Pergamon

Archives of Clinical Neuropsychology,Vol. 10, No. 3, pp. 211-223, 1995 Copyright 1995NationalAcademyof Neuropsychology

Printed in the USA.All rights reserved 0887-6177/95 $9.50 + .00

0887-6177(94)00041-7

Children's Color Trails

Jane Williams, Vaughn Rickert, John Hogan, and A. I. Zolten
University of Arkansas for Medical Sciences, Department of Pediatrics

Paul Satz, Louis F. D'Elia, Robert F. Asarnow, Ken Zaucha, and Roger Light
Neuropsychiatric Institute and Hospital, University of California, Los Angeles

Color Trails for Children was developed in response to the need for instruments which minimize cultural bias in neuropsychological testing. The test, similar in format to Trail Making, was designed to provide an evaluation of speeded visuomotor tracking while minimizing the influence of language. The present research involves two exploratory studies which examine the relationship between Color Trails for Children and Trail Making, factors that may affect performance times, and discriminant validity. Results indicate that the tests appear to measure the same neuropsychological domains, and administration o f Trail Making did not significantly alter performance times on Color Trails. Increasing age and 1(2 were related to quicker completion time for both tests. Females were found to complete Color Trails 2 and Trail Making Part B more quickly than males in this sample. Comparison between children diagnosed with learning disabilities, attention deficits, or mild neurological conditions and a preliminary standardization sample supported the discriminant validity of Color Trails to distinguish between normal controls and children with altered neuropsychological functioning. Comparison between clinical conditions indicated that Color Trails 2 was particularly sensitive in discriminating among the groups. Although further research is needed, results suggest that Color Trails has the potential to be an effective research and clinical tool in child neuropsychological assessment.

A continuing need exists for the d e v e l o p m e n t o f instruments w h i c h m i n i m i z e cultural bias in n e u r o p s y c h o l o g i c a l evaluation. In addition to clinical assessment, these tests are necessary for cross-cultural research in conditions affecting Address correspondence to: Jane Williams, PhD, University of Arkansas for Medical Sciences, Department of Pediatrics, 800 Marshall, Little Rock, AR 72202.
211

212

J. Williams et al.

neurological functioning. Due to its sensitivity in detecting altered neurological functioning in children, the traditional Trail Making Test (Reitan, 1959; Reitan & Davison, 1974) is one of the most frequently administered neuropsychological tests in English speaking countries (Reitan, 1971; Reitan & Herring, 1985; Rosin & Levett, 1989a). However, its application in cross-cultural contexts is potentially limited due to its reliance on the English alphabet in Part B. This feature of the test may adversely affect performances of individuals who are unfamiliar with the English language. The Color Trails Test (D'Elia & Satz, 1989) was developed as part of a neuropsychological assessment battery for the World Health Organization's (WHO) cross-cultural study of HIV-1 infection (Maj et al., 1991). The instrument relies on the concepts of number and color which are universally employed across cultures. It was designed to minimize the influence of language, including instructions which can be presented nonverbally with visual cues. The test, similar in format to Trail Making, was created in an attempt to develop a culturally fair assessment of speeded visuomotor tracking. Color Trails requires intact motor speed, attention, and visual scanning for quick and accurate performance. Versions of the test are available for both children and adults. Research with the Color Trails Test for Adults (D'Elia, Satz, & Uchiyama, in press) has suggested both discriminant validity (Maj et al., 1993) and robustness in its sensitivity across cultures (Maj et al., 1994). Color Trails for Children has the potential for being an effective neuropsychological instrument for use in cross-cultural research with children unfamiliar with the English language, as well as in the assessment of children who are illiterate, have reading and language disorders, or have limited educational experiences. It is hypothesized that Color Trails for Children will demonstrate the same discriminant validity as traditional Trail Making in the assessment of children with brain damage (Reitan, 1971; Reitan & Herring, 1985; Rosin & Levett, 1989a), Attention Deficit Disorder (Klee, 1986), hyperactivity (Johnston, 1986), and learning disabilities (Davis, Adams, Gates, & Cheramie, 1989). Research is needed concerning discriminant validity as well as potential factors that may affect performance on Color Trails for Children. Factors suggested to affect performance on traditional Trails include age, IQ, and gender (Rosin & Levett, 1989a). During childhood, quicker performances on the Trail Making Test have been demonstrated with increasing age (Rosin & Levett, 1989b) and increasing IQ (Horton, 1979; Rosin & Levett, 1989b). Differences in performance resulting from gender have not been supported (Leon-Carrion, 1989; Mittelmeir, Rossi, & Berman, 1989; Reitan, 1971; Rosin & Levett, 1989b). Differences in cultural context have also been shown to affect performance times with Spanish children performing significantly slower than American children on Trail Making (Leon-Carrion, 1989), while South African children with average and above average intelligence were found to perform

