Volume 34 Number 2
Winter 1995

American Indian children have the highest dropout rate of any ethnic group in the country and current statistics suggest that as many as 50% of them currently enrolled in school may not graduate (Bowker, 1993). Only 9.3% of American Indians, Eskimos, and Aleuts hold bachelor's degrees compared with 20.3% of all persons 25 years and older. The number of American Indian students pursuing a graduate degree increased only slightly during the period from 1971 to 1988 (Cross, 1991). Obviously, the academic pipeline for the American Indian people is leaking badly. Low teacher expectations and counseling of Indian students into vocationally-oriented curricula have been identified as among the factors contributing to this attrition (Bowker, 1993). Since college admission is often contingent upon standardized test scores and high teacher expectations are linked to high achievement test scores, low teacher expectations may directly influence the number of American Indian students who stay in the pipeline.

Kleinfeld and McDiarmid (1987) found that teachers in rural Native schools in Alaska held significantly lower expectations both of their students' abilities and prospects for college than did teachers in other schools. The teachers claimed that they were merely being realistic rather than displaying low expectations toward their students. Students attending the American Indian Science and Engineering Society (AISES) conference expressed concern over the inadequacy of counseling they had received about college academics and career choices. They reported feeling steered away from graduate school and the professions (Hoover & Jacobs, 1992).

At the turn of the century, desire to predict academic success led Binet to design a test for placing Paris boys into appropriate educational settings. In the United States as early as the 1890s, efforts were underway to survey the abilities of "the white race" with the goal of predicting "success in schooling" (Boring, 1950, p. 572). Predicting academic success for American Indian children was not an agenda item in the 1890s.

Existing intelligence tests correlate about r = +.50 with academic success for White children. These tests have some utility for doing what they were constructed to do, namely predict academic potential for White children. As a group, American Indian children score below average on many such standardized tests. Like Mexican-American, Puerto Rican, Chinese, and Japanese children, American Indian children tend to score higher on the nonverbal than on the verbal portions of these tests. Early efforts to construct culture-fair non-language assessment instruments were abandoned because they did not achieve the same level of predictive validity as the conventional tests (Cohen, Swerdlik, & Smith, 1992).

A pattern of performance has emerged from a number of studies of American Indian cognition which suggests this population may have an aptitude for mathematics (Bowker, 1993). Using Bannatyne's recategorization of the Wechsler Intelligence Scale for Children (WISC), this pattern has been described as better spatial than sequential ability and better sequential than verbal conceptual ability (Diessner & Walker, 1986). This cognitive style is similar to that which has been reported for Asian children in cross-cultural studies. However, studies comparing White and Asian children have been faulted for evaluating children in one country with norms obtained in another (Stigler & Stevenson, 1991). The problems of cultural equivalence in standardized cognitive ability testing has been explored in detail (Helms, 1992).

Efforts have been made to design noncognitive instruments which can predict the likelihood of academic success for minority students. The Noncognitive Questionnaire (NCQ) was devised as an alternative to standardized tests for use as an admission screening tool for ethnic minority students. Although the NCQ is reported to be a useful predictor of academic success for AfricanAmericans, international students, and Asian Americans, the NCQ items were not significantly related to either grade point average or to retention over three semesters for American Indians (Paris, Bennett, Sedlacek, & Kanak, 1993).

Given the difficulties inherent in assessing children with instruments normed for another culture, the failure of early efforts to devise culture-fair tests, and the lack of predictive validity for noncognitive measures for American Indians, the problem of finding a means for predicting academic potential for success for this population remains problematic. The concerns expressed by the AISES students are legitimate-Indian students with academic potential need to be advised about career options and tracked into college preparatory courses. Providing teachers and guidance counselors with objective and valid means of identifying potential for academic success for this population can facilitate these students remaining in the academic pipeline.

A new model of intelligence, derived from the work of the neuropsychologist Luria (1966), is emerging in the neurosciences. This new model views intelligence in terms of information processing capacity. The human brain has two fundamentally different ways of processing information: sequential processing and simultaneous processing (Springer & Deutsch, 1993; Zaidel, 1994). Sequential processing arranges and rearranges information, step-by-step, until it makes sense. Simultaneous processing, on the other hand, relies on mental representations of images. This kind of processing is called into play for thinking about relationships between things.

The study described here employed a battery of neuropsychological tests which assess sequential and simultaneous processing (see Note 1). The original study was part of Dingman's (1992) dissertation research. Three tests of simultaneous processing were found to be significant covariates to number of quarters in college, an important measure of academic success for American Indian students.

Subjects
American Indian students who participated in the study, 40 men and 40 women, were volunteers who were enrolled in colleges or universities in Montana. These students came from a variety of linguistic and cultural groups. Seventeen of the American Indian students (21%) indicated they had first learned a language other than English in their homes. Frequencies of self-reported tribal affiliations are shown in Table 1.

The mean percent of American Indian ancestry reported for the students was 61% (S. D. 29%). They had been enrolled in college an average of 5.7 quarters (S. D. 5, range 1 to 20 quarters), their mean reported grade point average was 2.70, and their mean age was 26 years (S. D. = 8 years). The vast majority of the American Indian students who participated in this study, 74 out of 80, reported that they intended to complete a four-year college degree.

