October 25, 2005

Linda Gottfredson devastates Evolutionary Psychology

"How did humans evolve such remarkable intellectual powers?"

As I've mentioned before, a major problem with orthodox Evolutionary Psychology of the Tooby-Cosmides mode is that it is heavily concerned with brain functioning, yet, because (understandably) it wishes to avoid being crucified for political incorrectness, it tries hard to ignore the vast amount of knowledge about the subject that has been generated over the last 100 years by IQ psychometricians. So, it generates theories like "domain-specific massive modularity" of brain function that are divorced from the reality of the importance of the general factor (g) of intelligence that psychometricians have uncovered going back to Spearman in 1904.

Now, one of the leading scholars of the current social impact of IQ, Linda S. Gottfredson, has reversed the situation by developing a theory of the evolution of intelligence based on how brainpower works today in her upcoming book chapter "Innovation, Fatal Accidents, and the Evolution of General Intelligence."

“How did humans evolve such remarkable intellectual powers?” This is surely one of the most enduring and captivating questions in the life sciences, from paleoanthropology to neuroscience. Modern humans (Homo sapiens sapiens) far exceed all other species in their ability to learn, reason, and solve novel problems. We are, most strikingly, the only species whose members routinely use words and other abstract symbols to communicate with each other, record ideas in material form, and imagine alternative futures. Perhaps for these reasons, we are the only species ever to have developed complex technologies that allow us radically to transform the physical environments we inhabit.

Human intelligence is tied in some manner to the large increase in brain size going up the human evolutionary tree. When the encephalization quotient (EQ) is used to measure brain size relative to body size, modern humans are three times as encephalized (EQ = 6) as other primates (EQ = 2) and six times the average for all living mammals (EQ = 1, the reference group). This phylogenetic increase represents a disproportionate expansion of the brain’s prefrontal cortext, which matures last and is most essential for the highest cognitive functions, including weighing alternatives, planning, understanding the temporal order of events (and thus cause-and-effect relations), and making decisions. Moreover, encephalization of the human line proceeded rather quickly, in evolutionary terms, after the first hominids (Australopithicines, EQ = 3) split off from their common ancestor with chimpanzees (EQ = 2) about 5 million years ago. Encephalization was especially rapid during the last 500,000 to one million years, when relative brain size increased from under EQ = 4 for Homo erectus (arguably the first species of Homo) to about EQ = 6 for living humans (the only surviving subspecies of Homo sapiens).

Brains are metabolically expensive. In humans they account for 2% of body weight but consume 20% of metabolic energy. Hence, the rapid increase in relative brain size suggests that higher intelligence conferred a strong adaptive advantage. Attempts to identify the selection forces driving up intelligence in the human environment of evolutionary adaptedness (EEA) often look to the ecological, behavioral, and life history correlates of encephalization, either in the paleontological record or comparative studies of living species.

Evolutionary psychologists agree that increases in brain size are crucial in tracing the evolution of man’s extraordinary intelligence, but they say relatively little about what that intelligence actually is. They agree that humans have impressive reasoning abilities, which in turn confer valuable behavioral flexibility, but they conceptualize human intelligence in very different ways. The debate has focused on whether intelligence is “domain specific” (e.g., has “massive modularity”) or “domain general."

Proponents of domain specificity emphasize the morphological modularity of the human brain, likening it to a Swiss army knife, and argue that human intellectual prowess consists of a large collection of separate abilities that evolved independently to solve different specific adaptive problems, such as “cheater detection”. Humans, they argue, have not evolved any meaningful content- and context-free general reasoning or learning ability, but are smart because the human brain evolved myriad “fast and frugal heuristics”. The domain generalists, emphasizing the highly interconnected circuitry of the brain’s distinct parts, argue that human intelligence is best understood as a generalized capacity that facilitates reasoning and adaptive problem solving, especially in novel, changing, or otherwise complex situations. These theorists acknowledge the modular elements of the brain and mind, but consider them subject to the more general learning and reasoning mechanisms that they believe humans have evolved.

The specific-versus-general debate parallels the one-versus-many-intelligences debate that raged in the psychometric study of intelligence for almost a century. By the 1990s, however, cumulated evidence had persuaded most psychometricians that there exists only a single general intelligence factor (called g) which, in addition, constitutes the common backbone of all human cognitive abilities, broad or narrow. Proponents of domain generality cite this conclusion to support their thesis for a domain-general intelligence, but their currently favored explanations for its evolution actually presuppose a social intelligence in competing with other humans for resources. This does not correspond well to psychometric g (or to any other measured trait).

This implicit shift in the explanadum has followed the field’s inability to better substantiate longstanding hypotheses about how mastery of the physical environment (e.g., food, predators, harsh climates), as distinct from the social environment, ratcheted up human intelligence.

This chapter aims to show not only that our species’ distinctive intelligence is domain general at the phenotypic, genetic, and functional levels, but also how a general intelligence could have evolved. Drawing evidence from sister disciplines not often consulted by evolutionary psychologists, I first describe how general mental ability, g, represents a suite of generic critical thinking skills that provides individuals with pervasive practical advantages in coping with many life challenges, especially when tasks are more complex.

A close look at the task requirements of jobs and daily self-maintenance in modern life reveals which task attributes contribute most to complexity, and hence to the functional advantages of higher g, in many domains of life. As will be illustrated, the cognitive demands of even the most mundane daily tasks are sufficient to put less intelligent persons at a higher relative risk for many unfavorable life outcomes, including premature death.

One particularly large class of deaths -- fatal accidents -- will be used to illustrate how individual differences in g might contribute to differential mortality as people go about their daily lives. The prevalence, etiology, and demographic patterning of accidental deaths in both modern and hunter-gatherer societies provide clues to how these could have winnowed away a group’s less intelligent members throughout human evolution: fatal accidents (unintentional injuries) kill a disproportionate number of reproductive-age males, their accidents are generally associated with provisioning activities, and preventing these is a cognitively demanding process. Accidents have a high chance component, are diverse in type, and rarely kill, which dulls our appreciation of them. But these attributes are also precisely what make them a potentially powerful force for evolving a general-purpose problem solving mechanism rather than hazard-specific hazard detection modules. Fatal accidents would supplement, rather than supplant, provisioning and other ecological challenges by which natural selection could have favored individuals who learn quickly and reason well. They would likewise supplement sexual selection, in which mate preferences drive the evolution of preferred traits. [Much more]

My published articles are archived at iSteve.com -- Steve Sailer

No comments: