Journal of Autism and Developmental Disorders, Vol. 31, No. 4, 2001

Attentional Processes in Autism Gerald Goldstein,1,2,3 Cynthia R. Johnson,2 and Nancy J. Minshew2

Attentional processes in individuals with high-functioning autism were compared with a matched control group. Participants for the study were 103 children and adults with autism and 103 control subjects. Measures administered corresponded to Mirsky et al.’s (1991) factor analysis of tests of attention. Diminished performance was noted on measures that loaded on the Focus-Execute and Shift factors, but not on the Sustain and Encode factors. For tests in which psychomotor speed was used as the score, and the difference between groups was significant, covariance analyses were performed, using tests of basic motor functions as covariates. This procedure led to attenuation to the point of nonsignificant differences in the case of some of the attention tests. Thus, this comprehensive analysis of attention in individuals with highfunctioning autism only found differences on measures in which the task placed demands on cognitive flexibility or psychomotor speed. Thus, purported attention deficits in autism may actually be primary deficits in complex decision making or psychomotor abilities. KEY WORDS: Autism; attentional processes.

INTRODUCTION

the definition of attention itself. Various attentional components and dimensions have been described in the literature. Deficits in the arousal, orienting, filtering, and gazing components of attention have been considered. One popular early model attempted to explain the propensity to engage in repetitive, stereotypical movements, respond in an atypical manner to the environment, and the failure to develop socially as being the result of an underlying deficit in the modulation of arousal (Dawson & Levy, 1989; Hutt, Hutt, Lee, & Ounsted, 1964; Ornitz & Ritvo, 1968). While not clearly supported and abandoned in recent times, both hyperarousal and hypoarousal were earlier asserted to be underlying deficits in autism (Barry & James, 1988; Hutt et al., 1964; Rimland, 1964). Another earlier, often-cited attentional model described the deficit as one of overselectivity, to explain the clinically observed intense focus on detail and failure to interpret multiple cues in the environment (Lovaas, Koegel, Schriebman, 1979; Pierce, Glad, & Schriebman, 1997). Rincover and Ducharme (1987) argued this apparent overselectivity was secondary to what they described as tunnel vision or overly selective gaze. This particular finding later was demonstrated to

Attentional processes in autism have been of interest for several decades. Attentional dysfunction has been implicated as a core deficit in this disorder as a result of the essential role attention plays in information processing as well as the clinical observations of the atypical responses of individuals with autism to their environment. Early neurobehavioral models generally postulated attentional dysfunction as originating at the sensory-perceptual or reflexive level of processing. More recent models have presumed the apparent attention deficits to be at the executive function or conceptual level. Despite the extended interest in the area of attentional processing in autism, there is little consensus on the nature of intact attentional abilities and deficits. This is in part due to a lack of consensus on 1

Veterans Administration Pittsburgh Healthcare System, Pittsburgh, Pennsylvania. 2 University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania. 3 Address all correspondence to Gerald Goldstein, Veterans Administration Pittsburgh Healthcare System, 7180 Highland Drive (151R), Pittsburgh, Pennsylvania 15206; e-mail:[email protected]

433 0162-3257/01/0800-0433$19.50/0 © 2001 Plenum Publishing Corporation

434 be related to developmental level and failed to be replicated with higher functioning and older individuals with autism (Burack, Enns, & Johannes, 1997; Enns & Akhtar, 1989; Pasto & Burack, 1995). Other investigators, in contrast, explored the possibility that attentional processes in autism were impaired as a result of an inability to filter out irrelevant stimuli (Bryson, Wainwright-Sharp & Smith, 1990; Burack, 1994). Recently, in a small sample of high-functioning individuals, deficits in reflexive visual orienting were reported (Casey, Gordon, Mannheim, & Rumsey, 1993). This finding was clinically appealing given the behavioral observations of individuals with autism described above and with perseveration on attending to irrelevant environmental stimuli. It has however failed to be replicated. However, Minshew, Luna, and Sweeney (1999) directly compared reflexive and volitional aspects of attention using rigorous laboratory eye movement procedures; no deficits were found in reflexive orienting but deficits were documented in volitional tasks dependent on frontal lobe. Burack and Iarocci (1995) found differences among a small number of lower functioning autistics in comparison with matched controls on a similar visual task but follow-up studies demonstrated that the abnormality in orienting was related to the informationprocessing demands of the directional cue (Burack et al., 1997). Two other studies reported results suggesting no overall visual orienting dysfunction in a small sample of high-functioning adolescents and adults, but did see deficits in disengaging and shifting attention as increasing processing demands were present in the task or when symbolic information was necessary for a task (Wainwright-Sharp & Bryson, 1993, 1996). Taken together, there are currently no consistent data to support arousal deficits, overfocusing, overselective gaze, poor filtering, or orienting deficits in individuals with autism. The possibility of deficits in sustained attention or vigilance as another attentional component in autism is compelling given the often-observed perseverative behaviors. In a review, Burack et al. (1997) reasoned that individuals with autism may have enhanced capabilities for sustaining attention, at least when self-determined. This component of attention is also of interest given the overlap with other developmental disorders. Sustained attention is the most commonly assessed component in individuals with Attention-deficit Hyperactivity Disorder (Yeates & Taylor, 1998) and is most often measured using continuous performance tasks. The studies including individuals with autism, while small in number, suggest no differences in visual sustained attention (Casey et al., 1993). Casey et al. provided limited evidence that autistic individuals perform

