been seen regularly, for many years, at the polyclinic of the unit. They were asked to participate in this study. The subjects with autism had to have a normal IQ to participate in the study. The diagnosis of autism had been established according to ICD-10 criteria. The language impaired children were selected to match the autistic sample as closely as possible with respect to age and non-verbal intelligence. To achieve this goal, the language impaired children were recruited on the basis of their diagnoses at the initial referral and assessment (age range 68). At that time, the language impaired children fullled the ICD-10 criteria for a specic language disorder (one and a half standard deviation discrepancy between the measure of the language tests and the measure on non-verbal intelligence tests). All of these children had been diagnosed by experienced speech pathologists, had been treated by speech pathologists and received special education based on their language dysfunction. Some of the older language impaired children were no longer receiving active speech services at the time of this study, but most of them still had speech therapy. The comparison group was recruited to match the clinical groups as closely as possible. The subjects were assigned to the three different groups by the two experienced child psychiatrists. All subjects had a non-verbal IQ above 85 (measured with the Kaufman Assessment Battery for r children, the Hamburg-Wechsler-Intelligenztest fu Kinder (HAWIK-R) or the Hamburg-Wechsler-Intel r Erwachsene (HAWIE-R) (11, 30, 31)). ligenztest fu The autistic children, the language impaired children and the comparison group were matched for age, sex and non-verbal IQ. Statistical analysis showed no signicant age or intelligence differences between the three groups. The characteristics of the sample are summarised in Table 1.

j Procedure
Several tests of the ``Testbatterie zur Aufmerksam fung'' were used to evaluate the attention keitspru performances of the subjects (34). A total of nine different subtests were used to evaluate the following functions: simple reaction time (``Alertness''), selective visual attention (``Quadrate''), selective ne''), sustained visual attenauditory attention (``To tion (``Vigilanz optisch''), sustained auditory attention (``Vigilanz akustisch''), incompatibility t''), Go/Nogo (``Eins aus zwei Be(``Inkompatibilita dingungen''), shift of attention (``Reaktionswechsel''), searching a visual matrix (``Scanning''). All subtests consist of non-verbal materials and responses are made through keypresses. The subjects were tested individually in a quiet room. Completing the test battery took two to three sessions, depending on the age of the subject. The alertness task is a simple reaction time test in which subjects have to press a key when a visual target (a cross in x-form) is presented on the screen. The target is presented in four blocks of 20 targets each. The measures used are the reaction time (median) and the number of omission errors. Simple reaction time tasks are considered to be a measure for general processing capacities. The selective visual attention task is a visuo-spatial task. The computer screen shows a 16- point conguration organised in a 4 4 square. Small crosses appear in different locations on the 16- point grid. The subject has to respond selectively whenever a specied pattern of crosses (four crosses arranged in a square) appears on the screen. The test consists of 100 presentations. The measures used are the reaction time of the correct responses (median) and the number of omission errors. In the selective auditory attention task a low (440 Hz) and a high tone (1000 Hz) are presented in an alternating, serial sequence of 200 tones. Whenever an irregularity is detected (two consecutive high tones or two consecutive low tones), the subject has to press a key. The measures used are the reaction time

M. Noterdaeme et al. Attention in children with developmental disorders

61

of the correct responses (median) and the number of omission errors. The sustained visual attention task lasts ten minutes. At the center of the computer screen a square, divided into a lower and an upper part is presented. A pattern of dots jumps alternatively between the lower and the upper square. Irregularities in the serial, alternating pattern have to be detected (dots blinking twice in the upper or lower square). The measures are the differences between the last ve minutes and the rst ve minutes of the test, in reaction time of the correct responses (median) and the differences in the rate of omission errors. The sustained auditory attention task lasts ten minutes. It is identical to the selective auditory attention tasks, except for the duration of the tasks. As in the sustained visual attention tasks, the measures are the differences between the last ve minutes and the rst ve minutes of the test, in reaction time of the correct responses (median) and the omission error rate. The selective and the sustained auditory attention tasks, as well as the visual sustained attention tasks require a continuous working memory capacity for successful performance. A correct reaction in these tasks depends on the ability to keep in mind and to compare two consecutive stimuli. Two consecutively identical stimuli is the target conguration, nonidentical stimuli should not prompt a reaction. The incompatibility task is constructed to check for interference (inhibition of prepotent responses). Decits in this test should be seen in patients with frontal lesions. A central xation point appears on the computer screen. Arrows pointing to the left or to the right are presented left or right of the central xation point. The subject is instructed to press the right key whenever the arrow points to the right, and the left key when the arrow points to the left, independent of the location of the arrow with respect to the central xation point. The test comprises 60 presentations (30 compatible and 30 incompatible presentations). Measures are the reaction time of the correct responses for all presentations (median) and the number of errors. The Go/Nogo task is designed to test the ability to inhibit reactions on irrelevant stimuli. According to the authors, performance on this test is impaired in patients with prefrontal lesions. Two stimuli are presented (``'' and ``+''). The subjects are instructed to react on the ``'' and not to press the key when the ``+'' is presented. The test comprises 40 stimuli presentations, 20 targets, 20 non-targets. Measures are the reaction time of the correct responses (median) and the number of commission errors. In the shift of attention task the subject has to continuously shift the focus of attention between two