Color Trails

213

significantly faster than a random sample of American children (Rosin & Levett, 1989b). The present research involves two studies that were designed to examine factors that may affect performance times and to explore the discriminant validity of Color Trails for Children. The first study addresses the strength of the relationship between Color Trails and Traditional Trails, order effects when both tests are administered, and factors that may alter performance on Color Trails for Children. The second study addresses discriminant validity by comparing performances of children with learning disabilities, attention deficits, and mild neurological disorder with an age-equivalent sample of the preliminary standardization group on Color Trails for Children. In addition, children with these disorders were compared with each other on Color Trails in an attempt to determine the sensitivity of this instrument to differentiate among the groups.

STUDY I
Method Subjects. Subjects were 223 children who had been evaluated for learning, emotional, or behavioral difficulties in an outpatient developmental center (n = 163) or through the local school district (n = 60). Each child was administered a standardized intellectual, academic, and behavioral assessment. The age of the children ranged from 5 years, 11 months to 16 years, 10 months (Mean = 11 years, 1 month; SD = 29 months). There were 69 females and 154 males. The mean Full Scale IQ was within the average range (Mean = 91; SD = 13.62; Range = 59-131). Instrument. Color Trails for Children consists of Part 1 and Part 2. Both parts are graphomotor tasks requiring the use of pencil and paper. Each part is printed on standard 8.5 by 11 inch paper, with a practice sample provided prior to actual test administration. Color Trails 1 (Figure 1) requires the child to quickly and c o r r e c t l y s e q u e n c e n u m b e r s f r o m 1 to 15. All odd n u m b e r s (1,3,5,7,9,11,13,15) are embedded within circles that have a pink background, while all even numbers (2,4,6,8,10,12,14) are embedded within circles that have a yellow background. Scoring consists of time in seconds from initiation to completion of task. The number of errors made are recorded. Color Trails 2 (Figure 2) contains duplicates of each number from 1 to 15 embedded within pink and yellow circles. The child is required to quickly connect the circles in ascending order, but alternating between pink and yellow colors. In other words, the child would connect Pink 1 with Yellow 2 then to Pink 3 and so on through the number 15. Scoring consists of time in seconds from initiation to completion of the task. The number of errors committed are recorded.

214

J. Williams et al.

~@ @ @ @ ~@ @

@ @ @ @

FIGURE 1. Color Trails 1 for Children. On the actual test protocol, all odd numbers are embedded in pink circles (shown here gray), while even numbers are in yellow circles (shown here white).

Procedure. Color Trails and traditional Trail Making were individually admin-

istered with each child given the practice task prior to the test proper. The child was told to work as quickly as possible and to try and not lift the pencil from the paper. If an error was made, the child was immediately directed to correct the error and start at the point where the mistake was made. In order to determine whether the order of administration had an effect on performance, a random subset of the subjects (n = 119) were alternately administered either Trails A and B followed by Color Trails 1 and 2 or Color Trails 1 and 2 followed by Trails A and B.

Color Trails

215

@ @ @

@ @

@ @
@ @

@ ~@

FIGURE 2. Color Trails 2 for Children. On the actual test protocol, each number is duplicated and embedded in separate pink (shown here gray) and yellow (shown here white) circles.

Results

As can be seen in Table 1, correlations between all types o f Trails were highly related ( p < .001). Means and standard deviations of the subjects who completed Trails A and B followed by Color Trails 1 and 2 (n = 61) as well as those subjects who completed Color Trails 1 and 2 followed by Trails A and B (n = 58) are found in Table 2. In order to determine if order of presentation affected time completion, a repeated three-way A N O V A with one between factor (administration) and two within factors (type and difficulty) was computed.

216

J. Williams et al.
TABLE 1 Correlations between Types of Trail M a k i n g Tests Type Trails A Trails B Color Trails 1 Color Trails 2 Trails A 1.00 Trails B .59 1.00 Color Trails 1 Color Trails 2 .74 .54 1.00 .74 .67 .69 1.00

There was not a siginficant three-way interaction for Administration x Type x Difficulty (F (1,117) = .113, p > .73). There were no two-way interactions for Administration x Type (F (1,117) = 2.05, p > .15) or Administration x Difficulty (F (1,117) = .10, p > .75) found. Main effects were noted for type (F (1,117) = 15.38, p < .001) and difficulty (F (1,117) = 362.81, p < .001). As expected, children's combined times on Trails A and B were significantly faster than combined times for Color Trails 1 and 2, while their combined times on Trails A and Color Trails 1 were significantly faster than combined times on Trails B and Color Trails 2. There was no main effect for administration (F (1,117) = .64, p > .42) which involved comparison of mean scores on all Trails. To further examine the possible effects of administration, Student's t-tests were computed by group on each type of Trails. Comparison of group scores based on administration for Trails A (t = 2. 17, p > .14), Trails B (t = 1.03, p > .31), Color Trails 1 (t = . 17, p > .68), and Color Trails 2 (t = .05, p > .82) did not indicate any significant differences in performance resulting from order of administration. Findings did not suggest improvement in performance on Color Trails resulting from having Trails A and B administered first. In order to analyze the effects of age, IQ, and gender, a three-way multiple analysis of variance (MANOVA) was performed (Table 3). Age consisted of three groups including children ages 6 - 8 years, 9-11 years, and 12

TABLE 2 M e a n Times in Seconds a n d S t a n d a r d Deviations According to Order of Administration Group Administration 1a Mean SD Administration 2 b Mean SD Trails A Trails B Color Trails 1 Color Trails 2

23.74 15.62 20.16 10.23

59.10 33.64 53.41 26.85

24.43 10.38 23.41 16.18

64.39 35.56 63.02 30.61

aAdministration 1 = Trails A and B fast. bAdministration 2 = Color Trails 1 and 2 first.

Color Trails
TABLE 3 Mean Times in Seconds on Trail Making and Color Trails According to Age, IQ, and Gender
Variable Age 6-8 yrs (n = 49) 9-11 yrs (n = 100) 12+ yrs (n = 74) Sex Female (n = 69) Male (n = 154) IQ FSIQ = < 83 (n = 68) FSIQ = 84-98 (n = 92) FSIQ = 99+ (n = 63) Trails A Trails B Color Trails 1 Color Trails 2