Goldstein, Johnson, and Minshew poorly on auditory continuous performance tasks in comparison to adult controls. In a more recent study, no differences on a visual continuous performance test and other indices of attention components under reflexive control were found (Pascualvaca, Fantie, Papgeorgiou, & Mirsky, 1998). Hence, no consistently observed deficits in sustained attention have yet been demonstrated. Further, impairment in shifting attention in autism has been proposed as a core deficit. A deficit in the ability to shift attention to extrapersonal space was offered as an explanation of the lack of interest in people (Ornitz, 1988; Townsend & Courchesne, 1994) and proposed to be related to parietal lobe dysfunction. Courchesne et al. (1994) have also hypothesized that individuals with autism are deficient in the ability to shift from one stimulus to another, to shift between modalities, and to disengage attention. This dysfunction was demonstrated on modality shift and reactiontime tasks involving substantial executive function ability, although they proposed a cerebellar origin (Courchesne et al., 1993, 1994; Courchesne, Akshoomoff, Townsend, & Saitoh, 1995). A series of other studies have explored the nature of the apparent deficit in shifting attention using the Wisconsin Card Sorting Test (WCST; Ozonoff, 1995a; Ozonoff, Pennington, & Rogers, 1991). Pascualvaca et al. (1998) concluded that the deficit in attention was at the level of executive control originating from frontal lobe dysfunction. This hypothesis was also supported by the saccadic eye movement findings of Minshew et al. (1999) who reported deficits only on volitional saccade tasks and none on tasks involving reflexive attentional abilities. The present study was designed to further examine aspects of attention relevant to several of the hypotheses about attention in autism. To provide a comprehensive assessment of attention and to clearly define its components, an empirically derived four-factor neuropsychological model was applied (Mirsky, Anthony, Duncan, Ahearn, & Kellam, 1991). This model was based upon a series of distinctions among types of attention originally formulated by Zubin (1975). In this model, attention is subdivided into the ability (a) to focus on a target object and perform a task in the presence of distracting objects, (b) to maintain vigilance over a sustained time period, (c) to adaptively shift focus of attention, and (d) to efficiently receive and interpret incoming information. A number of established procedures, that are now commonly used as clinical neuropsychological tests, were used in the Mirsky group factor analysis, each one having a substantial loading on one of the four factors extracted. The fac-

Attentional Processes in Autism tors derived from this model were termed FocusExecute, Sustain, Shift, and Encode. Focus-Execute received high loadings from the WAIS-R or WISC-R Digit Symbol/Coding, Letter Cancellation, Trail Making, and Stroop tests. A Continuous Performance Task (CPT) was the only test producing a high loading on the Sustain factor. Categories achieved, number correct, and error scores from the Wisconsin Card Sorting Test loaded on the Shift factor. The Encode factor received high loadings from the WAIS-R/WISC-R Digit Span and Arithmetic tests. The model was applied in the present study in order to clarify what components of attention might be intact and impaired in autism. An additional consideration also evaluated here is that many tests of attention utilize speed of performance as the response measure. Such tests load in the Mirsky group factor analysis on the Focus-Execute factor. Indeed, Mirsky et al. (1991) characterized that factor as reflecting perceptual-motor speed. It has been well established that individuals with autism commonly have psychomotor deficits in the form of slowness, awkwardness, or poor coordination (Bauman, 1992; Gillberg & Coleman, 1992; Hughes, 1996; Minshew, Goldstein, & Siegel, 1997; Rapin, 1997; Smith & Bryson, 1994). Thus, the direct interpretation of tests involving speed scores as measures of attention is confounded in autism by motor impairment, and it is necessary to account for the influence of that impairment on performance outcome. Based upon all of these considerations, it was hypothesized that attention tasks would be performed poorly by high-functioning individuals with autism only when such tasks make demands on complex decision making or psychomotor speed. METHOD Participants The participants for this study were 103 high-functioning (FS and VIQ ⱖ 70) individuals with autism and a comparison group of 103 medically healthy and neuropsychiatrically normal subjects. The diagnosis of autism was established through expert clinical evaluation in accordance with accepted clinical descriptions of high-functioning autism and two widely used research instruments, the Autism Diagnostic Interview (ADI or ADI-R; Le Couteur et al., 1989; Lord, Rutter, & Le Couteur, 1994) and the Autism Diagnostic Observation Schedule (ADOS; Lord, Rutter, & Goode, 1989). Examiner reliability in the administration and scoring of these instruments was established through ongoing training and consultation with developers of

