How and why scientists and researchers began moving to non-psychometric theories of intelligence?
Qi Li
The University of Georgia
The psychometric approach theory of intelligence has prevailed since Alfred Binet created back to the study of intelligence in the twentieth century, there is no doubt that psychometric theory of intelligence played an important role during that time. Psychometrics provided a good measurable way to study many different aspects of intelligence, and set standards to test intelligence. This approach, however, has been broadly criticized because it isolated intelligence rom many other factors (e.g., culture, social environment, ethics, gene, gender, nutrition) which may influence the development of intelligence So, researchers and scientists began to look for more diverse approaches to explore the nature of intelligence. This article focuses on discussing the process of this shift.
The earliest theory of intelligence backed to Darwin’s (1859) book, Origin of Species, which suggested that human intelligences evidenced by physical responses can be understood through scientific investigation because it is continuous with those of lower animals (Sternberg, 1990). In line with this, Galton (1883) proposed techniques for directly measuring the human behaviors (Sternberg, 1990). Galton argued that intelligence is composed of two general qualities: the capacity of labor and sensitivity. Capacity could be measured by the level of energy. Sensitivity could be measured how much information concerning external events reach us. It is the combination of Darwin’s theory and Galton’s measurement techniques that set up the psychophysical theory of intelligence
Psychophysical tests, however, ceased to play an important role in mental measurement by the turn of the century because they were found to show a chance pattern of correlations both with each other and with external criteria used to validate the tests (Sternberg, 1990). Wissler, for example, tested undergraduates at Columbia University using 21 of Cattell’s tests and found the tests correlated with each other at chance levels (Sternberg, 1990). The failure of these measures to correlate with each other and with external critieria suggested that they were not measuring a single construct of intelligence.
The next step in the development of psychometric tests occurred when, Binet and Simon devised tests to predict school performance and allow schools to correctly identify and school children with different mental abilities. This was done to insure that mentally retarded children received an adequate education. The differences between Binet and Simon’s research and Galton’s are not only the test items but also the original concerns. For Galton and his successor Cattell, the research grew out of pure scientific concerns and the belief that intelligence can be measured to the same exactness as the physical science. While, for Binet and Simon, the research grew out of practical educational concerns and the need to test various mental abilities related to school performance (e.g, vocabulary, verbal absurdities, comprehension) . The core of Binet and Simon’s conception of intelligence is judgment, which is manifested by three distinct elements: direction, adaptation and control. Binet developed mental test items which not only had construct validity but also differentiated between the performance of children of different ages and mental capacities. The psychometric approach became dominant as Binet and his successors attempted to document statically the mental abilities of human intelligence using the tests created by Binet.
As intelligence tests became more and more widely used, researchers began to explore the relationship between different IQ tests and the validity of test items. Statistical analysis of the different tests reveals a general factor, called “g”, which is defined as the largest common factor derived from IQ tests (Naglieri & Jenson, 1987). Since Binet’s research grew out of practical educational concerns, IQ tests and g theory are useful for predicting school achievement. However, IQ score can not explain individual intelligence in everyday life. For example, some individuals may have a high verbal ability but will perform poorly in storytelling because he or she does not know how to deal with the audience’s feedback. It will be difficult for IQ tests to explain the performance in non-academic settings because of the complexity of many tasks.
In addition, Kornhaber, Krechevsky & Gardner (1990) stated that “a focus on testing for an allegedly general ability is no longer tenable”(p.188) They claimed that IQ tests lack the ability to assess intelligent performance in authentic settings. Thus, Excessive reliance on psychometric tests not only isolated individuals from teachers who evaluated their performance in a social context but also separated individuals from the domains of knowledge valued by the society (Kornhaber et al., 1990). They claim that intelligence tests did not measure how people function when they are able to draw upon their experiences, feedback, and knowledge as they typically do.
Out of the dissatisfaction with IQ tests to measure and explain intelligence, scientists and researchers sought non-psychometric approaches to probe the nature of intelligence. Two of the influential theories are Sternberg’s triarchic theory of intelligence and Gardner’s theory of multiple intelligences. The two theories respond to the criticisms of intelligence theory by incorporating context as a part of theory.
Triarchic theory shifted the focus away from psychometric approach by proposing a general intelligence composed of three subtheories. The first is context of intelligence, which emphasizes the roles of adaptation to, selection of and shaping of environments in attaining fit to the environmental contexts where one lives. This subtheory states the appropriate understanding and measuring of human behavior should be in a given sociacultural setting. The second is experience and intelligence, which specifies the role of intelligence in individual’s interaction with the task or situation. This subtheory focuses on two kinds of ability: one is to deal with novel kinds of task and situational demands; another is to automatize the processing of information. The third is components of intelligence, which deals with the mechanisms by which intelligent behavior is accomplished. There are three kinds of process: metacomponents for planning, monitoring and evaluating of task performance; performance components used in actually performing the tasks one faces; and knowledge-acquisition components used in learning how to perform tasks.
Another shift from psychometric theory of intelligence is the Multiple Intelligences, developed by Gardner. Multiple intelligences theory is different from Sternberg’s triarchic theory which maintains a general intelligence by proposing six kinds of intelligences without any one being dominant or superior to the others. The assumption is that there are unique and independent intelligences and that individual can be encouraged to develop problem-solved abilities unique to individual intelligences.
In sum, scientists and researchers are moving away from psychometric theories to non-psychometric theories of intelligence. Due to the complexity and variousness of human behavior, it is necessary to correlate the human behavior with the context where the behavior happened. More comprehensive and deep researches focusing human behavior, context and their relationship are needed to contribute the knowledge of theories of intelligence.
References:
Gardner, H. (1983). Frames of the mind. New York: Basic books, Inc.
Kornhaber, M., Krechevsky, M., & Gardner, H. (1990). Engaging intelligence. Educational Psychologist, 25, 177-199.
Naglieri, J. A., & Jenson, A. R. (1987). Comparison of black-white differences on the wisc-r and the k-abc: Spearman's hypothesis. Intelligence, 11, 21-43.
Sternberg, R. J. (1990). Metaphors of mind: Conception of the nature of intelligence. Cambridge, MA: Cambridge University Press.