1 Introduction
People are perennially interested in human intelligence. At least since the Āclassical Greeks, ideas stressing the existence and the importance of intelligence have occupied discussions of the human condition, and intelligence is among the most widely recognized and commonly accepted concepts in scientific psychology. Furthermore, the notions that individuals differ in intelligence, that such differences can be measured and studied scientifically, and that those individual differences have meaningful consequences in life rank among the significant ideas of human civilization. As one intelligence researcher observed, āNo concept in the history of psychology has had or continues to have as great an impact on everyday life in the Western worldā as intelligence (Scarr, 1989, p. 75).
The construct of intelligence is essential to understanding what it means to be human. Yet, vital as intelligence is to our self-knowledge, the concept has successfully eluded consensual definition. The histories of philosophy and of science are littered with failed attempts. A brief review is instructive ā if how far back to begin is challenging ā so we mention here only a smattering that cover the ages. Aristotle in Nicomachean Ethics (Book X. 7.14ā15) stated that āintelligence ā¦ guides us and ā¦ gives us our notions of what is noble and divine ā¦ it is an activity of this part ā¦ that will be complete happiness.ā Thomas Hobbes (1651/1982) in Leviathan (Part 1: Man, Chapter 1: Sense) placed intelligence at the core of what it means to be human: āan intelligent being seen; which, coming into the understanding, makes us understand.ā Alfred Binet (1905) referred to intelligence as āJudgment, otherwise called āgood senseā, āpractical senseā, āinitiativeā, the faculty of adapting oneās self to circumstances ā¦ auto-critique.ā David Wechsler (1958, p. 7), author of the most widely used tests of intelligence today, asserted that intelligence is āthe aggregate or global capacity of the individual to act purposefully, to think rationally and to deal effectively with his [sic] environment.ā And then thereās Alexander Wissner-Gross (2016), who defined intelligence this way: āF = T ā S Ļ ā: āIntelligence is a force, F, that acts so as to maximize future freedom of action. It acts to maximize future freedom of action, or keep options open, with some strength T, with the diversity of possible accessible futures, S, up to some future time horizon, Ļ. In short, intelligence doesnāt like to get trapped.ā
Various scholars, symposia, synopses, and summaries have appeared from time to time that have aggregated multiple definitions (Legg and Hutter, 2007; Sattler, 2008; Wasserman and Tulsky, 2005). A turn-of-the-20th-century symposium entitled āIntelligence and Its Measurementā convened to achieve consensus on the meaning of intelligence; published in 1921 in the Journal of Educational Psychology, it produced myriad definitions. Intelligence was variously described by luminaries of the time as the āability to learnā (Buckingham), āthe power of good responses from the point of view of truth or factā (Thorndike), āthe ability to carry on abstract thinkingā (Terman), āthe ability of the individual to adapt himself [sic] adequately to relatively new situations in lifeā (Pintner), āthe capacity for knowledge and the knowledge possessedā (Henmon), and āthe capacity to acquire capacityā (Woodrow). Sixty-five years later, when two dozen prominent theorists were asked to define intelligence, they offered two dozen definitions (Sternberg and Detterman, 1986). Later still, Sattler (1992) discussed more than 20 separate (yet somewhat overlapping) definitions of intelligence, many of which led to the development of distinct scales to measure a variety of skills and abilities. One 1997 āconsensusā statement (endorsed by 52 āexpertsā ā out of 131 invited to sign) read as follows: āIntelligence is a very general mental capability that, among other things, involves the ability to reason, plan, solve problems, think abstractly, comprehend complex ideas, learn quickly and learn from experience.ā By 2007, surveying dictionaries, encyclopedias, and professional associations, Legg and Hutter collected 70+ definitions but opined that ācompiling a complete list would be impossibleā (p. 1). Wikipedia (the source of all continuously updated knowledge) defines intelligence āto include the capacity for logic, understanding, self-awareness, learning, emotional knowledge, reasoning, planning, creativity, and problem solving. It can be more generally described as the ability to perceive or infer information, and to retain it as knowledge to be applied towards adaptive behaviors within an environment or context.ā In summary, famously, and perhaps out of exasperation to cut the Gordian Knot, E. G. Boring (1923, p. 35) proclaimed nearly a century ago simply that āintelligence is what is measured by intelligence tests.ā
In āIntelligence: Knowns and Unknowns,ā a late 20th century state-of-the-art report, the Board of Scientific Affairs of the American Psychological Association (Neisser et al., 1996, pp. 77) declared:
Individuals differ from one another in their ability to understand complex ideas, to adapt effectively to the environment, to learn from experience, to engage in various forms of reasoning, to overcome obstacles by taking thought. Although these individual differences can be substantial, they are never entirely consistent: a given personās intellectual performance will vary on different occasions, in different domains, as judged by different criteria. Concepts of āintelligenceā are attempts to clarify and organize this complex set of phenomena. Although considerable clarity has been achieved in some areas, no such conceptualization has yet answered all the important questions, and none commands universal assent.
