Galton’s “ridiculous” intelligence tests

As some scholars tell the story, this so-called genius Francis Galton had some pretty stupid ideas about how to measure intelligence. Instead of measuring useful things like reasoning, knowledge, and creativity, Galton measured totally irrelevant things like visual acuity, grip strength, and reaction time! What an idiot! Anyone with the least bit of common sense would realize that the senses are not the same as intelligence. Galton, as the story goes, was blinded by his philosophical allegiance to empiricism and associationism, which emphasized the primacy of the senses in explaining human knowledge and reasoning. The world would have to wait for Alfred Binet to set things right.

As it turns out, the reason that Galton measured basic sensorimotor abilities like visual acuity, grip strength, and reaction time was that he was interested in… basic sensorimotor abilities like visual acuity, grip strength, and reaction time! He was interested in them for roughly the same reasons that medical doctors, developmental psychologists, and neuropsychologists are interested in them. For example, Galton published tables showing the ages at which sensory acuity and grip strength are at their peaks and the ages at which they tend to decline. This is useful, basic research.

In reading Galton’s publications of such findings, one will find no grand pronouncements about the nature of intelligence. Galton never claimed that basic sensorimotor abilities constituted the whole of intelligence. Nevertheless it is true that, from time to time, he claimed to have found evidence that the most intellectually able tend to have greater sensitivity and sensory discrimination (Galton, 1883, p. 20), though this evidence was never formally presented. Galton had no intention of measuring intelligence directly for diagnostic purposes. He was much more interested in discovering the precursors of intelligence. Galton, instead of measuring intelligence directly, preferred to infer intelligence from measures of eminence.

Many summaries of Galton’s work report that Galton’s hypotheses about sensory acuity and intelligence failed. Within the last few decades, Galton’s hypothesis has been revived. The correlation between sensory acuity and higher cognitive abilities is indeed positive as Galton predicted, but the effect is much smaller than he expected it to be (Jensen, 2006). It is unlikely that the relationship between sensory acuity and reasoning is direct (i.e., better input→better output). It is likely that the overall health of the body manifests itself (inconsistently and probabilistically) in both the sense organs and in the brain.

What can we learn from Galton’s “ridiculous” tests? We need to understand that there is an important difference between applied research that uses intelligence tests to forecast important life outcomes and more basic research that aims to explain the foundations of intellectual ability. Galton’s research, however crude, was of the latter type. Failing to understand this distinction leads us to scoff at perfectly reasonable research. For example, it is common to come across criticisms of research that aims to understand the relation between brain size and intelligence. The correlation between brain size and intelligence appears to be on the order of ρ = 0.3 to 0.4 (Lange, Froimowitz, Bigler, Lainhart, & Brain Development Cooperative Group, 2010). Reports of such findings bring to mind frightening visions of arbitrary bureaucrats in a future dystopia granting and denying privileges to people based on the size of their heads. No sane scholar is hoping this will happen. No scholar believes that head size per se causes higher or lower intelligence. Rather, head size is an imperfect indicator of one or more developmental processes that do have a more direct influence on intelligence. Studying head size is merely a stepping stone (it is hoped) to the discovery of such processes.

Roughly the same interpretive error occurs when it is announced that IQ tests are soon to be replaced with working memory tests because researchers have found that individual differences in working memory capacity explain much of IQ tests’ predictive validity.

Such announcements miss the point. If we care about individual differences in knowledge, reasoning, and creativity, we should measure these abilities directly. The fact that working memory plays a vital role in such abilities is useful to know, but such tests only supplement our understanding of intellectual processes in individuals. They help explain what might have gone wrong (or especially right) in an individual’s cognitive development. The correlations of working memory capacity and higher-order cognitive abilities are not nearly high enough that traditional cognitive ability tests are redundant. Research on working memory capacity (or processing speed or elementary cognitive tasks) is unlikely to ever lead to a replacement of direct measures of knowledge, reasoning, and creativity. Rather, such research helps us understand the origins of individual differences in these higher-order capacities.