217

33.97 21.73 15.94 22.68 25.07 30.62 22.15 18.87

80.84 54.90 38.17 52.52 63.42 69.74 58.20 45.98

39.24 23.17 18.17 25.13 28.59 34.42 24.03 22.12

92.35 60.89 44.56 57.89 73.98 82.39 60.94 54.47

years and above. IQ was divided into three groups based on the overall mean (M = 91.2; Range = 59-131) and standard deviation (SD = 13.6) for the entire sample. Group 1 consisted of children with FSIQs < 83, Group 2 consisted of children with FSIQs from 84 through 98, and Group 3 consisted of children with FSIQs > 99. There was not a siginficant three-way interaction for Age x IQ x Gender (Wilks' Lambda (16,611) = .95, p > .81). Nor were there any two-way interactions for Age x IQ (Wilks' Lambda (16,611) = .91, p > .23), Age x Gender (Wilks' Lambda (8,400) = .96, p > .51), or IQ x Gender (Wilks' Lambda (8,400) = .94, p > .09). Main effects were noted for Age (Wilks' Lambda (8,400) = .60, p < .001), IQ (Wilks' Lambda (8,400) = .79, p < .001), and Gender (Wilks' Lambda (4,200) = .91, p < .001). Follow-up analyses using Scheffe tests for significant differences indicated that the youngest (Group 1) and oldest group (Group 3) differed significantly from each other in time to complete all types of Trails. The youngest (Group 1) and middle group (Group 2) differed significantly in time to complete all types of Trails. The middle (Group 2) and oldest group (Group 3) differed significantly in time to complete all Trails except Color Trails 1 (p > .02). For the main effect of Gender, follow-up tests indicated significant differences between males and females on Trail Making Part B and Color Trails 2 with females performing significantly faster. Significant differences were not found for Trail Making Part A (p > .10) or Color Trails I (p > .04). For the main effect of IQ, follow up tests indicated significant differences between the lowest (Group 1) and highest group (Group 3) on all types of Trails. Differences between the lowest (Group 1) and middle group (Group 2) were significant for Trails A, Color Trails 1, and Color Trails 2, but not for Trails B (p > .01). The middle group (Group 2) and highest group (Group 3) differed significantly on Trails B, but not for Trails A (p > .13), Color Trails 1 (p > .60), or Color Trails 2 (p > .25).

218

J. Williams et al.

STUDY 2
Method Subjects. Subjects were 200 children from the original population who were

diagnosed with mild neurological disorders (n = 67), learning and/or language disabilities (n = 101), and learning and/or language disabilities with Attention Deficit Hyperactivity Disorder (n = 32). The rationale for selection of these groups was to examine a continuum of disorders from those postulated to involve subtle neurological findings, such as in Attention Deficit Hyperactivity Disorder, to those with documented neurological changes, such as in Traumatic Brain Injury. Children diagnosed with mild neurological disorders included primarily seizure disorders and mild to moderate closed head injuries. Children with epilepsy involved generalized or complex partial seizure disorders, while children with head injuries included mild to moderate brain insult generally including a loss of consciousness and positive MRI findings. None of these children had been diagnosed with learning disabilities or Attention Deficit Hyperactivity Disorder. The diagnosis of learning disability was based on state and federal guidelines with all children having a significant discrepancy between measured cognitive ability and academic achievement in reading, math, and/or written expression. The majority of children in this sample had reading and written expressive disorders. The diagnosis of language disorder was based on state and federal guidelines with children demonstrating a significant discrepancy between measured cognitive ability and acquired language skills. The diagnosis of Attention Deficit Hyperactivity Disorder resulted from significant elevations on the Conners Parent and Teacher Rating Scales (Conners, 1982), behavioral observations in the clinic, and cognitive factors such as computerized tests of attention and tasks assessing freedom from distractibility. The majority of children in the category involving two diagnoses had learning disabilities and attention deficits with less than 10% having language disorders and attention deficits. Age of the children ranged from 5 years, 11 months to 16 years, 10 months (M = 11 years, 1 month; SD = 29 months). There were 60 females and 140 males. The mean Full Scale IQ was within the Average range (M = 91; SD = 14; Range = 59-131) based on the Wechsler Intelligence Test for ChildrenRevised (WISC-R; Wechsler, 1974), Kaufman Assessment Battery for Children (KABC; Kaufman & Kaufman, 1983), or Stanford-Binet Intelligence Scale: Fourth Edition (Thorndike, Hagen, & Sattler, 1986). In addition to the clinical groups, normal controls were obtained from the preliminary standarization group for Color Trails for Children. This group was part of an NIH-funded study of mild head injuries in children (Asarnow, Satz, Lewis, Zaucha, & Light, 1993). It consisted of 388 children ages 8-16 from

Color Trails

219

the Downey Unified School District in Los Angeles County. The mean age of the children at the time of testing was 12 years, 1 month (SD = 27 months) with a mean PPVT-R standard score of 102 (SD = 17). Children were excluded from the standardization group if they were in special placement for any learning, attentional, or psychological disability.