A further contemporary challenge is recognition today that intelligence needs to be defined, assessed, and studied at (at least) three different levels: physiological, psychometric, and social.
The Latin root for the word intelligence translates roughly as āto comprehend or perceive.ā Perhaps, intelligence is best broadly defined as involving āthe ability to reason, plan, solve problems, think abstractly, comprehend complex ideas, learn quickly and learn from experienceā (Gottfredson, 1997, p. 13). No matter; intelligence is a subject of intensive study and continuing debate: See, for example, Dearyās (2001) Intelligence, Cianciolo and Sternbergās (2004) Intelligence, Golemanās (2005) Emotional Intelligence, Huntās (2011a) Human Intelligence, Bjorklundās (2011) Childrenās Thinking, Richieās (2016) Intelligence: All That Matters, and Haierās (2016) The Neuroscience of Intelligence.
If scientists cannot agree on how to define intelligence, many agree on its significance, many on the fact that people differ in how much intelligence they have, and many in the belief that intelligence can be measured. These certain features of intelligence are widely accepted.
ā¢ On significance: Intelligence is positively valued (Goodnow, 1986; Mugny and Carrugati, 1989), adaptive (Piaget, 1956; Sternberg, 1986), and involves solving problems (Laboratory of Comparative Human Cognition, 1982; Sternberg, 1986; Thorndike, Bregman, Cobb, and Woodyard, 1927). A lingering mystery is why there is so little consensus on exactly what it means for someone to be intelligent.
ā¢ On individual variation: Long ago, Francis Galton (1869) in Hereditary Genius sought to prove that people differed in their natural mental abilities. They do. A lingering mystery is why there is so little consensus on how and why one person is more intelligent than another.
ā¢ On measurement: Intelligence is an ability or cluster of abilities whose existence is normally inferred on the basis of performance on a test. What intelligence tests measure is tied to the issue of defining what intelligence is. Here, validity in its many construals comes into play. A lingering mystery is how best to measure intelligence.
Most definitions of intelligence used to determine the content of tests emphasize the face validity of specific abilities to (1) manipulate numbers; (2) think abstractly; (3) form relations among words, objects, or numbers; (4) adapt to novel environments; (5) apply past experience to the solution of current problems; and (6) adjust to real-world environments that are relevant to everyday life (Sattler, 1992). Moreover, for an intelligence test to qualify as a measure of intelligence, the test should have construct validity ā results that agree with some external, independent criterion of intelligence ā and predictive validity ā results that foretell future performance.
Educational and occupational attainment have been the most popular external validators of intelligence tests, meeting criteria of construct and predictive validity, as most intelligence tests have been designed to measure mental abilities that are important for success in academic work or in similar endeavors outside the classroom. Correlations between intelligence test scores and formal tests of reading, mathematics, or other subjects, and between intelligence test scores and school performance or grades, range between .40 and .70 (Brody, 1992; Jensen, 1980; Lavin, 1965; Vernon, 1947).
