Intelligence 38 (2010) 293–303

Contents lists available at ScienceDirect

Intelligence

Brain networks for working memory and factors of intelligence assessed in
males and females with fMRI and DTI
C.Y. Tang a,b,⁎, E.L. Eaves a, J.C. Ng a, D.M. Carpenter a, X. Mai a, D.H. Schroeder c, C.A. Condon c,
R. Colom d, R.J. Haier e
a

Department of Radiology, Mt. Sinai School of Medicine, NY, NY, USA
Department of Psychiatry, Mt. Sinai School of Medicine, NY, NY, USA
c
Johnson O'Connor Research Foundation, Chicago, IL, USA
d
Universidad Autonoma de Madrid, Spain
e
University of California, Irvine, CA, USA (Emeritus)
b

a r t i c l e

i n f o

Article history:
Received 24 September 2009
Received in revised form 21 March 2010
Accepted 22 March 2010
Available online 28 April 2010
Keywords:
Intelligence
Working memory
fMRI
DTI
Sex differences

a b s t r a c t
Neuro-imaging studies of intelligence implicate the importance of a parietal–frontal network. One
unresolved issue is whether this network underlies a general factor of intelligence (g) or other
specific cognitive factors. A second unresolved issue is whether males and females use different
parts of this network. Here we obtained intelligence factors (general, speed of reasoning, spatial,
memory, and numerical) from a large set of tests completed by 6929 young adults, 40 of whom (21
males, 19 females) also completed DTI and fMRI during a working memory n-back task. Within
brain areas activated during this task, correlations were computed between percent activation and
scores on the intelligence factors. The main findings were: (1) individual differences in activation
during the n-back task were correlated to the general intelligence factor (g), as well as to distilled
estimates (removing g) of speed of reasoning, numerical ability, and spatial ability, but not to
memory, (2) the correlations were mainly bilateral for females and unilateral for males, and (3)
differences in the integrity of the axonal connections were also related to the functional findings
showing that integrity of interhemispheric connections was positively correlated to some
intelligence factors in females but negatively correlated in males. This study illustrates the
potential for identifying aspects of the neural basis of intelligence using a combination of structural
and functional imaging.
© 2010 Elsevier Inc. All rights reserved.

1. Introduction
Neuro-imaging studies of the underlying structural and
functional anatomy of intelligence implicate areas throughout the brain, irrespective of the intelligence tests used. Jung
and Haier (2007) characterized these findings as mostly, but
not exclusively, in frontal and parietal areas. They proposed a
model of how these areas may form overlapping networks
underlying individual differences in intelligence: the ParietoFrontal Integration Theory—P-FIT. The P-FIT areas represent

⁎ Corresponding author. Department of Radiology, Mt. Sinai School of
Medicine, NY, NY, USA.
E-mail address: Cheuk.Tang@mssm.edu (C.Y. Tang).
0160-2896/$ – see front matter © 2010 Elsevier Inc. All rights reserved.
doi:10.1016/j.intell.2010.03.003

stages of information processing from posterior sensory
perception to abstraction in parietal areas to anterior
hypothesis testing and decision-making. Integration of
information among the areas is key. Based on functional
imaging studies that found inverse correlations between
regional brain activation and performance on intelligence
tests (Haier et al., 1988; Neubauer & Fink, 2009) the P-FIT
includes the hypothesis that efficient flow of information
around these networks is related to intelligence. Similar
networks have been identified for performance on fundamental cognitive tasks including aspects of attention and
memory (Cabeza & Nyberg, 2000; Naghavi & Nyberg, 2007;
Wager, Jonides, & Reading, 2004; Wager & Smith, 2003),
although inverse correlations with such tasks are not
reported, possibly because fundamental cognitive tasks used

