Using cerebral dominance for education programs

Using cerebral dominance for education programs

Leon J. Zalewski

FOR ALMOST 2 DECADES, many educators have promoted the concept that hemispheric dominance, or hemisphericity, is a fundamental part of learning style. Hemisphericity has surfaced repeatedly in a variety of periodicals, conferences, workshops, and self-help books and is even discussed in introductory educational psychology texts (Dembo, 1988). Well-known educators, such as Madeline Hunter (1976) and E. P. Torrance (1981, 1982), recommended that schools modify existing instructional methods and assessment procedures based on the existence of hemispheric specialities or dominance (Hardyck & Haapanen, 1979; O’Boyle, 1986; Springer, 1989).

Torrance (1981) stated that left-hemisphere functioning involves activities such as sitting erect and thinking, writing nonfiction, and learning algebra, whereas right-hemisphere functioning includes activities such as lying down and thinking, writing fiction, and learning geometry. According to Taggart and Torrance (1984), verbal and logical thought processes take place in the left cerebral hemisphere, whereas spatial and intuitive thinking take place in the right hemisphere. Left-hemisphere tactics can be characterized as “structured, verbal, facts, sequences, outline, and logical,” and right-hemisphere strategies are “open-ended, spatial, ideas, relationships, summary, and intuitive” (Taggart & Torrance, p. 2). More recently, Torrance and Rockenstein (1988) bisected creativity into behaviors associated with left-and right-hemisphere thinking.

Certain instructional methods, curricula, and assessment instruments currently used in schools are based on the notion of hemisphericity (Druckman & Swets, 1988; O’Boyle, 1986). For instance, Hunter (1976) stated that school curricula are primarily aimed at left-brained learners. Others corroborate this perspective, arguing that the educational agenda of most schools revolves around such left-brained subjects as reading, writing, and mathematics (Cooke & Haipt, 1986; Hopkins, 1984; Kane & Kane, 1979). Advocates for whole-brain curricula are concerned that teachers are unwittingly discouraging creativity with a disproportionate amount of left-brain instruction (Bogen, 1977; Samples, 1975; Torrance, 1981; Torrance & Rockenstein, 1988). Lord (1984) claimed that students who are high achievers in science tend to be more right-brained in their method of information processing, implying that science instructors should teach problem solving as a right-brained activity.

McCarthy, Hammond, and Samples (1985) developed a science methods text integrating Piagetian theory, learning styles, and hemispheric dominance research (see also McCarthy, 1980). Cooke and Haipt (1986), in a National Education Association publication, proposed an integrative or whole-brain teaching/learning model, and certain math educators have suggested that the implications of the literature on hemispheric dominance should be incorporated into the study of their discipline (Creswell & Gifford, 1988). Other whole-brain training programs were extensively reviewed in Druckman and Swets (1988). Some other educators even contend that the differences between the two hemispheres and the dominance (or lack thereof) can be used to determine certain childhood problems (see the review in Gardner, 1978, 1986).

There is a considerable amount of experimental support for the concept of hemispheric laterality, or dominance–especially in the domains of language and spatial abilities–but the proposed relationship between these findings and educational practice seems overly simplistic and perhaps even spurious (Calvin, 1983; O’Boyle, 1986; Springer, 1989). Although Gardner (1986) argued that the issue has produced a number of “excessive claims” about the processing differences between the two hemispheres, Hellige (1990) believes that “there is no scientific foundation for the idea |hemisphericity~” at all. According to Levy (1985), cognitive processing takes place very rapidly between the two hemispheres and therefore cannot be localized on one particular side of the brain. Similarly, Das (1989a), drawing on the seminal work of A. R. Luria (1973), theorized that cognition can be understood as the interdependent functioning of three neuropsychological subsystems: arousal and attention, coding, and planning. Das, Kirby, and Jarman (1979) examined both forms of the coding function, that is, simultaneous and successive processing, in which neither cognitive function is localized in one hemisphere or the other (see also Das 1984a, 1984b, 1989a). Springer and Deutsch (1989) apparently concur with Das’s position stating: “Almost any human behavior or higher mental function, however, clearly involves more than the actual specialities of either hemisphere and utilizes what is common to both hemispheres”.