435 these instruments. Potential autistic subjects were excluded if found to have an associated neurological, genetic, infectious, or metabolic disorder such as tuberous sclerosis, fragile X syndrome, or fetal cytomegalovirus infection. Participants with autism were comparable in age and general intelligence to those in recent shifting attention and executive function studies (Ozonoff, Strayer, McMahon, & Filloux, 1994), and to the earlier studies of Rumsey and Hamburger (1988) and Lord et al. (1989). Particular care was taken to exclude subjects with Asperger disorder by requiring ADI and clinical interview documentation of delayed and disordered language development typical of autism. Exclusion criteria for control subjects were a current or past history of neurological or psychiatric disorder, a developmental cognitive disorder, learning disability, poor school attendance, or family history of developmental cognitive disorders, autism, or neuropsychiatric disorder. All subjects with autism and controls obtained WISC-R (n ⫽ 114) or WAIS-R (n ⫽ 92) (Wechsler, 1974, 1981) Full-scale IQ and Verbal and Performance IQ scores of 70 or above. Demographic data are presented in Table I. There were no significant differences between the autism and control groups with respect to age, gender distribution, educational level, and IQ levels. Attention Battery Procedures A battery consisting of tests highly comparable or identical to those used in the Mirsky group factor analysis was administered by trained technicians closely supervised by an experienced neuropsychologist. These tests assess the various components of attention as contained in the Mirsky model (Mirsky et al., 1991) including encoding, focusing, sustaining, and shifting of attention. Encode. Digit Span and Arithmetic subtests of the WAIS-R or WISC-R. The scaled scores were used as dependent measures. Table I. Demographic Data for the Autism and Control Groups

Age Years of education Socioeconomic status Verbal IQ Performance IQ Full-scale IQ % Female

Autism group

Control group

M

SD

M

SD

18.15 8.27 3.42 99.60 95.65 97.57 13.6

10.14 4.57 1.49 17.03 12.82 14.73

18.96 9.33 3.70 99.94 98.60 99.12 10.7

10.10 4.28 1.24 10.80 10.14 10.29

436 Focus-Execute. Letter Cancellation (Talland, 1965). The stimulus is an array of letters; the subject is asked to mark out a target letter. Time to completion was the dependent measure. Stroop Color Word Inference Test (Stroop, 1935): Subjects read a list of color names, then name the colors of small patches, and finally name the color of words printed in conflicting ink (e.g., word blue is printed in green and the subject must say “green”). Time score for the final list was the dependent measure. Trail Making A & B (Reitan & Wolfson, 1993): Trail Making A requires connecting a series of circled numbers in numerical order as quickly as possible. In Trail Making B, numbers and letters must be connected in alternate order (e.g., 1 to A to 2 to B). Time scores were the dependent measures. Digit Symbol/ Coding: The adult or child form of this substitution task was used, depending upon whether the subject received the WISC-R or WAIS-R. The scaled score was the dependent measure. Sustain. Complex Continuous Performance Test (CPT; Neuchterlein, 1991). Subjects view letters on a computer screen and are instructed to respond by pressing the computer spacebar or mouse button following the presentation of a target letter. Scores used were number of correct responses, number of incorrect responses, and reaction times. The version of the CPT used by the Mirsky group was slightly different in detail from the version used in the present study, but both procedures had the same attentional requirements. That is, both procedures used the AX method in which the participant responds to the letter X only when it follows an A. Shift. Wisconsin Card Sorting Test (Heaton, Chelune, Talley, Kay, & Curtiss, 1993). The subject must sort a series of 128 cards containing geometric forms according to color, shape, or number. Unbeknown to the subject, the relevant concept is changed several times during the course of the test. Scores used in the present study were number of categories successfully completed, total number of correct responses, and number of perseverative errors. Motor Tests. To evaluate the contribution of motor dysfunction to group differences, the Halstead Finger Tapping Test (Reitan & Wolfson, 1993) and the Klove Grooved Pegboard (Matthews & Klove, 1964) were administered. Strength of grip in kilograms was determined with a band dynamometer. Data Analysis It was hypothesized that the subjects with autism would perform poorly on measures of attention only when the task involved a higher level executive func-

