g factor
In plain wordsThe common factor behind the finding that people who do well on one kind of mental test tend to do well on others. Researchers also look for a similar general factor across AI models.
The general factor of cognitive ability extracted by factor analysis from the positive correlations among scores on diverse cognitive tests — vocabulary, arithmetic, spatial reasoning, memory — a pattern called the positive manifold. In hierarchical models such as the Cattell–Horn–Carroll theory, g sits above broad abilities including fluid intelligence (reasoning on novel problems) and crystallized intelligence (acquired knowledge). It accounts for a large share of individual differences in test performance and predicts educational and occupational outcomes, but what it represents — a single underlying capacity, the product of many interacting processes, or merely a statistical summary — is debated. In AI evaluation, factor analyses of language modelslarge language model, n.: A computer program trained on enormous amounts of text to predict the words that come next. At vast scale, that simple skill lets it chat, answer questions, translate, summarize and write code.' benchmarkbenchmark, n.: A fixed, shared test — a set of questions or tasks with an agreed way of scoring — used to compare AI systems with one another and to track progress over time. scores ask whether a similar general factor runs across models; findings vary with the models and tasks sampled, some showing a dominant general factor and others several distinct but correlated abilities.
Origin
Introduced by the psychologist Charles Spearman in '"General Intelligence," Objectively Determined and Measured' (American Journal of Psychology, 1904). Raymond Cattell distinguished fluid and crystallized intelligence from the 1940s, and John Carroll's 1993 reanalysis of hundreds of datasets grounded the hierarchical model.
Usage note
Not to be confused with artificial general intelligenceartificial general intelligence, n.: AI that could learn and do almost any mental task a capable adult can, rather than excelling at one narrow job. Experts disagree on the exact bar and on when, or whether, it will arrive.: a family of models can share a strong general factor across benchmarks without any of them being generally intelligent in the AGI sense.
Related
- intelligenceintelligence, n.: The ability to learn, understand, reason and solve problems — especially new ones — in order to reach goals. Experts disagree about how to define it precisely, for people and even more for machines.
- artificial general intelligenceartificial general intelligence, n.: AI that could learn and do almost any mental task a capable adult can, rather than excelling at one narrow job. Experts disagree on the exact bar and on when, or whether, it will arrive.
- benchmarkbenchmark, n.: A fixed, shared test — a set of questions or tasks with an agreed way of scoring — used to compare AI systems with one another and to track progress over time.
- jagged intelligencejagged intelligence, n.: The uneven pattern of AI abilities: today's systems can solve very hard problems yet fail at some a child could manage, so their skills form a spiky profile rather than a smooth one.
- universal intelligenceuniversal intelligence, n.: A mathematical definition of intelligence as an agent's average ability to achieve goals across every possible environment, with simpler environments counting more.
Sources
- Spearman, C. (1904). 'General Intelligence,' Objectively Determined and Measured. The American Journal of Psychology 15(2), 201–292.
- Carroll, J. B. (1993). Human Cognitive Abilities: A Survey of Factor-Analytic Studies. Cambridge University Press.
Mentioned in
intelligenceintelligence, n.: The ability to learn, understand, reason and solve problems — especially new ones — in order to reach goals. Experts disagree about how to define it precisely, for people and even more for machines.