Editing G factor (psychometrics) (section)

===Mental energy or efficiency===

Charles Spearman reasoned that correlations between tests reflected the influence of a common causal factor, a general mental ability that enters into performance on all kinds of mental tasks. However, he thought that the best indicators of ''g'' were those tests that reflected what he called ''the eduction of relations and correlates'', which included abilities such as [[Deductive reasoning|deduction]], [[Inductive reasoning|induction]], problem solving, grasping relationships, inferring rules, and spotting differences and similarities. Spearman hypothesized that ''g'' was equivalent with "mental energy". However, this was more of a metaphorical explanation, and he remained agnostic about the physical basis of this energy, expecting that future research would uncover the exact physiological nature of ''g''.<ref>Jensen 1998, 18–19, 35–36, 38. The idea of a general, unitary mental ability was introduced to psychology by [[Herbert Spencer]] and [[Francis Galton]] in the latter half of the 19th century, but their work was largely speculative, with little empirical basis.</ref>

Following Spearman, [[Arthur Jensen]] maintained that all mental tasks tap into ''g'' to some degree. According to Jensen, the ''g'' factor represents a "distillate" of scores on different tests rather than a summation or an average of such scores, with factor analysis acting as the [[distillation]] procedure.<ref name="Jensen 2002"/> He argued that ''g'' cannot be described in terms of the item characteristics or information content of tests, pointing out that very dissimilar mental tasks may have nearly equal ''g'' loadings.  Wechsler similarly contended that ''g'' is not an ability at all but rather some general property of the brain. Jensen hypothesized that ''g'' corresponds to individual differences in the speed or efficiency of the neural processes associated with mental abilities.<ref>Jensen 1998, 91–92, 95</ref> He also suggested that given the associations between ''g'' and [[#Elementary cognitive tasks|elementary cognitive tasks]], it should be possible to construct a [[ratio scale]] test of ''g'' that uses [[time]] as the unit of measurement.<ref>Jensen 2000</ref>