People are not all the same. There are young people, old people, and people in between. Nevertheless, we do make distinctions, ranging from binary decisions about the right to vote to the minimum age at which one qualifies for a pension. Many of these distinctions are based on the underlying presumption that intelligence grows, and then declines. There are mandatory retirement age requirements for air traffic controllers and commercial aviators, largely because of our beliefs about changes in their cognitive capabilities. In the United States federal judges are appointed for life. Many of the individual states have mandatory age limits for judges.

There are men and women, a sharper biological distinction than young and old. Historically, many societies have assumed that men and women have different cognitive capacities. In contemporary post-industrial societies there is a presumption of equality. But does “equality” mean “identity?” Should we regard a low incidence of women in corporate law practice with suspicion? Should the acceptable percentage of women in corporate law practice be the same as the acceptable percentage of women working as helicopter pilots?

People are identified with, and self-identify with, various racial and ethnic groups. Do these groups differ in intelligence? To some people the answer is obvious – they do. Other people believe that even the suggestion of a difference is evidence that the speaker is prejudiced.

The quotation from The Bell Curve: Intelligence and Class Structure in American Life, is a pretty strong statement about the importance of intelligence. When Herrnstein and Murray made it they were attacked as elitist and antidemocractic. Other people, with impeccable democratic credentials, had said similar things in a less contentious way. Just a few years before Herrnstein and Murray wrote, Robert Reich, a sociologist who had served as Secretary of Labor in the Clinton administration, wrote that work has shifted from emphasizing the manipulation of objects to the manipulation of abstract ideas, varying from programming a robot to analyzing a financial system. It follows that skill in manipulating abstract concepts, intelligence, has become progressively more valuable over time. To what extent do cognitive tests predict such skill?

Problems in Investigating the Relationship between Intelligence and Success

This chapter examines the relation between intelligence and success in three broad regions; academics, the workplace, and personal life. These studies are not easy to do, for several reasons. First, we have to specify what we mean by success in each arena. Next, we have to select quantitative measures of success. Observable measures, such as grade point average or money earned, are frequently only partially satisfactory measures for our criteria.

Walter Lippmann had more to say about intelligence than was quoted in Chapter 1.

This criticism strikes at the heart of psychometric theory. Lippmann believed that “real” intellectual talent has little to do with a talent for test taking. His criticisms have been echoed, with surprisingly little change, in today's world.

The gist of the modern criticisms of intelligence testing is that the tests capture a very narrow slice of human cognition. The following quotes are from books written for the general public:

Shakespeare was right (again). Every experience we have leaves an imprint on our brains, and from it, on our minds. Clearly physical experiences can change the brain. It has been claimed that early species of Homo got a leg up on the evolutionary ladder when they began to eat fish. In modern days, lecithin, a substance found in a number of foods, including fish and eggs, has been studied to see if it can enhance learning. (The results are mixed.) We do not stop there; we concern ourselves with the social environment. If we do not believe that social experiences can affect the brains of children, why do we have decency ratings for movies and television programs? And why would it be possible to sell video programs for infants with names like Baby Einstein? How might we manipulate the environment to improve intelligence?

There are two ways this question can be interpreted. The less interesting interpretation is “Can the environment be manipulated to improve test scores?” A more interesting question is “Can the environment be manipulated to improve general mental competence?” The answer to both questions is “yes.”

Environment is a catch-all term. In discussing environmental effects on intelligence it is useful to make a distinction between the physical environment and the social environment. The physical environment involves things like nutrition, air pollutants, and disease – anything that makes itself felt by direct physical action.

Introduction

Thomas Jefferson and John Adams, the men who wrote in the Declaration of Independence that “all men are created equal,” believed in a natural aristocracy! They weren't hypocrites. Jefferson and Adams believed that all men had equal rights, not that they had equal talents. Adams's lifelong support of his alma mater, Harvard University, and Jefferson's investment in the University of Virginia showed how strongly both men supported the development of the natural aristocracy. However, neither Jefferson nor Adams said how the natural aristocrats were to be identified. Today we partially rely on tests to do this.

We all have an intuitive feeling about what a cognitive test is, for you are unlikely to grow up in post-industrial society without taking one. This chapter expands on intuitive notions by describing some of the typical tests, chosen to represent different types, and by commenting on the aspects of intelligence that they do, or do not, evaluate.

Most textbooks on testing stress the distinction between individual testing and testing people in groups. An equally important distinction is between tests composed of batteries of subtests and single-format tests.