Beyond definition and measurement of intelligence, the very structure of intelligence has been elusive and controversial. What we think we know about the structure of intelligence originates in theoretical models that reciprocally shape the way intelligence has been defined, measured, and studied. Historically, two models of intelligence have dominated scientistsā understanding of its structure. Some theoreticians and researchers have contended that intelligence consists of a general ability that accounts for individual differences in performance in a wide variety of intelligence tasks and can be measured psychometrically; others have held that a single general ability provides an incomplete description of intelligence and argued for the greater relevance and validity of multiple abilities for specialized particular performance situations. This contrast has deep roots in philosophy and epistemology for it calls to mind the Greek philosopher and lyric poet Archilochus (c. 680āc. 645 b.c.) who distinguished people who view intellect as of a piece (āhedgehogsā) and people who favor fragmentation of intellect into components (āfoxesā); for modern treatments, see Gould (2011) and Berlin (2013). Hedgehogs believe in the significance of a singular intellectual capacity, that individuals are born with a certain amount of intelligence, and that intelligence can be measured and arrayed dimensionally so that individuals can be ordered in terms of an amount. An equally venerable tradition on the trail of foxes valorizes numerous distinct functions of mind, allows for profiles of individuals according to different manifestations, and calls on multiple domains of analysis. The school of hedgehogs (following Spearman, 1927) subscribes to the monoarchic āgeneral factorā view of intelligence; the school of foxes (following Thurstone, 1935) posits a loose confederation of āprimary mental abilities.ā Contemporary theoreticians of intelligence have merged the two into a composite hierarchical structure (following Carroll, 1993), as we do here. Accounts of the history of intelligence testing and of the leading figures in that history, as well as of the many controversies they generated, can be found in several sources (see Fancher, 1985; Gould, 2011; Sokal, 1987; Zenderland, 1998).
Following a short general introduction to mental testing, we briefly and critically discuss these three traditional conceptions of the structure of intelligence. We then proceed to consider the theoretical framework that undergirds the empirical study we present in this volume. There is long-standing and good agreement that human children and adults have intelligence and differ among themselves in intelligence, but there has been less confidence until relatively recently in the nature of the structure of intelligence. The work we relate in this volume speaks to the question of structure. Note that in this volume children have played key roles in the telling of this story of mental testing and intelligence.
Mental Testing
A significant impairment to addressing central questions about intelligence rests on the fact that intelligence per se cannot be observed. Rather, intelligence is inferred from behavior. Intelligence is a Kantian noumenon ā something that exists apart from human sense and perception. The dominant approach to measuring intellectual ability or abilities has been through standard psychometric mental tests; but theorists who contend that there are many different intelligences suggest by corollary that only a few have been captured this way. A mental test usually consists of collections of tasks, uses visually or auditorily presented verbal, mathematical, or symbolic materials, and requires examinees to perform the tasks. Successful performance on the tasks in mental tests is thought to call on one or more kinds of intellectual ability, where other characteristics (diligence, motivation) and other factors (gender, socioeconomic status) are thought to be less critical. Tests are usually administered individually, and testing requires an examiner with special training. The process of administering intelligence tests has changed remarkably little in the century since mental testing originated.
Intelligence test theory turns on certain assumptions: that different tasks yield reliable measures of characteristic behaviors of the individuals who perform the tasks; that individuals can be ordered on a scale reflecting different degrees of ability on those tasks; that abilities are distributed normally so that measurements can be transformed to scales with at least interval properties; and that successful performance on the tasks strictly reflects mental ability or cognitive competence of the individual.
As is by now well known, the first practically useful intelligence test was developed in the early 1900s by Alfred Binet with his collaborators, Victor Henri and ThƩophile Simon (Binet and Henri, 1898; Binet and Simon,1904). The introduction of (nearly) universal primary education had brought into French elementary schools numbers of children who had not previously attended school and seemed not to profit from normal classroom teaching. These children were deemed to be of apparently below-average intelligence and in need of special education. The problem was to devise a quick and inexpensive way of identifying them.
Based on extensive trial and error, Binet and his colleagues devised a series of tasks of increasing difficulty for 4-, 6-, 8-, and 10-year-old children (Binet and Simon, 1911, 1916). Binet realized that, by tapping a variety of cognitive tasks at different levels of difficulty and organizing the items according to the age levels at which children of normal intellectual functioning were likely to succeed, Binet and Simon could also arrive at a scale that would classify childrenās levels of mental functioning based on the number of items children passed at the various levels. In 1905, they introduced a new test for measuring intelligence, called the BinetāSimon scale. The test consisted of different kinds of simple-to-complex tasks. The first version was composed of clinical procedures that did not result in a summary score, but later revisions provided such a summary score (āmental levelā) that still later was developed into the concept of the intelligence quotient (Intelligenzquotient or IQ, defined roughly as mental age divided by chronological age by Stern, 1912) and soon after adopted by Terman (1916) at Stanford University in the United States (hence the StanfordāBinet). Binet thus designed the precursor to the modern-day intelligence test. By the 1920s, mental tests were used in schools to measure childrenās intell...