Procedure. Procedure for the clinical groups was identical to the previous
study. Normal controls were administered Color Trails for Children only.

Resul~
Statistical analysis of group differences involved the use of one-way analysis of covariance (ANCOVA) on Color Trails 1 and Color Trails 2 with age as a covariate due to the older mean age of the standardization group. Children below 8 years and above 16 years were eliminated from the clinical groups to equate the age range of all groups. Comparisons indicated a significant difference on Color Trails 1 (F(3,563) = 14.77, p < .001) and Color Trails 2 (F(3,563) = 58.47, p < .001) between the groups (Table 4). Followup analyses using Scheffe tests for significant differences indicated that the normal controls were quicker than the Learning Disabled, Mild Neurological, or Learning Disabled/Attention Deficit groups on both Color Trails 1 and 2. Statistical assessment of group differences involved the use of one-way analysis of variance (ANOVA) on Trail Making A, Trail Making B, Color Trails 1, and Color Trails 2. As seen in Table 5, comparisons did not indicate a significant difference on Trails A (F(2,197) = 1.65, p > .19), Trails B (F(2,197) = .94, p > .39), or Color Trails 1 (F(2,197) = 2.85, p > .06) among the Learning Disabled, Mild Neurological, or Learning Disabled/Attention Deficit groups. A significant difference was found for Color Trails 2 (F(2,197) = 3.14, p < .05). Follow-up analysis using Scheffe tests indicated that the Learning Disabled group performed significantly quicker than the Learning

TABLE 4 Comparison of Mean Times in Seconds and Standard Deviations of Clinical Groups and Normal Controls on Color Trails Con~'ols (n = 388) LD (n = 93) Mild Neurological (n = 58) LD + ADHD (n = 29)

Group Color Trails 1 Mean

SD
Color Trails 2 Mean

17.68 7.9 37.22 15.6

21.22 10.3 55.61 23.1

25.27 15.3 60.48 24.8

27.37 12.2 76.76 38.9

SD

Note. LD = Learning Disabled; LD + ADHD = Learning Disabled with

Attention Deficits.

220

J. Williams et al.

Disabled/Attention Deficit group on Color Trails 2. The Mild Neurological group did not differ from either the Learning Disabled or Learning Disabled/ Attention Deficit groups.
Discussion

Present findings suggest a high correlation between Color Trails for Children and Trail Making, indicating that these tasks measure similar neuropsychological domains. The relationships are similar, although stronger than those suggested in adult studies with Color Trails. Maj et al. (1993) found significant relationships between Trails A and Color Trails 1 (r = .41, p < .05) and Trails B and Color Trails 2 (r = .50, p < .001) in a sample of normal subjects in which the order of administration was counterbalanced. In the present study, order of administration in which Trail Making was given first did not suggest that previous experience with Trails A and B had a significant impact on quickness to complete Color Trails. Consistent with previous findings, increasing age and IQ resulted in quicker performance times on Color Trails for Children and Trail Making. In contrast to previous studies, gender was a differentiating factor for Color Trails 2 and Trails B with girls performing more quickly on these two tasks. Results from comparisons of performance times on Color Trails between clinical groups and normal controls suggest discriminant validity for this instrument. Color Trails appears to be sufficiently sensitive to differences in neuropsychological functioning as normal controls were found to perform significantly more quickly than children with learning disabilities, mild neurological conditions, or learning disabilities with attention deficits.