In contrast, the concept of hemispheric dominance implies that simultaneous processing is a right-brained function and that successive processing is typically associated with left-brain activity. In light of this apparent conflict, using the results of hemispheric laterality research to guide and direct educational practice is highly questionable (Das, 1989b; O’Boyle, 1986). Springer and Deutsch (1989) challenged advocates of cerebral dominance theory to provide empirical evidence supporting their ideas (see also Springer, 1989).


As Hellige (1990) pointed out, recent popularized accounts have claimed that hemisphericity can be easily assessed and used to classify individuals as right-brained, left-brained, or whole-brained. One instrument currently used by educators to measure cerebral dominance is the Human Information Processing Survey (HIP Survey; Taggart & Torrance, 1984). The HIP Survey was originally designated as Form C, Your Style of Learning and Thinking (SOLAT; Torrance, Reynolds, Riegel, & Ball, 1977). Both the HIP Survey and the SOLAT were developed to sample the specialized cerebral functions of the right and left hemispheres (Taggart & Torrance). The test manual for the HIP Survey presents very limited reliability and validity data; however, Taggart and Torrance reviewed a number of empirical studies that evaluated the SOLAT’s psychometric properties. Unfortunately, most of the literature Taggart and Torrance reviewed has not been published and is thus unavailable for further scientific analysis. In addition, the papers discussed in Taggart and Torrance have severe methodological shortcomings. For instance, limited sample sizes were reported in the majority of the studies, and concurrent validity and construct validity were examined using instruments whose psychometric properties were inadequate (e.g., Torrance Test of Creative Thinking, Creative Motivation Scale, What Kind of Person Are You?, Rorschach Inkblot). When the issue of reliability was addressed, only the SOLAT’s test-retest coefficients were significant. However, these reliability estimates were inconsistent; they ranged from .55 to .86.

Das (1989b) reviewed the HIP and its predecessor, the SOLAT. Using Fitzgerald and Hattie’s (1983) research, Das found these instruments completely lacking in empirical support. Moreover, the theoretical foundation for these tests is extremely weak. Das’s (1989b) assessment was that the HIP (and, by extension, the SOLAT) has “such an unjustified theoretical base that its validity as a measure of thinking style of left-brained and right-brained people is questionable, hence its unreliability will be of only academic interest to the users”.

Given the methodological problems of previous validation studies and the dubious nature of the hemisphericity construct, a statistical reassessment of the SOLAT was required. Thus, we evaluated the SOLAT’s underlying psychometric properties and their implications for educational practice.


Subjects and Procedure

The investigation was conducted in three phases:

Phase I. To determine construct validity, we administered the SOLAT to a group of undergraduate and graduate students (N = 235) at a small midwestern university that was located in a large metropolitan area and served a multiethnic student population (median age = 35).

According to cerebral dominance theory, the SOLAT categorizes individuals as left-brained, right-brained, or whole-brained (integrated processors). We hypothesized that if the SOLAT has construct validity, a principal factor analysis performed on the SOLAT would approximate a three-factor solution corresponding to the three modes of information processing.

Phase 2. To determine the internal consistency and test-retest reliability coefficients, we administered the SOLAT to a second sample (N = 124) at the same university. Six classes of undergraduate and graduate students completed the survey approximately 6 weeks apart.

Phase 3. To further examine construct validity, we administered the SOLAT to both normal and brain-injured (BI) adults. The BI sample consisted of 74 adults who had sustained a brain injury after the age of 13, were enrolled in a Southern California community college, and possessed normal fluid intelligence. The median age for the BI group was 28 years. The mental ability of the BI group was ascertained from the subjects’ total scores on the Raven Standard Progressive Matrices (Raven, 1958). (The average number of correct responses was 46.04, SD = 8.62.) Due to the nature of the subjects’ brain injuries, the SOLAT was individually administered by a trained assistant. A comparison group of normal subjects (N = 74) was randomly selected from the group of university students discussed in Phase 1.