Materials
A self-report questionnaire was used to collect demographic information from students on sex, age, quarters-in-college, college grade point average, future academic plans, bilingualism, and ethnic background, including percent of American Indian ancestry and tribal affiliation.

The Cognitive Laterality Battery (CLB) consists of eight tests of cognitive function, presented along with instructions, on 35mm slides and audio cassettes using a sound/sync projector. Devised by Harold Gordon (1983), the CLB consists of tests selected to assess qualitatively different cognitive functions. Four of the tests assess Sequential processing: Serial Numbers, Serial Sounds, Word-Production-by-Letter, and Word-Production-by-Category. The other 4 tests assess Simultaneous processing: Form Completion, Localization, Orientation, and Touching Blocks. These tests have good internal reliability and the CLB had been administered to more than 4000 people in four countries by 1985 (Gordon, 1985).

A complete description of all eight tests in the battery and their norms are available from Dr. Gordon or in the literature (Gordon, 1983). Three tests of particular importance to the results presented here, Localization, Form Completion, and Orientation, are described below. These measures are only weakly correlated to more familiar tests such as the Block Design subtest on the Wechsler Adult Intelligence Scale (WAIS). Gordon (1983) reports that the correlation between performance on the Block Design subtest and the Orientation test of the CLB is r =.35. The correlation between the Block Design and the Form Completion tests is reported to be only r = .25. The Form Completion test is a nearly pure measure of simultaneous processing capacity, as is the Orientation test when a mental rotation strategy is used by the test-taker.

Localization
The Localization test frame, in which an "X" is displayed, is projected on the screen for 3 seconds. A blank frame, printed in the answer booklet, has the same proportional dimensions as the one which appeared on the screen. The student's task is to mark on his answer sheet, the exact location of the X as it appeared on the screen. A student's score on this task is his total error measured in millimeters over 24 trials. The resulting score was multiplied by -1.00 for this analysis in order to make this test conform to the direction of the others in the CLB.

Orientation
Each stimulus, presented for 15 seconds, consists of a slide showing 3 three-dimensional constructions of 10 stacked cubes. Two of the constructions are identical and the third is a mirror image of the other two. The 3-D constructions are rotated in space around a vertical axis. The student's task is to indicate in his answer booklet which 2 of the 3 constructions are identical. Twenty-four stimulus slides are presented and the student receives credit for correctly identified pairs of figures.

Form Completion
Each stimulus slide consists of six incomplete silhouette drawings of common objects or scenes. Four of these stimulus slides are presented for 45 seconds each. The subject's task is to identify each of the drawings in his answer booklet. The student receives credit for each drawing correctly identified.

Procedure
Prior to each administration of the battery, an envelope containing a questionnaire and the answer booklet for the Cognitive Laterality Battery was placed at each seat in the testing room. A statement explaining the research was read to participants at the beginning of the session. After questions had been answered, students were asked to open their envelopes and to complete both sides of the questionnaire. The CLB was then presented on a sound/sync slide projector.

Because the dependent variables (scores on the eight subtests of the CLB) in this study are correlated, a Multivariate Analysis of Variance (MANOVA) covariate analysis was used to investigate whether measures of academic success were related to performance on any of the CLB tests.

MANOVA covariate analysis for college grade point average and performance on the 8 tests of the CLB did not approach statistical significance. MANOVA covariate analysis for another measure of academic success, quarters-of-college, was significant (F = 1.88, p =.07, alpha = .10). Univariate F-tests for the covariate of quarters of college was significant for 3 of the Simultaneous processing measures in the battery: Orientation (t = 1.88, p = .06), Form Completion (t = 1.95, p = .05) and Localization (t = 2.30, p = .02).

This investigation revealed that three tests of Simultaneous processing are positively correlated with number of quarters of college enrollment for American Indian students. American Indians who remain enrolled in college may be especially adept at using their simultaneous processing capacity for integrating and synthesizing many parallel problems at the same time. Learning style preferences reported for American Indians may reflect an underlying cognitive strength for simultaneous processing (Walker, Dodd, & Bigelow, 1989). The verbal-sequential tests in the Cognitive Laterality Battery were not significant covariates of either measure of academic success for these students.

Localization, Orientation and Form Completion tests are not measures which are typically included on standardized intelligence tests. However, the Kaufmann Assessment Battery for Children (K-ABC) (Kaufman & Kaufman, 1983) which was developed using an information processing model of intelligence includes a Gestalt Closure test and a test of Spatial Memory (Cohen, Swerdlik & Smith, 1992). The Gestalt Closure and Spatial Memory test of the K-ABC assess the same cognitive functions as the Form Completion, Localization, and Orientation tests of the CLB.

Since capacity for simultaneous processing appears to be related to remaining in college for American Indian students, it is hoped that American Indian norms will be established for the K-ABC and that other measures will also be constructed for use with this population. These measures may provide objective and valid means for identifying American Indian students with academic potential who are not currently being identified.

1. The findings reported here are from a doctoral dissertation which was supported by the Minority Fellowship Program of the American Psychological Association. Certain of the results obtained appeared to have educational implications for American Indian students which warranted their submission to the Journal of American Indian Education.