294

C.Y. Tang et al. / Intelligence 38 (2010) 293–303

in imaging studies usually are chosen to minimize individual
differences in performance.
Since intelligence tests tap more than one cognitive
domain, it remains to be seen if the P-FIT or other models of
brain networks represent a general factor of intelligence (g),
common among all tests, or more specific group factors like
memory or spatial ability. So far, only two structural imaging
studies have extracted a g-factor score from a battery of tests
and then correlated these and other more specific group factor
scores (with g removed) to gray matter (Colom et al., 2009;
Haier et al., 2009). Both studies showed similar results for a
spatial factor, but not for a g-factor, suggesting that there may
not be a single neural basis for g (Haier et al., 2009). Johnson
et al. (2008) also, for example, extracted other cognitive factors
with g removed in a small sample and showed some gray
matter correlates different than those associated with g.
Functional imaging studies of networks related to intelligence have not yet used g-factor scores as dependent variables,
instead relying on single tests like the Raven Progressive
Matrices Test (Gray, Chabris, & Braver, 2003; Haier et al., 1988;
Lee et al., 2006; Prabhakaran, Smith, Desmond, Glover, &
Gabrieli, 1997). The interpretation of functional imaging data,
moreover, is constrained by the task performed during the
imaging, unlike structural imaging (Toga & Thompson, 2005),
so task selection is a key element of research design. Two
functional imaging studies with fMRI (Gray et al., 2003; Waiter
et al., 2009) have used the working memory n-back task
because working memory is highly related to intelligence
(Colom, Abad, Quiroga, Shih, & Flores-Mendoza, 2008; Colom
et al., 2005; Colom, Rebollo, Palacios, Juan-Espinosa, &
Kyllonen, 2004; Engle, 2002; Grabner, Fink, Stipacek, Neuper,
& Neubauer, 2004; Kane, Hambrick, & Conway, 2005;
Oberauer, Schulze, Wilhelm, & Suss, 2005). These two fMRI
studies of intelligence using the n-back test report that
activation in frontal and parietal areas is correlated to single
intelligence test scores. Inverse correlations are not reported.
All the subjects were males.
In addition to issues about the use of single test scores
rather than factor scores, another unresolved issue concerns
sex differences. A number of imaging studies show male/
female differences related to intelligence (Haier, Jung, Yeo,
Head, & Alkire, 2005; Luders et al., 2008; Schmithorst &
Holland, 2007; Sowell et al., 2007) and other cognitive
abilities (Haier & Benbow, 1995; Jausovec & Jausovec,
2008). Findings regarding brain efficiency also show strong
sex differences (Neubauer, Fink, & Schrausser, 2002) that may
be related to any number of brain differences between males
and females (e.g. (Luders et al., 2004; Rabinowicz, Dean,
Petetot, & de Courten-Myers, 1999). Due to cost, most
imaging studies focus on one sex (usually males) or partial
out sex when male and female samples are combined to
increase statistical power at the cost of obscuring any actual
sex differences. In our view, separate analyses for males and
females are required to explore any differences that may be
unique to the study sample or to a more general finding.
Here we extend the two previous n-back studies to
determine fMRI correlates of intelligence using factors derived
from a battery of tests rather than a single test score; these
factors are independent of the g-factor. Based on the structural
studies of (Colom et al., 2009) and (Haier et al., 2009), we
hypothesize that individual factors will have functional

correlates different from the g-factor. Further, we present
analyses separately for males and females to test whether the
P-FIT areas differ in activation during the non-verbal n-back
memory task and whether inverse correlations consistent with
brain efficiency may be stronger in males, as suggested by
Neubauer and Fink (2009). The P-FIT also noted the potential
importance of individual differences in white matter connectivity, especially the arcuate fasciculus; and there is some
suggestion that white matter may be more important for intelligence in women than in men (Haier et al., 2005). Therefore, we added a second imaging method, Diffusion Tensor
Imaging (DTI), to characterize white matter tracts among any
areas identified with functional imaging as related to n-back
performance. DTI provides information on the integrity of the
axonal connections in the brain (Basser, 1997). By combining
DTI with fMRI it is possible to provide a more comprehensive
picture of structural and functional integration during cognitive performance (Fjell et al., 2008). For example Schmithorst
and Holland (2007) found differences in activated brain areas
between boys and girls during a verbal task as well as
differences in white matter pathways (Schmithorst, Holland,
& Dardzinski, 2008). Older girls showed greater inter-hemispheric connectivity. Yu et al (2008) computed correlations
between the integrity of several tracts (corpus callosum,
cingulum, uncinate fasciculus, optic radiation, and corticospinal tract) and intelligence. The 79 participants (men and
women; mean age 23.8) were divided in two groups:
average and high intelligence. White matter integrity was
assessed by fractional anisotropy (FA). The results, controlling for age and sex, showed that high intelligence
participants display more white matter integrity than
average intelligence participants only in the right uncinate
fasciculus. The authors concluded that the right uncinate
fasciculus is an important neural basis of intelligence
differences. There were no separate analyses by sex, but a
sample of 15 participants with mental retardation was also
studied. These participants were compared with the 79
healthy controls and they showed extensive damage in the
integrity of the brain white matter tracts: corpus callosum,
uncinate fasciculus, optic radiation, and corticospinal tract.
A recent paper (Chiang et al., 2009) reported the first
study combining a genetically informative design and a DTI
approach for analyzing the relationships between the white
matter integrity and human intelligence. Intelligence was
assessed by the Multidimensional Aptitude Battery, which
provides measures of general intelligence, verbal (information, vocabulary, and arithmetic), and non-verbal intelligence
(spatial and object assembly). The sample included 23 pairs
of identical twins and 23 pairs of fraternal twins (males and
females but all pairs were same sex; mean age 25 years).
White matter integrity, quantified using fractional anisotropy
(FA), was used to fit structural equation models (SEM) at each
point in the brain. They then generated three-dimensional
maps of heritability. White matter integrity was found to be
under strong genetic control in bilateral frontal, bilateral
parietal, and left occipital lobes. FA measures were correlated
with the estimate of general intelligence and with non-verbal
intelligence in the cingulum, optic radiations, superior frontooccipital fasciculus, internal capsule, callosal isthmus, and
the corona radiate. Further, common genetic factors mediated
the correlation between intelligence and white matter

