Questioning our concept of intelligence

New theories of intelligence contend that conventional I.Q. tests do not adequately sample the range of tasks needed to assess intelligence. In addition, new theories do not support a “general factor” of intelligence. They suggest there is more than one kind of intelligence and that different intelligences don’t overlap all that much.

For example, Howard Gardner, Harvard University, espouses a theory of multiple intelligences comprising seven abilities that he believes are distinct and relatively independent:

  1. linguistic intelligence, used in reading a novel, writing a poem or article or giving an extemporaneous talk
  2. logical-mathematical intelligence, used in solving math problems and proving logical theorems
  3. spatial intelligence, used in finding one’s way in unfamiliar territory or figuring out how to pack something most efficiently
  4. musical intelligence, used in remembering a tune, singing or composing
  5. bodily-kinesthetic intelligence, used in dancing or performing a sport
  6. interpersonal intelligence, used to figure out what other people mean from what they say or the way they look and deciding the appropriate way to respond
  7. intrapersonal intelligence, used in understanding ourselves — why we take rejection poorly or tend to be confident

Gardner contends that conventional I.Q. tests measure only the first two and sometimes the third. He notes that the form of measurement on these tests is quite limited, often using only multiple-choice items or very restricted kinds of performances.

Three types of ability

Another recent theory of intelligence is Robert J. Sternberg’s triarchic theory of human intelligence. As he describes it in a recent article in The American Scholar, his theory developed from his observations of graduate students at Yale. Students applying to the graduate program usually test very high on traditional I.Q. tests and earned very good grades and high S.A.T. scores.

However, through the years, Sternberg observed that some of these top students did not seem capable of original thought, while others, some of whom had been accepted with reservations about their ability, were outstanding in the doctoral program.

Sternberg’s theory holds that intelligence has three major aspects: analytical, creative and practical. Conventional good test takers and good students tend to excel in analytical intelligence but not necessarily in the creative or practical aspects of intelligence. Sternberg contends that the reason traditional I.Q. tests predict school achievement is that schools, like these tests, emphasize analytical skills far more than creative or practical ones. In fact, he states that practical and creative skills are sometimes discouraged in schools.

Sternberg reports that traditional intelligence tests have changed little in almost a century. The first major test was designed to distinguish students who were genuinely lacking in academic abilities from those who posed behavior problems but had academic potential. Throughout the development of intelligence tests there has been a close association between tests and academic performance, which the tests were primarily designed to predict.

Matching ability to instruction

In a five-year study, Sternberg and colleagues sought to determine if prediction of college performance would improve if broader tests were used to measure potential. Even more importantly, the researchers asked whether students would perform better if they were taught, and their achievement assessed, in ways that reflected their own patterns of ability. In this study of Yale graduate students researchers specifically looked at the effects of matching versus mismatching abilities to instruction.

All students were evaluated for memory as well as for analytical, creative and practical abilities. Memory was measured by standard multiple-choice, factual recall questions. Analytical tasks required students to compare and contrast two scientific theories or critique an experiment. Creative measures required students to generate their own theories and experiments, and practical measures had students show how theories and experiments could be applied to specific case studies.

Results revealed that the three abilities were related but distinct from one another. No general ability factor was indicated across the different measures. Analytical ability always contributed to the prediction of academic achievement, but achievement was more accurately predicted when all three abilities were measured.

Most important, students performed better when the kind of instruction they received matched rather than mismatched their patterns of ability. In other words, students achieved better when they were taught in a way that recognized and encouraged their particular pattern of skills. Therefore, Sternberg reports, when we teach and assess more broadly, taking into account creative and practical skills, we find that many students are considerably more intelligent than we might have expected.

Recognizing pragmatic and creative abilities

Sternberg points out that this information is particularly important in a society as stratified as ours is along racial and ethnic lines. In this study, students who were identified as high-analytic looked like a typical group of “good students”: mostly white, and from economically and socially privileged backgrounds. But students chosen for high creative or practical scores were much more diverse racially, ethnically and economically.

Sternberg suggests that by creating a rather closed system that rewards one type of student on ability tests, in classrooms and in college admissions, we may be missing a lot of what other segments of our population have to offer our society. And our failure to recognize these students’ abilities can have quite serious implications for their careers. We have created a system that values certain kinds of abilities, but not others.

The so-called cognitive elite is no fact of nature; it is something we have created and it is based on a limited kind of cognitive ability. As an analogy, Sternberg writes that if we were to admit only tall people to competitive colleges, we would eventually find that most individuals in professional fields and executive positions were tall. But we know better than to assume that being tall is a prerequisite for these positions.

Incomplete view of intelligence

Sternberg concludes that these studies using new theories of intelligence show that conventional views of intelligence are not wrong, but just very incomplete. They deal with only a small portion of what some scientists now believe intelligence to comprise. Although theories vary, researchers like Sternberg and Gardner are united by their belief that I.Q. represents too narrow a conception of human intelligence. In addition, their studies support the notion that intelligence can be modified.


“What Should We Ask About Intelligence?” The American Scholar Volume 65, Number 2, Spring 1996 pp. 205-217.

Published in ERN September/October 1996 Volume 9 Number 4

Leave a Reply

  • (will not be published)