Introduction

The epigraph, taken from the 1909 Gilbert and Sullivan operetta Princess Ida, almost gets the scientific facts right. In the interest of scientific accuracy the lines should read

for there are some key genetic differences between apes and monkeys. We and our other ape cousins have lost those elegant monkey tails, developed a different limb structure, and have bigger brains. The ape-monkey genetic changes took place twenty million years ago; five millions years ago more genetic changes produced the hominids. Somewhere around 100,000 years ago our own species, Homo sapiens, a very big-brained species with a limb structure that is fine for walking but terrible for climbing trees, began to spread from our place of origin in Africa, and by somewhat less than 15,000 years ago (the date is still debated) had reached the southern tip of South America, thus populating all the continents except Antarctica.

Genetic change did not stop as humans wandered across the globe. A very large majority of the genes that define humanity are shared by all humans, but there are significant genetic differences between populations. Swedes, Tanzanians, Chinese, and the Quechua Amerindians who live in the Andes differ in their appearance – within the range of human differences – and they differ in other ways as well.

A colleague of mine once observed that the ending sections of all scientific discussions could be collapsed to a single word – much. Much has been learned and much remains to be learned. That is certainly true of intelligence. But let us be more specific.

A Summary

Scientific research on intelligence is just over a century old. There has been progress; it just has not been as rapid as we might have hoped. A great deal of this research has come in spurts. As in many other sciences, these spurts have occurred when technological or historic developments open up new sources of relevant data. In research on human intelligence that has happened three times. The development of group testing initiated by the Army Alpha examination in World War I showed that valid intelligence measures could be obtained without expensive one-on-one interviews. The result was an explosion of applications, and data, in both academic and industrial settings. In the 1930s the development of mathematical techniques for handling multivariate data made possible the evaluation of sophisticated psychometric models of human intelligence. The impetus of this development was magnified with the development of computers, which made computation-heavy methods of analysis possible. Since about 1985 we have been busy digesting data from a new source – the extensive neuroscientific measurements that are now possible, These include noninvasive techniques for looking directly at indicators of functioning in the normal brain.

Overview

This chapter and the next discuss theories of intelligence. This chapter describes the role that theory plays in our efforts to understand intelligence. The discussion is necessary because there has sometimes been a lack of clarity about what theories of intelligence are and how they are to be used. Hopefully the discussion here will provide some principles to apply in analyzing specific theories, as is done in Chapter 4.

The current chapter begins with a brief discussion of theory in general. It then moves to a discussion of some special problems facing theories of intelligence. Finally, I contrast two different goals for theories in psychology, and consider why these goals may produce different theories.

Scientific and Nonscientific Theories

Outsiders sometimes see science as a process of collecting facts. Theories are seen as secondary. Sometimes they are even derided. For example, proponents of intelligent design, the belief that life on Earth is the result of the actions of some all-powerful being, have insisted that schoolchildren be taught that the theory of evolution is “only a theory.” This is not meant as a compliment to Charles Darwin.

Psychometric models of intelligence assume that variations in cognitive performance across different situations can be summarized by individual differences in a fairly small number of basic cognitive dimensions, such as general reasoning, verbal facility, and visual-spatial reasoning. In general, the effort to find these dimensions has been quite successful. In the last analysis, though, psychometric models are summaries of test scores, and are relevant to cognition in general to the extent that test performance itself is relevant to individual differences in thinking, as revealed in daily life. In Chapter 1, I argued that test performance is relevant, as an imperfect sample of the cognitive skills we use on a daily basis. This chapter concentrates on theories of cognition within that imperfect sample. The following chapter discusses expansion of these theories.

What Are Psychometric Models?

Whenever we summarize scores on different pieces of performance we are implicitly assuming that there is some common thread underlying each piece. This is a very common assumption.

The two presidents were right. President Bush (or his speechwriter) described the brain accurately – a very complex bit of circuitry. President Johnson's description of Ford may not have been accurate, but he was right that bouncing the brain is not a good thing. A modern playwright put the matter in a slightly different way.

Every expression of intelligence is due to actions of the brain. What actions a brain will take in a given situation depends upon both the brain's structure and its history. This chapter will focus on individual differences in brain structures and processes related to intelligence. The topic is exciting. Findings are coming in so fast that it is hard to make sense of all of them. My computer literature search for papers on BRAIN AND INTELLIGENCE retrieved 5,648 citations. Neither I nor anyone else has read them all.