Phase 1. The dimensionality of the SOLAT was determined by a principal factor analysis, with the maximum likelihood procedure used for the communality estimates and Promax oblique rotations. The number of factors extracted was based on the preliminary eigenvalues, the scree plot, and the percentage of variance each factor contributed to the total variance. In marking specific factors, the minimum correlation of an item with a factor was set at .30, and correlations of this magnitude or higher were retained for subsequent analyses.

Of the 50 items used to assess cerebral dominance, 41 were found to be useful in defining the dimensions of the scale. Items 1, 8, 28, 31, 41, 43, 44, and 49 were eliminated from further analysis because of their low correlations with the seven factors. These items were used to assess such traits as memory for names and faces, ability to find direction, a preference for either algebra or geometry, and a preference for either dogs or cats.

The factor structure for the 41 items used in the analysis is shown in Table 1. A total of seven factors were extracted, accounting for 100% of the variance. Factor 1, which contributed 39% of the variance, comprised seven items that primarily involved a preference for words or images in learning and explanations. Factor 1 was labeled Verbal vs. Visual. The second factor, which contributed 17% of the variance, was defined by eight items that involved approaches to solving problems. Factor 2 was defined as Style of Problem Solving. The third factor, which explained 13% of the variance, was defined by seven items that involved various aspects of planning and organization. Factor 3 was called Planning and Organization. The fourth factor, which accounted for 10% of the variance, was defined by five items that assessed TABULAR DATA OMITTED insight from poetry, symbols, or inventiveness. Factor 4 was labeled Creativity. The fifth factor, which explained 7% of the variance, was defined by five items generally involving humor and playfulness and was called Humor. The sixth factor, which also contributed 7% of the total variance, was defined by five items that involved the intellect versus emotions. Factor 6 was labeled Logical vs. Intuitive. The seventh factor contributed 6% of the variance and was defined by four items related to a preference for reading either fiction or nonfiction books. Factor 7 was labeled Reading Preference.

The factor intercorrelations are presented in Table 2. Note that nearly all the correlations between the factors were negative, and relatively small. The strongest correlation, which was negative, appeared between Factor 1 (Verbal vs. Visual) and Factor 3 (Planning and Organization), r = -.26. Overall, it seems that the seven dimensions of the SOLAT are independent and unrelated to each other.


Reference Axis Correlations

Factor 1 2 3 4 5 6 7

1 1.00

2 -.11 1.00

3 -.26 -.20 1.00

4 .00 -.22 -.23 1.00

5 -.23 -.13 -.01 -.00 1.00

6 -.14 .16 -.07 -.02 -.04 1.00

7 -.02 -.09 .02 -.11 -.07 -.11 1.00

Phase 2. The internal consistency (Cronbach’s alpha) and test-retest reliability coefficients for each factor are presented in Table 3. The data show low to moderate reliability coefficients for the seven dimensions during Time 1. The highest reliability estimate (r = .61, for Factor 1) accounted for only 36% of the total variability. The internal consistency reliability coefficients for Factor 5 (Humor) and Factor 7 (Reading Preference) are clearly unacceptable (r = .35 and r = .34, respectively). The internal consistency reliabilities computed for Time 2 across each dimension were equally low, ranging from .26 for Factor 5 to .59 for Factor 1. The test-retest reliability coefficients computed from Time 1 to Time 2 were, at best, marginally acceptable, ranging from .65 to .77.

Phase 3. We conducted a one-way ANOVA between the normal group (N = 74) and the BI group (N = 74) on each dimension of the SOLAT. For six of the seven factors, there were no significant differences between the two groups. The exception was for Factor 3 (Planning and Organization). The normal subjects’ responses indicated a somewhat more right-brained processing style (M = 0.30, SD = 2.56), whereas the responses of the BI subjects (M = -0.89, SD = 2.59) reflected a tendency toward left-brained processing, F(1, 146) = 7.87, p |is less than~ .01. The means and standard deviations for each group on the seven factors are presented in Table 4. Note the large variability on each factor within each group.