TABLE 5 Comparison of Mean Time in Seconds and Standard Deviations on the Trail Making Test and Color Trails According to Clinical Group

Group Trails A Mean

SD

Learning Disabled (n = 101) 21.18 10.87 55.24 30.64 22.84 12.77 59.96* 28.55

Mild Neurological LearningDisabled+ADHD (n = 67) (n = 32) 22.90 10.25 56.43 28.04 27.31 15.84 64.06 29.36 25.56 18.41 63.44 30.47 27.90 12.59 75.50* 38.40

Trails B Mean
SD

Color Trails 1 Mean

SD

Color Trails 2 Mean

SD

*< .05

Color Trails

221

Examination of group performances indicated that children with learning disabilities, uncomplicated by attention deficits, were the quickest in completion of all types of Trails even though they performed significantly more slowly than normal controls. Children with mild neurological variability, such as epilepsy, were not significantly different from the learning disabled children, but consistently performed at a slower pace. The factor that resulted in the slowest performance on all types of Trails was when attention deficits were present. The poorer performance by children with attention deficits supports past research demonstrating the sensitivity of Trail Making in differentiating this group. It would be of interest to determine whether children in previous studies with learning disabilities who performed significantly slower than normal controis on Trail Making had problems with attention. In one study, it was noted that performance on subtests from the freedom from distractibility factor of the WISC-R (Arithmetic, Digit Span, and Coding) was a discriminating variable for a group of boys with reading disabilities (McManis, Figley, Richert, & Fabre, 1978). Significantly lower scores on these subtests would suggest that these learning disabled children, who had slower times on Trails, may have had difficulty with attention to task. In the present sample, Color Trails 2 appeared to be the most sensitive of the tasks in discriminating among groups of children with altered neuropsychological functioning. Three possible explanations are offered concerning this increased sensitivity. First, the variable of color may more significantly interfere with the peformance of children diganosed with altered neuropsychological functioning. This would be supported by the near signficant finding (p = .06) of Color Trails 1 among the clinical groups. Second, Color Trails 2 is similar to Trails B in that it requires an ability to shift cognitive sets which is not found in Trail Making Part A or Color Trails 1. In past studies, Trail Making Part B has demonstrated greater sensitivity to cerebral dysfunction in contrast to Part A (Horton, 1979). Trail Making Part B has been shown to distinguish between normal controls and children with Attention Deficit Hyperactivity Disorder (Boucugnani & Jones, 1989), probably due to its sensitivity to frontal lobe functioning (Boucugnani & Jones, 1989; Shute & Huertas, 1990). The similarities in task demands may contribute to Color Trails 2 increased discrimination. Third, Color Trails 2 may reduce the confounding effect of language found in Trail Making Part B. Previous research has demonstrated that an experimental Trail Making test, based on spatial rather than language cues, was more sensitive in distinguishing between children with learning disabilities and normal controls than traditional Trails (Davis et al., 1989). The present research represents an exploratory analysis in the use of Color Trails for Children. A standardization sample of normal controls stratified by age, IQ, gender, and geographic location is needed before the test can be used with great confidence. Findings suggest that Color Trails has the potential of being an effective tool in test batteries used to detect altered neuropsychological

222

J. lkflliams et al.

functioning in children. Confounding effects introduced by language, illiteracy, reading disabilities, or educational experience may be minimized with this instrument, making it an asset in cross-cultural research.
Acknowlegements - - Special appreciation is expressed to Amy Hendon and Joe Smith of Arkansas Children's Hospital for graphics and computer assistance.