Goldstein, Johnson, and Minshew tion or a psychomotor component. Analyses of variance (ANOVAs) were computed to determine significant differences between the two groups on the attention measures administered. Using the motor tests as covariates, analyses of covariance (ANCOVAs) were conducted when there was a significant difference between the two groups on attention measures with a motor component. These tests included Letter Cancellation, the Trail Making Test and Digit Symbol/Coding, all of which use speed of skilled hand movement as the dependent measure, and the Stroop Test, which uses speed of speech. Preliminary analysis of the data indicated that there was a wide age range in both groups, leaving open the possibility that developmental changes could significantly influence the findings. The sample was split at the median age (16 years) and the final analyses were performed as two-way analyses of variance with group (autism vs. control) and age range (older vs. younger) as independent variables. While significant main effects for age might be anticipated as reflecting normal development, significant interactions would indicate developmental differences between autistic and control participants, since age differences found would not be independent of presence or absence of autism. Using Mirsky et al.’s (1991) published principalcomponents factor analysis, it was hypothesized that the autistic subjects’ performance on the Shift and Focus-execute factors would be poorer compared to controls secondary to deficits in executive functioning and psychomotor function, respectively. Factor scores were computed by multiplying test scores by the salient loadings provided by Mirsky et al. (1991) in their factor analysis. The total score for each factor divided by the number of tests with salient loadings was the dependent measure for each factor. Group comparisons of these factor scores were made in addition to those made for the individual tests. RESULTS Results from the comparisons of attention measures are presented in Table II. Significant main effects for group were found for Digit Symbol/Coding, Letter Cancellation, the Stroop Test, Trail Making A and B, and the categories achieved and perseverative error scores of the Wisconsin Card Sorting Test. With the exception of the Wisconsin Card Sorting Test measures, these instruments all have a psychomotor component. Significant main effects for age were found for Arithmetic, the Trail Making Test (A and B), and the Stroop Test. The only significant Group ⫻ Age interaction was

Attentional Processes in Autism

437 Table II. Analyses of Variance for Attention Test Data

Autism Test Digit Span Arithmetic Digit Symbol/Coding Letter Cancellation Trail Making A (sec.) Trail Making B (sec.) Stroop Interference CPT Mean RT # Correct # Incorrect WCST Categories achieved # Correct # Perseverative errors

Controls

Old

Young

M

SD

M

SD

M

SD

M

SD

Fautism(g)

Fage(a)

Fgxa

10.01 9.81 7.20 86.69 32.66 65.85 32.49

4.04 3.90 3.00 34.89 18.47 44.08 11.75

10.35 10.15 10.38 67.39 21.42 48.67 36.35

2.53 2.68 2.34 24.80 8.97 21.58 11.81

10.21 9.45 8.92 74.29 33.09 72.03 37.45

3.33 3.21 3.22 4.58 16.17 37.84 11.58

10.15 10.49 8.40 87.00 19.93 40.11 31.18

3.50 3.43 2.96 5.91 10.60 22.10 11.66

0.51 0.87 46.75c 7.48b 48.47c 20.47c 4.30a

0.00 5.37a 0.03 3.28 63.20c 59.07c 12.58c

0.24 0.51 0.01 1.09 11.74c 7.30b 1.01

0.61 26.93 15.87

0.25 5.31 33.12

0.55 26.76 5.97

0.13 5.15 9.46

0.61 27.79 11.88

0.25 4.47 30.95

0.54 25.35 11.08

0.11 6.00 15.06

0.79 0.01 2.35

1.64 3.65 0.01

1.43 1.71 0.05

4.71 68.77 20.31

1.78 15.04 15.24

5.34 70.77 12.60

1.41 9.74 10.25

5.23 69.41 14.83

1.39 11.53 12.93

4.85 70.19 17.80

1.82 13.60 14.53

6.92b 1.23 16.29c

2.26 0.22 1.98

0.42 1.61 0.73

p ⬍ .05. p ⬍ .01. c p ⬍ .001. a b

are contained in Table III. Covarying the Finger Tapping Test altered the results of the Stroop Test with regard to statistical significance, and reduced the level of significance of Letter Cancellation from p ⬍ .01 to p ⬍ .05. When the Grooved Pegboard Test was used as a covariate, the difference for the Stroop test again became nonsignificant, the significance level for Letter Cancellation went from p ⬍ .01 to p ⬍ .05, and the significance levels for Trail Making A and B went from p ⬍ .001 to p ⬍ .01. Covarying grip strength resulted in nonsignificant group differences on Letter Cancellation and the Stroop Test. None of the motor test covariates attenuated group differences on Trail Making A or B and Digit Symbol/Coding to the point of nonsignificance. Hence, covarying all of the motor tests resulted in nonsignificant differences between the two groups on the Stroop Test, and covarying grip strength produced a nonsignificant difference for Letter Cancellation.