To investigate potential differences in learning style or processing differences, we conducted an ANOVA using only the BI subjects. The independent variable was the localization of the cerebral damage–left hemisphere, right hemisphere, or diffuse (bilateral damage), and the dependent measure was the factor score on each of the seven dimensions. According to cerebral dominance theory, a subject would not prefer the learning style corresponding to the site of his or her brain injury. For example, a person with a right-hemisphere lesion would not prefer activities or tasks that require right-brained processing. The results of the seven ANOVAs indicated no significant differences in the total score on each dimension for the right-brain, left-brain, or bilateral BI students. The means and standard deviations for each group are shown in Table 5.


Internal Consistency and Test-Retest Reliability Coefficients

Internal consistency(a)

Factor Time 1 Time 2 Test-Retest

1 0.61 0.59 0.77

2 0.57 0.54 0.71

3 0.56 0.51 0.65

4 0.53 0.45 0.70

5 0.35 0.26 0.62

6 0.45 0.32 0.68

7 0.34 0.35 0.75

a Cronbach’s alpha coefficients.


Means and Standard Deviations for Normal and Brain-Injured

Adults on Each Dimension of the SOLAT

Normal Brain-injured

Factor M SD M SD

1 1.19 3.24 0.61 3.10

2 -2.08 3.41 -3.04 2.73

3 0.30 2.56 -0.89 2.59

4 0.28 2.30 -0.16 2.30

5 -0.30 2.14 -0.59 1.98

6 0.27 2.32 -0.39 1.86

7 0.38 2.07 -0.27 1.82

Note. N = 74.

A correlational analysis between the total score on each factor and the localization of the brain injury is presented in Table 6. There were no significant relationships between the location of the brain injury and the subjects’ responses on each of the seven factors.


Means and Standard Deviations for Left-Hemisphere,

Right-Hemisphere, or Bilateral Brain-Injured Adults on

Dimensions of the SOLAT

Left-hemisphere Right-hemisphere Bilateral brain

injury injury injury

Factor M SD M SD M SD

1 1.00 3.09 0.41 2.73 0.48 3.46

2 -3.19 2.48 -2.88 3.14 -3.07 2.64

3 -0.48 2.06 -0.75 2.79 -1.31 2.79

4 -0.19 2.44 -0.33 2.14 0.00 2.39

5 -0.29 1.92 -0.91 2.02 -0.55 2.01

6 -0.57 1.66 -0.46 1.77 -0.21 2.09

7 -0.14 1.82 -0.42 1.59 0.38 1.97


Pearson Product-Moment Correlations Between Localization of

Brain Injury and Total Score on Each Dimension of the SOLAT

Factor Injury

1 -0.07

2 0.05

3 -0.04

4 -0.03

5 -0.13

6 0.02

7 -0.07

Note. All correlations were nonsignificant.


Although the SOLAT’s psychometric value has been largely unsubstantiated, educators have used this paper-and-pencil survey in an attempt to measure cerebral dominance, or hemisphericity. Some well-known educational researchers advise that classroom instruction should be modified in accordance with students’ preferred mode of information processing. Thus, in theory, a student with a left- or a right-brain preference should be taught with this preference in mind, and the student’s curriculum and materials should also correspond to this preference. Other educators, suggesting that instruction should be directed at whole-brain thinking, emphasize an integrative learning-teaching model.

Although these ideas seem logical, the SOLAT does not have sufficient reliability and construct validity to warrant further consideration by teachers. Contrary to cerebral dominance theory as discussed by Taggart and Torrance (1984), the SOLAT’s factor structure approximates seven dimensions rather than three underlying and distinct modes of processing. When we administered the SOLAT to BI adults, we hypothesized that subjects with right- or left-hemisphere damage would not prefer a corresponding mode of information processing. However, the results of our study indicated that localization of brain injury has little or no effect on the various dimensions of the SOLAT. We further hypothesized that the factor structure for the normal and BI groups would be substantially different. In fact, normal and BI adults largely responded to the items in a similar fashion. We examined the SOLAT’s internal consistency and test-retest reliabilities, and, for the most part, the reliability estimates were quite low. Only the test-retest reliabilities corresponded to those presented by Taggart and Torrance (1984).

