The 'g' Factor: What Charles Spearman Discovered
In 1904, the British psychologist Charles Spearman published a paper that would shape the next century of intelligence research. He noticed that performance on disparate cognitive tests — sensory discrimination, mathematics, vocabulary, mental rotation — was positively correlated. People who did well on one test tended to do well on others.
In 1904, the British psychologist Charles Spearman published a paper that would shape the next century of intelligence research. He noticed that performance on disparate cognitive tests — sensory discrimination, mathematics, vocabulary, mental rotation — was positively correlated. People who did well on one test tended to do well on others. Spearman called this pattern of positive correlations the 'positive manifold' and proposed that it reflected a single underlying general factor he labeled g.
The methodological tool Spearman developed to study g was factor analysis, a statistical technique that identifies latent variables (factors) that account for shared variance across observed measures. When applied to a large battery of cognitive tests, factor analysis consistently extracts a single dominant factor that accounts for 30 to 50% of the variance in scores. The remaining variance is captured by group factors (broad abilities like Gf and Gc) and by test-specific variance.
g is not a thing in the brain. It is a statistical regularity in cognitive data. The most common modern interpretations of g are functional: g may reflect the efficiency of fronto-parietal communication, or the speed of basic neural processing, or the integrity of white-matter tracts that support coordinated cognition. Different researchers favor different mechanisms, and the empirical evidence allows multiple compatible accounts.
What is uncontroversial is that g is the single best predictor of life outcomes that psychometrics has ever identified. Across thousands of studies, g correlates with academic achievement (r ≈ 0.5), job performance in complex roles (r ≈ 0.4), income (r ≈ 0.3), and health behaviors (r ≈ 0.2). These correlations are statistically robust but not enormous; g is one important predictor among many, not a deterministic ceiling.
Critics of g — most prominently Stephen Jay Gould in 'The Mismeasure of Man' (1981) and Howard Gardner in 'Frames of Mind' (1983) — have argued that g is a statistical artifact of the test selection process, and that human cognitive ability is better described as a set of independent intelligences. The empirical evidence for the positive manifold is strong enough that most contemporary psychometricians accept g as a real and useful construct, while acknowledging that single-number summaries lose information about subtest profiles.