for the Trail Making Test. On Part A, the older autistic participants were substantially slower than the controls (M ⫽ 43.23 vs. 24.98 seconds) than was the case for the younger participants (M ⫽ 22.77 for the autism group and 16.76 for the controls). The same phenomenon occurred in the case of Trail Making B. The older autistics had a mean of 88.70 seconds with a mean of 58.09 seconds for the controls. In the younger groups, M ⫽ 43.49 for the autism group and M ⫽ 36.33 seconds for the controls. It appears that the younger group of autism participants performed more closely to the level of their age peers than was the case for the older participants. The results of a series of ANCOVAs performed for only those tests that produced a significant difference between the autistic subjects and controls and had a motor component to the task (Digit Symbol/Coding, Stroop, Letter Cancellation, and Trail Making A and B)

Table III. Analyses of Covariance for Tests of Motor Function Autism Test Digit Symbol Letter Cancellation (seconds) Trail Making A (seconds) Trail Making B (seconds) Stroop Interference-Correct

Controls

Covariates ( p)

M

SD

M

SD

F

p

Tapping

Pegboard

Grip

7.20 86.69 32.66 65.85 32.49

3.00 34.89 18.47 44.08 11.75

10.38 67.39 21.42 48.67 36.35

2.34 24.80 8.97 21.58 11.81

48.51 7.14 28.69 11.79 4.76

⬍.001 ⬍.01 ⬍.001 ⬍.001 ⬍.05

⬍.001 ⬍.05 ⬍.001 ⬍.001 ⬎.05

⬍.001 ⬍.05 ⬍.01 ⬍.01 ⬎.05

⬍.001 ⬎.05 ⬍.001 ⬍.001 ⬎.05

438

Goldstein, Johnson, and Minshew

Comparisons between groups on the Mirsky factor scores are contained in Table IV. Significant differences were found for the Focus-execute and Shift factors, but not for the Encode and Sustain factors.

DISCUSSION This comprehensive analysis of attentional functioning in individuals with carefully diagnosed highfunctioning autism demonstrates that the major dysfunctions relative to controls are on those measures of attention that require cognitive flexibility or utilize psychomotor speed, as opposed to accuracy or span of apprehension. That is, differences were noted only on the Focus-Execute and Shift factors of the Mirsky group model, and not the Vigilance and Encode factors. Hence, we were unable to confirm the view that individuals with autism have difficulties in encoding information and sustaining attention over time. The sample used covered a broad age range, but only one test, Trail Making, showed a different age difference pattern in autism subjects and controls, suggesting that the findings overall are not confounded by developmental considerations. For the most part, both the older and younger autistic subjects showed the same significant differences, or lack of them, from controls. The significant interactions for Parts A and B of Trail Making appear to be largely attributable to the superior performance of both the younger and older control groups on this test, and the particularly poor performance of the older autism group. These findings are consistent with several previous studies. Normal performance at repeating digits and calculating have previously been reported in autism (Minshew, et al. 1992; Minshew, Goldstein, & Siegel, 1997). Normal performance at a sustained attention task involving the CPT has also been reported recently by Pascualvaca et al. (1998). Results in both of these areas demonstrate that even challenging attentional tasks that

Table IV. Comparison of Factor Scores Between the Autism and Control Groups Autism group

Control group

Factor

M

SD

M

SD

F

p

Encode Focus Shift Sustain

7.54 34.14 29.42 12.40

2.58 9.83 6.02 9.87

7.79 28.21 27.75 9.55

1.59 5.46 4.21 2.98

0.85 3.37 2.18 1.66

⬎.05 ⬍.001 ⬍.05 ⬎.05

do not have a conceptual or psychomotor component do not differentiate high-functioning individuals with autism from controls. The present results also indicate that some of the tests contained in the Mirsky model do not identify dysfunction after adjustments are made for motor function; unequivocal attention dysfunction was not identified using these procedures, even though they assess attention as defined in several ways. These definitions correspond to the major ways in which attention is defined in the literature, that is, encoding sequences of information into short-term memory, sustaining concentration, resisting distraction or “freedom from distractibility” (Cohen, 1957), and shifting focus of attention. The data also suggest that in the case of autism, unequivocal evaluation of attention could not be accomplished using dependent measures based on speed of movement. Significant intergroup differences remained on the Trail Making Test and Digit Symbol/ Coding after covariance analysis. The motor activity required for these tests is different from the Stroop and Letter Cancellation Tests, which only require repetition of a simple response. Both the Trail Making Test and Digit Symbol/Coding require coordinated, skilled sequential movements. Thus, when tasks involving simple, repetitive movement were used, covariance analysis substantially attenuated the difference between the autism and control groups. When tasks involved skilled coordinated movement, the covariance analysis also attenuated differences, but less substantially. However, the motor skill dysfunction associated with autism in itself may have produced such differences, without the necessity of postulating an attention deficit. Support for this view comes from the previously reported finding that individuals with autism did more poorly than controls on Part A but not Part B of the Trail Making Test (Minshew et al., 1992). Part B is a conceptual task, while Part A primarily assesses intactness of rapid coordinated movement. Significant differences between individuals with autism and controls may be found on experimental measures of attention that assess such processes as conceptual reasoning, executive function, rapid decision making, and problem solving, abilities that are widely believed to be impaired in autism (McEvoy, Rogers, & Pennington, 1993; Minshew et al. 1997; Ozonoff, 1995b; Ozonoff et al., 1991, 1994). With regard to the Mirsky model shift factor, there is evidence from other research indicating that individuals with autism do not have difficulty with elementary perceptual shift tasks, but do have difficulty when shifting must be accomplished at a conceptual level (Minshew, et al., 1999).