categories of geometrical symbols (round and angular forms). Each presentation contains a variation of both geometrical categories and the test consists of 100 consecutive presentations. Subjects have to switch from one type of geometrical category to the other after each presentation, beginning with an angular symbol as a target in the rst presentation, followed by a round symbol in the next presentation. According to the authors, the test is a measure for exibility and shift of attention capacity. Performance on the test is impaired in patients with frontal lesions. Measures are the reaction time of the correct responses (median) and the number of switching errors. The visual scanning task is a searching task. Subjects have to identify a target symbol embedded in a matrix of 25 symbols. The symbols are squares with one open side, the target symbol being the square with the upper side open. The symbols are arranged in ve rows and columns of ve symbols each. The scanning tasks comprises 100 consecutively presented matrices which have to be searched according to a given rule (searching from left to right in each row, starting with the upper row). Subjects have to press one key if they detect the target in the matrix and press another key if they do not identify the target. According to the authors, this tasks is a complex visual task, tapping different resources. Performance on the task can be impaired through decits in sustained attention or planning decits. Patients with frontal lesions have problems with the successful completion of this task. The target symbol is present in 50 matrices and absent in the remaining 50 matrices. Measures are the median of the searching time for targets and the number of omission errors made in the search. The test battery was developed by Zimmermann and Fimm (34). Norms were established for adults (age range 20 to 69). The subtests Go/Nogo, shift of attention, scanning and incompatibility are theoretically constructed to tap executive functions and proved to be sensitive to frontal lesions. Therefore, for the present study they were classied as being representative of the executive function domain. The remaining tasks were classied as attention tasks.

j Statistical analysis
Not all variables used fullled the criteria of normal distribution and homogeneity of variance. Therefore, group differences were calculated with the KruskalWallis one-way analysis of variance by ranks. To establish the signicant differences between the groups, we used post-hoc tests and calculated the critical difference between the mean group ranks (3).

62

European Child & Adolescent Psychiatry, Vol. 10, No. 1 (2001) Steinkopff Verlag 2001

Within group differences between the auditory and the visual modality were calculated with the Wilcoxon-test.

Results
The results are summarised in Tables 2 and 3.

j Attention functions
The tests alertness, sustained auditory and visual attention as well as selective auditory and visual attention are representative of attention functions. Statistical analysis showed that the language impaired and the autistic children did not differ from the comparison children on the alertness test. The language impaired were signicantly slower and made more errors than the comparison children and the autistic children on the test of sustained auditory attention. There were no signicant group

differences on the sustained visual attention. However, the language impaired had a signicantly higher overall error rate on this task than the autistic children or the comparison children (median error rate language impaired children: 16, median error rate autistic children: 5, median error rate comparison children: 5). As the errors were evenly distributed between the rst half and the second half of the sustained visual attention task for all groups, there were no signicant group differences on the subtraction score. Within the language impaired group, the ability to maintain a constant level of attention over time was impaired for the auditory modality only (Wilcoxon test, p < 0.0001). The two other groups did not show signicant differences between the two modalities (autistic group: Wilcoxon test p < 0.7; comparison group: Wilcoxon test, p < 0.8). The language impaired children also made more errors than the two other groups on the auditory selective attention task, whereas no statistical significant group differences emerged on the selective