Moreover, the results of our study provide further support for the conclusions discussed in two reviews of the SOLAT (Das, 1989b; Fitzgerald & Hattie, 1983). Both reviews confirmed that the SOLAT is theoretically unsound, lacks a coherent scoring procedure, has faulty items with low reliabilities, and has little, if any, concurrent validity. “What can a reviewer say in regard to a test for which there is no empirical or theoretical basis, especially when the test seems to attribute multidimensional human behavior as managerial thinking to brain lateralization?” (Das, 1989b, p. 144).

Clearly, the fundamental problem with the SOLAT lies in the misguided attempt to link highly sophisticated neuroanatomical functioning in the right and left cerebral hemispheres with three discreet modes of information processing. As various researchers in neurology and neuropsychology have pointed out, the notion of hemisphericity is founded primarily on inconclusive evidence (Calvin, 1983; Das, 1989b; Druckman & Swets, 1988; Gazzaniga, 1985; Hellige, 1990; O’Boyle, 1986; O’Boyle & Hellige, 1989; Springer, 1989; Springer & Deutsch, 1989). Thus, brain lateralization research has limited application in education.

Although teachers need not alter or supplement their existing curriculum or teaching methods on the basis of a student’s SOLAT score, they should be sensitive to individual differences in cognitive ability. Understanding how children attend to a task, code various types of information, and deploy cognitive strategies can assist teachers in the educational process (Das, 1984a, 1986b; Das et al., 1979). In addition, rather than using a single teaching method, educators can use multiple methods of instruction to enhance student learning (O’Boyle, 1986). Finally, there is little empirical support for educational programs that supposedly train students to compensate for hemisphericity through teaching integrative processing techniques (Druckman & Swets, 1988; Hellige, 1990), and teachers would do well to avoid such quick fixes.

The notion of cerebral dominance has limited theoretical or practical value for educators, and any attempt to measure hemisphericity using the SOLAT will lead to highly questionable results. Teachers should not “rely on hemispheric laterality as a panacea for all our educational woes . . .” (O’Boyle, 1986, p. 43), but should use extreme caution when applying neuropsychological research to the classroom.


Bogen, J. E. (1977). Some educational implications of hemispheric specialization. In M. C. Wittrock, J. Beatty, M. S. Gazzaniga, H. J. Jerison, S. D. Krashen, R. D. Nebes, & T. J. Teyler (Eds.), The human brain (pp. 133-152). Englewood Cliffs, NJ: Prentice-Hall.

Calvin, W. H. (1983). Left brain, right brain: Science or the new phrenology? In W. H. Calvin (Ed.), The throwing madonna: Essays on the brain (pp. 102-107). New York: McGraw-Hill.

Cooke, J. K., & Haipt, M. (1986). Thinking with the whole brain: An integrative teaching/learning model (K-8). Washington, DC: National Education Association.

Creswell, J. L., & Gifford, C. (1988). Implications of right/left brain research for mathematics educators. School Science and Mathematics, 88, 118-131.

Das, J. P. (1984a). Aspects of planning. In J. R. Kirby (Ed.), Cognitive strategies and education performance (pp. 35-50). New York: Academic Press.

Das, J. P. (1984b). Intelligence and information integration. In J. R. Kirby (Ed.), Cognitive strategies and education performance (pp. 13-31). New York: Academic Press.

Das, J. P. (1989a). A system of cognitive assessment and its advantage over IQ. Paper presented at the School of Education, University of California, Riverside.

Das, J. P. (1989b). Human Information Processing Survey. In J. C. Conoley & J. J. Kramer (Eds.), The tenth mental measurements yearbook (pp. 363-364). Lincoln, NE: The Buros Institute of Mental Measurements, The University of Nebraska Press.

Das, J. P., Kirby, J. R., & Jarman, R. (1979). Simultaneous and successive processes. New York: Academic Press.