Attentional Processes in Autism The distinction appears to be in the informationprocessing complexity of the task, particularly when adaptive shifting requires multiple-step decision making. A conceptual shift, for example, may involve changing the relevant dimension in a concept identification procedure; materials previously correctly sorted by size must be sorted by color. Furthermore, numerous other studies involving the Wisconsin Card Sorting Test have produced equivocal results, with several reports of normal functioning by autism samples on several of the measures derived from this test (Minshew et al., 1997, Ozonoff, 1995a). In the present study, the group with autism did not differ from controls on the correct response score of the Wisconsin Card Sorting Test. Considering the literature as a whole, differences in various scores from the Wisconsin Card Sorting Test between individuals with autism and normal controls may be a matter of the presence of autism in combination with general ability level. Thus, evidence for impaired shifting of attention in autism appears to be associated with various aspects of complex information processing such as planning decision strategies or concept formation, and not with perceptual shifting of focus of the type that may be mediated by cerebellar function (Courchesne et al., 1993). Furthermore, this deficit, at least when measured by such procedures as the Wisconsin Card Sorting Test, may not appear in higher functioning individuals with autism. These findings indicate that the well-established cognitive deficits and their associated abnormal behaviors associated with autism are not the result of a failure to incorporate information, or to sustain concentration, or to resist distraction. There was no evidence for any of these phenomena. Therefore, it appears that if individuals with autism do appear to have attentional deficits, they would be at the conceptual level, perhaps involving executive abilities and monitoring of novel information, as has been suggested elsewhere (Ozonoff, 1995b; Ozonoff et al., 1994; Pascualvaca et al., 1998). Such considerations as the ability to organize information and the capacity to monitor ongoing events and make rapid adjustments are likely to be relevant considerations. This conclusion is also supported by the recent experimental literature (e.g., Burack et al., 1997). The autism and control samples included a very broad age range, and represent a possible limitation of the study. It is known that certain abilities improve over time while others deteriorate over time in autism (Szatmari, Bartolocci, Bremner, Bond, & Rich, 1989). However, our examination of age differences only showed a significant interaction between age range and presence or absence of autism for the Trail Making Test.

439 Thus, our study does not, in general, suggest the presence of unusual developmental changes on the tests given in the subjects with autism. Developmental studies using age-appropriate materials are nevertheless needed to evaluate this matter in more detail, since changes in younger or older individuals on different tests have not been ruled out. For example, very young children with autism could have severe, generalized attentional difficulties that get better over the years. The present study, like the Mirsky model, used mainly clinical tests, and it is difficult to compare results from such tests with those coming from studies using experimental procedures. Thus, more advanced prospective studies relating the construct of attention to working memory and executive function across developmental stage appear to be a very worthwhile area of investigation for future studies. ACKNOWLEDGMENT This work was supported by a grant from the Edith L. Trees Charitable Trust, NINDS Grant NS33355 and NICHD Collaborative Program of Excellence in Autism to Dr. Minshew and by the Medical Research Service, Department of Veterans Affairs. REFERENCES Barry, R. J., & James, A. L. (1988). Coding of stimulus parameters in autistic, retarded, and normal children: evidence for a twofactor theory of autism. International Journal of Psychophysiology, 6, 139–149. Bauman, M. L. (1992). Motor dysfunction in autism. In A. B. Joseph & R. R. Young (Eds.), Movement disorders in neurology and neuropsychiatry (pp. 659–661). Boston: Blackwell Scientific. Bryson, S. E., Wainwright-Sharp, J. A., & Smith, I. M. (1990). Autism: A developmental spatial neglect syndrome? In J. T. Enns (Ed.), The development of attention: Research and theory (pp. 405–427). Amsterdam, The Netherlands: Elsevier North Holland. Burack, J. A. (1994). Selective attention deficits in persons with autism: Preliminary evidence of an inefficient lens. Journal of Abnormal Psychology, 103, 535–543. Burack, J. A., Enns, J. T., & Johannes, E. A. (1997). Attention and autism: Behavioral and Electrophysiological Evidence. In D. J. Cohen & F. R. Volkmar (Eds.), Handbook of autism and pervasive developmental disorders (2nd ed. pp. 226–247). New York: J. Wiley. Burack, J. A., & Iarocci, G. (1995). Visual filtering and covert orienting in autism. Paper presented at the meeting of the Society for Research in Child Development, Indianapolis, IN. Casey, B. J., Gordon, C. T., Mannheim, G. B., & Rumsey, J. M. (1993). Dysfunctional attention in autistic savants. Journal of Clinical and Experimental Neuropsychology, 15, 933–946. Cohen, J. (1957). A factor-analytically based rationale for the Wechsler Adult Intelligence Scale. Journal of Consulting Psychology, 6, 451–457. Courchesne, E., Akshoomoff, N. A., Townsend, J., & Saitoh, O. (1995). A model system for the study of attention and the cerebellum:

440 Infantile autism. Electroencephalography and Clinical Neurophysiology-Supplement, 44, 315–325. Courchesne, E., Townsend, J., Akshoomoff, N. A., Saitoh, O., Yeung-Courchesne, R., Lincoln, A. J., James, H. E., Haas, R. H., & Lau, L. (1994). Impairment in shifting attention in autistic and cerebellar patients. Behavioral Neuroscience, 108, 848–865. Courchesne, E., Townsend, J. P., Akshoomoff, N. A., Yeung-Courchesne, R., Press, G. A., Murakmi, J. W., Lincoln, A. J., James, H. E., Saitoh, O., Egaas, B., Haas, R. H., & Schriebman, L. (1993). A new finding: Impairment in shifting of attention in autistic and cerebellar patients. In S. H. Broman, & J. Grafman (Eds.), Atypical deficits in developmental disorders: Implications for brain function (pp. 101–137). Hillsdale, NJ Erlbaum. Dawson, G., & Levy, A. (1989). Arousal, attention and the socioemotional impairments of individuals with autism. In G. Dawson (Ed.), Autism: Nature, diagnosis, and treatment (pp. 144 –173). New York: Guilford. Enns, J. T., & Akhtar, N. (1989). A developmental study of filtering in visual attention. Child Development, 60, 319–337. Gillberg, C., & Coleman, M. (1992). The biology of the autistic syndromes (2nd ed.). London: MacKeith. Heaton, R. K., Chelune, C. J., Talley, J. L., Kay, G. G. & Curtiss, G. (1993). Wisconsin Card Sorting Test manual, revised and expanded. Odessa, FL: Psychological Assessment Resources. Hughes, C. (1996). Brief report: Planning problems in autism at the level of motor control. Journal of Autism and Developmental Disorders, 26, 99–107. Hutt, C., Hutt, S. J., Lee, D., & Ounsted, C. (1964). Arousal and childhood autism. Nature, 204, 908–909. Le Couteur, A., Rutter, M., Lord C., Rios, P., Robertson, S., Holdgrafer, M., & Mclennan J. (1989). Autism diagnostic interview: A standardized investigator-based instrument. Journal of Autism and Developmental Disorders, 19, 363–387. Lord, C., Rutter, M., & Goode, S. (1989). Autism Diagnostic Observation Schedule: A standardized investigator-based instrument. Journal of Autism and Developmental Disorders, 19, 185–212. Lord, C., Rutter, M., & Le Couteur, A. L. (1994). Autism Diagnostic Interview Revised: A revised version of a diagnostic interview for caregivers of individuals with possible pervasive developmental disorders. Journal of Autism and Developmental Disorders, 24, 659–685. Lovaas, O. I., Koegel, R., & Schriebman, L. (1979). Stimulus overselectivity in autism. A review of research. Psychological Bulletin, 86, 1236–1254. Matthews, C. G., & Klove, H. (1964). Instruction manual for the Adult Neuropsychology Test Battery. Madison: University of Wisconsin Medical School. McEvoy, R. E., Rogers, S. J., & Pennington, B. F. (1993). Executive function and social communication deficits in young autistic children. Journal of Child Psychology and Psychiatry, 34, 563–578. Minshew, N. J., Goldstein, G., & Siegel, D. J. (1997). Neuropsychologic functioning in autism: profile of a complex information processing disorder. Journal of the International Neuropsychological Society, 3, 303–316. Minshew, N. J., Luna, B., & Sweeney, J. A. (1999). Oculomotor evidence for neocortical systems but not cerebellar dysfunction in autism. Neurology, 52, 917–922. Minshew, N. J., Goldstein, G., Muenz, L. R., & Payton, J. B. (1992). Neuropsychological functioning in non-mentally retarded autistic individuals. Journal of Clinical and Experimental Neuropsychology, 14, 740–761. Mirsky, A. F., Anthony, B. J., Duncan, C. C., Ahearn, M. B., & Kellam, S. G. (1991). Analysis of the elements of attention: A neuropsychological approach. Neuropsychology Review, 2, 109–145. Nuechterlein, K. H. (1991). Vigilance in schizophrenia and related disorders. In S. Steinhauer, J. Gruzelier, & J. Zubin (Eds.), Handbook of schizophrenia (Vol. 5, pp. 397–433). Amsterdam, The Netherlands: Elsevier.