Table 2 Median of reaction times (in ms) on the attention and the executive function tasks Attention tasks Alertness Sustained attention* Auditory Autism Specific language disorder Comparison group Chi-square value (Kruskal-Wallis) p 258 255 255 0.43 0.8 ns 19 61 16 7.3 0.02 Visual 52 36 19 0.05 0.9 ns Selective attention Auditory 520 603 446 1.8 0.4 ns Visual 1047 961 886 3.2 0.2 ns 4419 3284 2614 5.6 0.05 432 469 446 0.09 0.9 ns Executive function tasks Scanning Go/Nogo Interference 603 477 402 7.5 0.02 Shift of attention 1312 1117 920 9.2 0.009

* The measures for the sustained attention tasks refer to the differences in the median of the reaction time between the last five minutes and the first five minutes of the task Table 3 Median of error rates on the attention and the executive function tasks Attention tasks Alertness Sustained attention* Auditory Autism Specific language disorder Comparison group Chi-square (Kruskal-Wallis) p 1 1 1 1.8 0.3 ns 1 6 1 15.2 0.005 Visual 0 0 0 3.8 0.1 ns Selective attention Auditory 1 5 1 6.3 0.05 Visual 4 5 3 1.5 0.4 ns 5 8 5 2.3 0.3 ns 3 5 3 4 0.1 ns Executive function tasks Scanning Go/Nogo Interference 7 7 6 2 0.3 ns Shift of attention 15 10 3 19.7 0.0001

* The measures for the sustained attention tasks refer to the differences in the median of the error rates between the last five minutes and the first five minutes of the task

M. Noterdaeme et al. Attention in children with developmental disorders

63

visual attention task. Within the language impaired group, there was no signicant difference between the auditory and the visual selective attention task (Wilcoxon test, p < 0.3). The two other groups had a better performance on the auditory task than on the visual selective attention task (autistic group: Wilcoxon test p < 0.01; comparison group: Wilcoxon test, p < 0.01). The post-hoc analysis showed that the autistic children did not differ from the comparison children on any task or measure.

j Executive functions
The tests scanning, Go/Nogo, interference and shift of attention represent more complex functions and are categorised as executive functions. Reaction times and number of errors are shown in Tables 2 and 3. Statistical analysis showed that the language impaired children and the autistic children differed signicantly from the comparison group on the shift of attention task (reaction time as well as errors). Posthoc analysis showed that there were no statistically signicant differences between the two clinical groups on this task. The autistic children differed signicantly from the comparison group on the interference task and the scanning task (only on the reaction time measure; the error rate did not differ signicantly between the three groups). Post-hoc analysis showed no signicant differences between the language impaired children and the comparison group. The Go/Nogo task did not show signicant statistical differences between the groups.

Discussion
Children with a specic language disorder and children with autism share a number of characteristics. Both groups of children have poor social skills and are delayed in the acquisition of functional language and communication skills. In addition, these children often have decits in attention and executive functions. In the last few years, there has been an increasing number of studies devoted to the analysis of the neurocognitive decits associated with certain types of developmental disorders. Whereas a substantial amount of research on these topics has been devoted to autistic individuals, relatively little attention has been paid to the analysis of these functions in language impaired children. The type of decit in these children is not clear, and hence the present study focuses on this issue.