Dembo, M. H. (1988). Applying educational psychology in the classroom (3rd ed.). New York: Longman.

Druckman, D., & Swets, J. A. (Eds.). (1988). Enhancing human performance: Issues, theories, and techniques. Washington, DC: National Academic Press.

Fitzgerald, D., & Hattie, J. A. (1983). An evaluation of the ‘Your Style of Learning and Thinking’ Inventory. British Journal of Educational Psychology, 53, 336-346.

Gardner, H. (1978). What we do and don’t know about the two halves of the brain. The Harvard Magazine, March-April, 24-27.

Gardner, H. (1986). Notes on cognitive development: Recent trends, new directions. In S. L. Freidman, K. A. Klivington, & R. W. Peterson (Eds.), The brain, cognition, and education (pp. 259-285). New York: Academic Press.

Gazzaniga, M. S. (1985). The social brain: Discovering the networks of the mind. New York: Basic Books.

Hardyck, C., & Haapanen, R. (1979). Educating both halves of the brain: Educational breakthrough or neuromythology? Journal of School Psychology, 17, 219-230.

Hellige, J. B. (1990). Hemispheric asymmetry. Annual Review of Psychology, 41, 55-80.

Hopkins, R. L. (1984). Educating the right brain: Why we need to teach patterning. Clearing House, 58, 132-134.

Hunter, M. (1976). Right-brained kids in left-brained schools. Today’s Education, 65, 45-48.

Kane, N., & Kane, M. (1979). Comparisons of right and left hemisphere functions. The Gifted Child Quarterly, 7, 7-15.

Levy, J. (1985, May). “Right brain, left brain: Fact and fiction.” Psychology Today, pp. 38-44.

Lord, T. R. (1984). A plea for right brain usage. Journal of College Science Teaching, November, 100-102.

Luria, A. R. (1973). The working brain. New York: Penguin Books.

McCarthy, B. (1980). The 4 MAT system. Arlington Heights, IL: EXCELL.

McCarthy, B., Hammond, B., & Samples, R. E. (1985). 4-mat in science: Toward wholeness in science education. Barrington, IL: EXCELL.

O’Boyle, M. W. (1986). Hemispheric laterality as a basis for learning: What we know and don’t know. In G. D. Phye & T. Andre (Eds.), Cognitive classroom learning: Understanding, thinking, and problem solving (pp. 21-48). New York: Academic Press.

O’Boyle, M. W., & Hellige, J. B. (1989). Cerebral hemisphere asymmetry and individual differences in cognition. Learning and Individual Differences, 1, 7-35.

Raven, J. C. (1958). Standard progressive matrices. Cambridge, England: Lewis & Co.

Samples, R. E. (1975). Are you teaching only one side of the brain? Learning, 3, 24-30.

Springer, S. P. (1989). Educating the two sides of the brain: Separating fact from speculation. American Educator, Spring, 32-37.

Springer, S. P., & Deutsch, G. (1989). Left brain, right brain. New York: W. H. Freeman.

Taggart, W., & Torrance, E. P. (1984). Human information processing survey: Administrator’s manual. Bensenville, IL: Scholastic Testing Service.

Torrance, E. P. (1981). Implications of whole-brained theories of learning and thinking for computer-based instruction. Journal of Computer-Based Instruction, 7, 99-105.

Torrance, E. P. (1982). Hemisphericity and creative functioning. Journal of Research and Development in Education, 15, 29-37.

Torrance, E. P., Reynolds, C. R., Riegel, T. T., & Ball, O. E. (1977). Your Style of Learning and Thinking, Forms A & B: Preliminary norms, abbreviated technical notes, scoring keys, and selected references. Gifted Child Quarterly, 21, 563-573.

Torrance, E. P. & Rockenstein, Z. L. (1988). Styles of thinking and creativity. In R. R. Schmeck (Ed.), Learning strategies and learning styles (pp. 275-290). New York: Plenum.

COPYRIGHT 1992 Heldref Publications

COPYRIGHT 2004 Gale Group