TRENDS in Cognitive Sciences Vol.6 No.4 April 2002
http://tics.trends.com 1364-6613/02/$ – see front matter © 2002 Elsevier Science Ltd. All rights reserved. PII: S1364-6613(00)01853-2
169 Review
Robert Plomin*
F.M.Spinath
Social, Genetic and
Developmental Psychiatry
Research Centre, Institute
of Psychiatry, King’s
College London, Denmark
Hill, London, UK SE5 8AF.
*e-mail:
r.plomin@iop.kcl.ac.uk
Inter-individual differences in performance on
diverse psychometric tests of cognitive abilities –
such as verbal, spatial, memory and processing speed
– intercorrelate about 0.30 on average and a general
factor (an unrotated first principal component)
accounts for about 40% of the total variance of these
psychometric tests [1]. g is not the whole story of
cognitive abilities– group factors representing specific
abilities also represent an important level of analysis
– but trying to tell the story of cognitive abilities
without g loses the plot entirely. Arecent article in
TICS made the case for the existence and importance
of g, which is sometimes called ‘intelligence’ [2].
A companion paper described the cognitive and
psychophysical correlates of psychometric g [3].
In this review, we consider g and its cognitive
correlates from a genetic perspective. By ‘genetic’,
we mean ‘heritable’ (inherited DNA differences among
individuals) rather than ‘innate’ (evolutionarily
constrained differences among species).
A stronger case has been made for substantial
genetic influence on g than for any other human
characteristic. Dozens of studies including more than
8 000 parent-offspring pairs, 25 000 pairs of siblings,
10 000 twin pairs, and hundreds of adoptive families
all converge towards the conclusion that genetic
factors contribute substantially to g [4]. Estimates
of the effect size, called heritability (see Ref. 4 for
explanation), vary from 40 to 80% but estimates
based on the entire body of data are about 50%,
indicating that genetic variation accounts for about
half of the variance in g. Breaking down the research
by age, heritability can be seen to increase almost
linearly from infancy (about 20%) to childhood
(about 40%) to adulthood (about 60%) [5]. Genetic
research has moved beyond merely estimating
heritability. One new direction that is the focus of the
present review is multivariate genetic analysis,
which provides important new insights into the
structure of cognitive processes linked to g.
Multivariate genetic analysis of psychometric tests
A technique called multivariate genetic analysis
examines the extent to which genetic and
environmental factors mediate the phenotypic
covariance between variables. It also yields a statistic
called the ‘genetic correlation’, which is the extent to
which genetic effects on one trait correlate with
genetic effects on another trait independent of the
heritability of the two traits. That is, although all
psychometrically assessed cognitive abilities are
moderately heritable, the genetic correlations
between cognitive abilities could be anywhere from
0.0, indicating complete independence, to 1.0,
indicating that the same genes influence different
cognitive abilities. Multivariate genetic analysis is
described in Box 1.
Multivariate genetic analyses of psychometric
tests consistently find that genetic correlations are
very high – close to 1.0 – in adolescence and adulthood
[6,7]. That is, if a gene were identified that is
associated with a particular psychometrically
assessed cognitive ability, the same gene would be
expected to be associated with other cognitive
abilities as well. In other words, although
psychometric g accounts for about 40% of the total
phenotypic variance of diverse cognitive tests, genetic
g accounts for nearly all of the genetic variance.
Models of cognitive mechanisms and genetic g
Can genetic g as inferred from psychometric tests be
explained in terms of more basic cognitive
mechanisms? Although it might seem obvious to
cognitive scientists that the answer is yes, there is a
problem. In cognitive science, it is generally assumed
that cognitive processes are independent or modular.
Although Fodor’s original concept of modules as
innate and invariant information-processing units [8]
has been watered down to the notion of domain
specificity, it remains a pervasive view in cognitive
science from older lesion studies and newer
neuroimaging research that cognition consists of
many discrete and independent processes [9–11].
How can modular cognitive processes explain such a
molar outcome as genetic g?
In cognitive science, modularity is considered from
a normative perspective, focusing on species-typical
processes as assessed for example by neuroimaging or
by disrupting normal processes with drugs, lesions or
knockout genes. The provenance of g, however, is
individual differences – why people differ in their
Two recent articles in this journal made the case for the existence and
importance of g and reviewed research on cognitive and psychophysical
correlates of psychometric g.This review considers g from a genetic
perspective.Multivariate genetic research indicates that g accounts for nearly
all of the genetic variance of diverse psychometric cognitive tests (genetic g).
Recent research suggests not only that elementary cognitive tasks are
genetically linked to psychometric g but also that genetic g pervades these
tasks. Contrary to the assumption of modularity that dominates cognitive
science, genetic g exists in the mind as well as in psychometric tests.
Genetics and general cognitive
ability ( g )
Robert Plomin and Frank M. Spinath