The results of this study show that both groups of subjects reach a normal functional level on certain tasks and have decits in different domains. Simple reaction time tasks (alertness) are considered to be a global measure for information processing capacity. This measure shows no signicant difference between both clinical groups and the comparison group. As a matter of fact, the three groups perform equally well on all measures of this task: reaction time and standard deviation are almost identical and neither group shows an increased rate of omission errors. Kail analysed reaction time studies in language impaired children and found that these children were slower than control children on verbal as well as on non-verbal tasks. He concluded that the decit could not be explained through a specic problem in the language system, but by a more general problem in information processing (10). All the tasks of this study are non-verbal and it is therefore not possible with these data to compare the subjects' performance on these two types of tasks. On the basis of the simple non-verbal tasks used, we can, however, not support the hypothesis of a decit of generally slower information processing. The performance of the autistic subjects on sustained attention tasks and on selective attention tasks is again comparable to the performance of the comparison group on all measures of these paradigms in both the visual and the auditory modality. For the language impaired group, clear decits can be demonstrated in the auditory attention functions, whereas there seems to be no problem in the visual modality. So the question arises as to how specic the decits in the auditory modality are for the language impaired group. A closer look at the results and at the construction of the auditory and visual tasks indicates that what might be specic for the language impaired subjects is not so much the distinction between auditory and visual modality but a much closer interaction between attention and working memory on certain types of tasks. The task used for the evaluation of auditory attention (selective attention as well as sustained attention) consists of a serial, alternating presentation of two different acoustic stimuli. Every once in a while the alternating pattern is interrupted and two identical stimuli are presented one after the other. This sequence of two identical stimuli is dened as the target. Correct reactions on this task continuously require working memory skills, as each stimulus has to be stored in memory long enough to be compared with the next stimulus. The two visual tasks put a quite different load on working memory. Whereas the visual sustained attention paradigm is analogous to the auditory task and thus also taps working memory skills, the

64

European Child & Adolescent Psychiatry, Vol. 10, No. 1 (2001) Steinkopff Verlag 2001

selective attention task requires the identication of one denite spatial pattern with minimal load on memory skills. An analysis of the results in the visual modality indicates that the language impaired children only perform as well as the other groups on the selective visuo-spatial attention task. In the sustained visual attention task (requiring working memory), the language impaired subjects produce a considerably higher overall number of errors than the autistic subjects and the comparison group. The apparent intact performance in the visual sustained attention task is due to a lack of a signicant decrease in performance between the rst and the second test half in this group, but the overall performance level is much worse for the language impaired group than for the two other groups. We interpret these ndings as an indication that in language impaired children selective attention and sustained attention are particularly impaired when the tasks require working memory skills and that this is at least partly independent of the modality. There is some support for this hypothesis in the literature. Several studies have reported decits in auditory processing and in memory in language impaired children (12, 28, 29). The study of Lincoln et al. (12) might be particularly relevant for the understanding of our results. These authors examined a group of older language impaired adolescents (mean age = 17.7) and autistic subjects (mean age = 24.4) with approximately the same paradigm as Tallal and co-workers (rst phase of the experiment: association between tones and specic keypresses; second phase of the experiment: response to increasingly longer sequences of two tones). Their results showed that the autistic children had no difculties at all with the auditory processing tasks, whereas the language impaired children showed the expected problems. The data also demonstrated that the auditory processing decits were closely linked to deciencies in short-term auditory memory. The difculties involved decits for tone sequences presented rapidly or relatively slowly and the difculties increased with the length of the tone sequences. According to the models on attention, selective attention is more a function of the dorsolateral parietal cortex, whereas working memory skills are considered to be functions of the prefrontal cortex. Sustained attention is at least partly relying on subcortical structures. Attention decits in language impaired children might be due to problems in the connection between the posterior and the anterior attention network and the subcortical sites relevant for attention control (15). In this study, decits in the executive functions are present in both clinical groups. As a matter of fact, although the autistic children perform a bit worse

than the language impaired children on these tasks, there is really no signicant difference between the two groups. The decit in the executive domain in autistic children has been widely documented by numerous studies (2026). Our autistic subjects show decits on tasks requiring exibility and planning. The two tasks focussing on the inhibitory capacity only show minor signicant differences between the clinical groups and the controls, although there is an indication that the autistic individuals needed more time to achieve a comparable level of performance as measured by the number of errors. This result is in accordance with other studies which found that executive functions are not evenly impaired in autistic children: exibility/planning decits seem to be prominent in these children, whereas inhibition problems were more pronounced in ADHD children (21, 26). The decits in the executive domain of our language impaired group are also in accordance with literature ndings. As a matter of fact, there are only very few studies concentrating on executive functions in this group of children (32). The similarity between the autistic children and the language impaired children on the shift of attention task is striking and can be interpreted in different ways. First, it may be an indication that the language impaired children and the autistic children share a decit in planning and exibility and thus in certain functions attributed to the frontal lobes. Second, the possibility should be considered that the apparent similarity in the results could be produced through the involvement of different mechanisms. Observation during task performance shows that the language impaired children have problems in naming and labelling visual material: they clearly use the strategy, but they are slow and not efcient in using it. As this is an important strategy, especially for solving the task we used in this study, it might be hypothesised that the problems for the language impaired group could rather be a result of difculties in naming and labelling strategies than an inherent exibility problem. Naming/labelling strategy was not overtly observed in the autistic group, and therefore it might be possible that these children do not have the idea to use this as a strategy at all and solve the tasks through entirely different processes. Some evidence for the use of other strategies in the autistic group can be inferred from the interviews we did with all children after each session. We asked them about difculties and strategies they used in different tasks. None of the autistic children reported the use of naming/labelling as a useful strategy, whereas most of the language impaired children did. The strategy was not made explicit to the subjects and it was up to them to discover the most efcient way to solve the