Goldstein, Johnson, and Minshew Ornitz, E. M. (1988). Autism: A disorder of directed attention. Brain Dysfunction, 1, 309–322. Ornitz, E. M., & Ritvo, E. R. (1968). Perceptual inconstancy in early infantile autism. Archives of General Psychiatry, 18, 76–98. Ozonoff, S. (1995a). Reliability and validity of the Wisconsin Card Sorting Test in studies of autism. Neuropsychology, 9, 491–500. Ozonoff, S. (1995b). Executive functions in autism. New York: Plenum Press. Ozonoff, S., Pennington, B. F., & Rogers S. J. (1991). Executive function deficits in high-functioning autistic individuals: Relationship to theory of mind. Journal of Child Psychology and Psychiatry, 32, 1081–1105. Ozonoff, S., Strayer, D. L., McMahon, W. M., & Filloux, F. (1994). Executive function abilities in autism and Tourette syndrome: An information processing approach. Journal of Child Psychology and Psychiatry, 35, 1015–1032. Pascualvaca, D. M., Fantie, B. D., Papageorgiou, M., & Mirsky, A. F. (1998). Attentional capacities in children with autism: Is there a general deficit in shifting focus. Journal of Autism and Developmental Disabilities, 28, 467–478. Pasto, L., & Burack, J. A. (1995, June). A developmental study of visual filtering: Can windows facilitate filtering efficiency? Poster presented at the meeting of the Canadian Psychological Association, Charlottetown, Canada. Pierce, K., Glad, K. S., & Schriebman, L. (1997). Social perception in children with autism: An attentional deficit? Journal of Autism and Developmental Disorders, 27, 265–282. Rapin, I. (1997). Current concepts: Autism. New England Journal of Medicine, 337, 97–104. Reitan, R. M., & Wolfson, D. (1993). The Halstead-Reitan Neuropsychological Test Battery: Theory and clinical interpretation (2nd ed.). S. Tucson, AZ: Neuropsychology Press. Rimland, B. (1964). Infantile autism. New York: Appleton-CenturyCrofts. Rincover, A., & Ducharme, J. M. (1987). Variables influencing stimulus overselectivity and “tunnel-vision” in developmentally disabled children. American Journal of Mental Deficiency, 91, 422–430. Rumsey J. M., & Hamburger, S. D. (1988). Neurophysiological findings in high-functioning men with infantile autism, residual state. Journal of Clinical and Experimental Neuropsychology, 10, 201–221. Smith, I. M., & Bryson, S. E. (1994). Imitation and action in autism: A critical review. Psychological Bulletin, 116, 259–273. Stroop, J. R. (1935). Studies of interference in serial verbal reactions. Journal of Experimental Psychology, 18, 643–662. Szatmari, P., Bartolucci, G., Bremner, R., Bond, S., & Rich S. (1989). A follow-up study of high-functioning autistic children. Journal of Autism and Developmental Disorders, 19, 213–225. Talland, G. A. (1965). Deranged memory. New York: Academic Press. Townsend J., & Courchesne, E. (1994). Parietal damage and narrow “spotlight” spatial attention. Journal of Cognitive Neuroscience, 6, 220–232. Wainwright-Sharp, J. W., & Bryson, S. E. (1993). Visual orienting deficits in high functioning people with autism. Journal of Autism and Developmental Disorders, 26, 423– 438. Wainwright-Sharp, J. A., & Bryson, S. E. (1996). Visual-spatial orienting in autism. Journal of Autism and Developmental Disorders, 26, 423–438. Wechsler, D. (1974). Wechsler Intelligence Scale for Children (Rev. ed.). San Antonio, TX: Psychological Corp. Wechsler, D. (1981). WAIS-R manual. San Antonio, TX: Psychological Corp. Yeates, K. O., & Taylor, H. G. (1998). Neuropsychological assessment of older children. In G. Goldstein, P. D. Nussbaum, & S. R. Beers (Eds.), Neuropsychology (pp. 35–61). New York: Plenum Press. Zubin, J. (1975). Problem of attention in schizophrenia. In M. L. Kietzman, S. Sutton, & J. Zubin (Eds.), Experimental approaches to psychopathology (pp. 139–166). New York: Academic Press.