Deary I J
Department of Psychology, University of Edinburgh, 7 George Square, Edinburgh EH8 9JZ, UK.
Novartis Found Symp. 2000;233:58-72; discussion 73-8. doi: 10.1002/0470870850.ch4.
Psychometric intelligence attracts a converging consensus about its phenotypic structure. Mental ability test scores have proven predictive validity. However, although individual differences in mental abilities can be measured, they are not understood. A long-standing aim of the 'London School' of British psychologists, since Galton and Spearman, is to understand the origins of psychometric intelligence differences in terms of individual differences in brain processes. The history of this research is described, as is the rise in interest since the 1970s. The first problem, met since antiquity, is to discover the relevant levels of brain function. Thus, aspects of brain function that 'explain' psychometric intelligence differences are sought at psychometric, cognitive, psychophysical, physiological, neurochemical and genetic levels. The growing points and dead-ends within each of these levels are identified. Special attention is given to research that crosses levels of description of brain function. Two types of multi-level brain function research are discussed, 'correlational' and 'circumstantial/experimental,' and examples of each are described. Illustrating both approaches, there is a detailed account of research on inspection time that discusses how psychometric intelligence-brain process correlations at one level (psychophysical) may be expanded using event-related potentials, psychopharmacology and functional magnetic resonance imaging.