M. Noterdaeme et al. Attention in children with developmental disorders

65

tasks. In an interesting study on executive functions in autism Russell et al. (27) reported that the problems in the executive domain were most prominent when novel rules had to be applied and when the response was non-verbal. They hypothesised that children with autism are challenged by executive functions, because they are unlikely to encode rules verbally. Our results offer some support for this idea; especially the shift of attention tasks involved the application of a novel rule and responses were nonverbal keypresses. This study was undertaken as a rst attempt to evaluate a wide range of attention and executive functions in two groups of developmentally impaired children. The answer as to which mechanisms are involved in both groups requires a more detailed analysis of the executive functions.

Summary
Our results show that both groups of children have a different attention test prole. Autistic children do not show any decits on tasks tapping sustained attention, selective attention or general alertness. Their performance on these tasks is comparable with the performance of the normally developing children in this study. Language impaired children have specic problems on attention functions which also require some working memory skills, indicating a close link between these two neurocognitive domains. The decits in the executive functions are found in both groups, and hence they do not discriminate between the two groups of developmentally impaired children.

References
1. Baddeley AD (1995) Working memory. In: Gazzaniga MS (ed) The Cognitive Neurosciences. MIT Press, Cambridge, pp 755763 2. Barthlen-Weis M, Breuer-Schaumann A rungen in (1994) Sprachentwicklungssto Kombination mit anderen Teil rungen. In: Martinius J, leistungssto runAmorosa H (eds) Teilleistungssto nchen, pp 6772 gen. Quintessenz, Mu 3. Bauer F (1986) Datenanalyse mit SPSS. Springer Verlag, Berlin 4. Beitchman JH, Wilson B, Brownlie EB, Walters H, Inglis A, Lancee W (1996) Long-term Consistency in Speech/language proles: I. Developmental and academic Outcomes. Journal of the American Academy of Child Psychiatry 35:804814 5. Beitchman JH, Wilson B, Brownlie EB, Walters H, Inglis A, Lancee W (1996) Long-term Consistency in Speech/language proles: II. Behavioral, emotional, and Social Outcomes. Journal of the American Academy of Child Psychiatry 35:815825 6. Burack JA (1994) Selective attention decits in persons with autism: preliminary evidence of an inefcient attention lens. Journal of Abnormal Psychology 103:535543 7. Campbell TF, McNeil (1985) Effects of presentation rate and divided attention on auditory comprehension in children with an acquired language disorder. Journal of Speech and Hearing Research 28:513520 8. Courchesne E, Townsend J, Akshoomoff N, Yeung L, Courchesne R, Press G, Murakami J, Lincoln A, James H, Saitoh O, Haas R, Schreibman L, Lau L (1992) A new nding in autism: impairment in shifting attention. In: Broman S, Grafman J (eds) Atypical Cognitive Decits in Developmental Disorders: Implications for Brain Function. Erlbaum, Hillsdale, NJ, pp 101138 Garretson HB, Fein D, Waterhouse L (1990) Sustained attention in children with autism. Journal of Autism and Developmental Disorders 20:101104 Kail R (1994) A method for studying the generalized slowing hypothesis in children with specic language impairment. Journal of Hearing and Speech Research 37:418421 Kaufman AS, Kaufman NL (1991) Kaufman Assessment Battery for Children (K-ABC) (Deutschsprachige Fassung von Melchers und Preu). Hogrefe, ttingen Go Lincoln AJ, Dickstein P, Courchesne E, Elmasian R, Tallal P (1992) Auditory processing in non-retarded adolescents and adults with developmental receptive language disorder and autism. Brain and Language 43:613622 Minshew NJ, Goldstein G, Muenz LR, Payton JB (1992) Neuropsychological functioning in nonmentally retarded autistic individuals. Journal of Clinical and Experimental Neuropsychology 14:749761 Minshew NJ, Goldstein G, Siegel DJ (1997) Neuropsychological functioning in autism: prole of a complex information processing disorder. Journal of the International Neuropsychological Society 3:303316 Mirsky AF (1996) Disorders of attention: a neuropsychological perspective. In: Lyon GR, Krasnegor NA (eds) Attention Memory and Executive Func16. tion. Paul H. Brooks Publishing, Baltimore, pp 71159 Nicolay L (1995) Aufmerksamkeits-, Programm- und Handlungssteuerung rten Erstkla lern. Der von sprachgesto dagoge 26:116 Sprachheilpa Noterdaeme M, Amorosa H (1999) Evaluation of emotional and behavioral problems in language impaired children. European of Child and Adolescent Psychiatry 8:7177 Noterdaeme M, Amorosa H (1998) lligkeiten bei sprachentVerhaltensauffa rten Kindern: die Child wicklungsgesto Behavior Checklist als Screening-In r Kinderstrument. Monatsschrift fu heilkunde 146:931937 Noterdaeme M, Minow F, Amorosa H (1999) Anwendbarkeit der Child Behavior Checklist in der Differentialdi rter Kinder. agnostik entwicklungsgesto r Kinder- und JugendpsyZeitschrift fu chiatrie und Psychotherapie 27:183188 Ornitz EM (1988) Autism: a disorder of directed attention. Brain Dysfunction 1:309322 Ozonoff S, Jensen J (1999) Brief report: specic executive function prole in three neurodevelopmental disorders. Journal of Autism and Developmental Disorders 29:171177 Ozonoff S, McEvoy RE (1994) A longitudinal study of executive function and theory of mind development in autism. Development and Psychopathology 6:415431 Ozonoff S, Strayer DL (1997) Inhibitory function in nonretarded children with autism. Journal of Autism and Developmental Disorders 27:5965 Ozonoff S, Pennington BF, Rogers SJ (1991) Executive function decits in