REFERENCES
Asamow, R. E, Satz, P., Lewis, R., Zaucha, K., & Light, R. (1993) The UCLA studies of mild closodhead injury in children and adolescents [Summary]. Journal of Clinical and Experimental Neumpsychology, 15( 1), 20. Boucugnani, L. L., & Jones, R. W. (1989). Behaviors analogous to frontal lobe dysfunction in children with attention deficit hyperactivity disorder. Archives of Clinical Neuorpsychology, 4, 161-173. Conners, C. K. (1982). Parent and teacher rating forms for the assessment of hyperkinesis in children. In P. A. Keller & L. G. Ritt (Eds.), Innovations in clinical practice: Vol 1. A resource book (pp. 257-264). Sarasota, FL: Professional Resource Exchange. Davis, R. D., Adams, R. E., Gates, D. O., & Cheramie, G. M. (1989). Screening for learning disabilities: A neuropsychological approach. Journal of Clinical Psychology, 45, 423-429. D'Elia, L., & Satz, P. (1989). Color Trails I and 2. Odessa, FL: Psychological Assessment Resources. D'Elia, L., Satz, P., & Uchiyama, C. (in press). Standardization and validation of the Color Trails Testfor Adults. Odessa, FL: Psychological Assessment Resources. Horton, A. M. (1979). Some suggestions regarding the clinical interpretation of the Trail Making Test. Clinical Neuropsychology, 1, 20-23. Johnston, C. W. (1986). The neuropsychological evaluation of attention deficit disorder. Psychiatric Annals, 16, 47-51. Kaufman, A., & Kaufman, N. (1983). Kaufman assessment battery for children. Circle Pines, MN: American Guidance Service. Klee, S. H. (1986). The clinical psychological evaluation of attention deficit disorder. Psychiatric Annals, 16, 43-46. leon-Carrion, J. (1989). Trail making test scores for normal children: Normative data from Spain. Perceptual and Motor Skills, 68, 627-630. Maj, M., D'Elia, L., Satz, P., Janssen, R., Zaudig, M., Uchimyama, C. Starace, F., Galderisi, S., & Chervinsky, A. (1993). Evaluation of two new neuropsychoiogical tests designed to minimize cultural bias in the assessment of HIV-1 seropositive persons: A WHO study. Archives of Clinical Neuo rpsychology, 8(2), 123-135. Maj, M., Janssen, R., Satz, P., Zaudig, M., Starace, E, Boor, D., Sughondhabirom, B., Bing, E. G., Luabeya, M. K., Ndetei, D., Riedel, R., Shulte, G., & Sartorius, N. (1991). The world health organization's cross-cultural study on neuropsychiatric aspects of infection with the human immunodeficiency virus 1 (HIV-1). British Journal of Psychiatry, 159, 351-356. Maj, M., Satz, P., Janssen, R., Zaudig, M., Starace, E, D'Elia, L., Sughondhabirom, B., Mussa, M., Naber, D., Ndetei, D., Shulte, G., & Sartorius, N. (1994). WHO neuropsychiatric AIDS study, cross-sectional phase II: Neuropsychological and neurological findings. Archives of General Psychiatry, 51, 51-61. McManis, D. L., Figley, C., Richert, M., & Fabre, T. (1978). Memory-for-Designs, Bender--Gestalt, Trail Making Test, and WISC-R performance of retarded and adequate readers. Perceptual and Motor Skills, 46, 443-450. Mittelmeier, C., Rossi, J. S., & Berman, A. (1989). Discriminative ability of the Trail Making Test in young children. International Journal of Clinical Neuropsychology, 11, 163-I 66.

Color Trails

223

Reitan, R. M. (1959). Manual for administration of neuropsychological test batteries for adults and children. Tucson, AZ: Reitan Neuropsychological Laboratories. Reitan R. M. (1971). Trail Making Test results for normal and brain-damaged children. Perceptual and Motor Skills, 33, 575-581. Reitan, R. M., & Davison, L. A. (Eds.). (1974). Clinical neuropsychology: Current status and applications. Washington, I)C: Hemisphere Press. Reitan, R. M., & Herring, S. (1985). A short screening device for identification of cerebral dysfunction in children. Journal of Clinical Psychology, 41,643-650. Rosin, J., & Levett, A. (1989a). The trail making test: A review of research with children. South African Journal of Psychology, 19, 6-13. Rosin, J., & Levett, A. (1989b). The trail making test: Performance in a nonclinical sample of children aged l0 to 15 years. South African Journal of Psychology, 19, 14-19. Shute, G. E., & Huertas, V. (1990). Developmental variability in frontal lobe function. Developmental Neuropsychology, 6, l-11. Thorndike, R. L., Hagen, E., & Sattler, J. (1986). Stanford-Binet intelligence scale (4th ed.). Chicago: The Riverside Publishing Company. Wechsler, D. (1974). Wechsler intelligence scale for children-Revised. New York: The Psychological Corporation.