9.

17.

10.

18.

11.

19.

12.

20. 21.

13.

14.

22.

15.

23.

24.

66

European Child & Adolescent Psychiatry, Vol. 10, No. 1 (2001) Steinkopff Verlag 2001 with autism: Implications for the core executive dysfunctions in the disorder. Journal of Autism and Developmental Disorders 29:103112 28. Tallal P, Newcombe F (1978) Impairment of auditory perception and language comprehension in dysphasia. Brain and Language 5:1324 29. Tallal P, Stark R, Kallman C, Mellitts D (1980) Developmental processing decits and verbal processing. Neuropsychologia 18:273284 30. Tewes U (ed) (1983) Hamburg-Wechs r Kinder (HAWIK-R). ler-Intelligenztest fu Revision 1983. Huber Verlag, Stuttgart 31. Tewes U (ed) (1991) Hamburg-Wechsr ler-Intelligenztest fu Erwachsene (HAWIE-R). Huber Verlag, Stuttgart 32. Weyandt LL, Willis WG (1994) Executive functions in school-aged children: potentional efcacy of tasks in discriminating clinical groups. Developmental Neuropsychology 10:2738 33. World Health Organisation (1991) ICD10 Internationale Klassikation psychi rungen-ICD 10, Kapitel V (F). scher Sto Dilling H, Mombour W, Schmidt MH (eds), Huber, Bern 34. Zimmermann P, Fimm B (1993) Test fung batterie zur Aufmerksamkeitspru (TAP). Version 1.0. Handbuch Teil 1. rselen Psytest, Wu

high-functioning autistic individuals: relationship to theory of mind. Journal of Child Psychology and Psychiatry 32:10811105 25. Ozonoff S, Strayer DL, McMahon WM, Filloux F (1994) Executive function abilities in autism and Tourette syndrome: an information processing approach. Journal of Child Psychology and Psychiatry 35:1015 1032 26. Pennington B, Ozonoff S (1996) Executive functions and developmental psychopathology. Journal of Child Psychiatry and Psychology 37:5187 27. Russell J, Jarrold C, Hood B (1999) Two intact executive capacities in children