295

C.Y. Tang et al. / Intelligence 38 (2010) 293–303

integrity. This latter finding suggested a common physiological mechanism and common genetic determination.
DTI studies of intelligence are relatively new and there are
not yet data in adult samples, so our analyses are exploratory.
Since integrity of white matter could relate to efficient flow of
information, we generally expect positive correlations with
intelligence factors.
2. Method
2.1. Participants
The sample was the same as reported in a previous study
focused on structural MRI assessments of gray matter only
(Haier et al., 2009). During 2002–2003, 6889 individuals
sought consultation from the Johnson O'Connor Research
Foundation (JOCRF), a non-profit organization dedicated to
using psychometric assessments for vocational guidance.
Each completed the same battery of eight cognitive tests
listed below in one of 11 testing centers in major cities
throughout the United States. The mean age was 25.4 years
(SD = 10.6); there were 3722 males (mean age = 25.0,
SD = 10.2, and there were 3207 females (mean age = 25.9,
SD = 11.0). In addition, participants who completed the same
test battery in 2006 and 2007 in the New York City center
were invited to return for structural and functional MRI, and
Diffusion Tensor Imaging (DTI) at Mt. Sinai Medical Center.
All who volunteered were screened using our own structured
interview to exclude anyone with a major medical or
psychiatric illness including a history of head injury and
substance abuse. After informed consent was obtained, the
final 40 participants in this study completing imaging
included 21 males (mean age = 26.62, SD = 4.60) and 19
females (mean age = 26.63, SD = 4.90). JOCRF clients are
generally college-bound or college-educated, and they tend
to be spread across the ability range of that population. This is
a somewhat narrower range than for the general population,
and so any correlations found are likely to be somewhat
attenuated relative to the larger population. In a recent study
(Condon & Schroeder, 2005), it was found that for undergraduate education, 39.0% of JOCRF examinees attend “most”
or “very difficult” schools, per Peterson's guide to four-year
colleges, 41.5% “moderately difficult” schools, and 16.2%
“minimally difficult” or two-year schools, while 3.3% do not
attend college.
2.2. Intelligence testing and factors
The eight tests in the JOCRF battery were: Inductive Speed
(IS), Analytical Reasoning (AR), Number Series (NS), Number
Facility (NF), Wiggly Block (WB), Paper Folding (PF), VerbalAssociative Memory (VAM), and Number Memory (NM). A
description of these tests, including the constructs they
measure and their reliabilities can be found in Haier et al.
(Haier, 2009) along with confirmation that this battery loads
on four factors—Speed of Reasoning (IS and AR), Numerical
(NS and NF), Spatial (WB and PF), and Memory (VAM and
NM) in addition to a g-factor. The means and standard
deviation for all tests are shown in Table 1. The correlations
among the tests and factors are shown in Table 2. In terms of
gender differences, the eight JOCRF tests show mean-level

differences that are more-or-less in line with those reported
in the literature (Halpern, 2000).The largest difference in
favor of males in the JOCRF population is for Wiggly Block (.51
standard deviation units), a test of spatial visualization, while
the largest difference in favor of females is for VerbalAssociative Memory (.45 SD units). In terms of variability,
although there have been reports of greater variability among
males (Halpern, 2000), for the JOCRF population on these
tests, the SDs for males and females are very similar, with the
exception of Paper Folding, for which the male SD is 18.4%
larger than the female SD. Finally, in terms of correlations
among the tests, they are quite similar for the two genders.
An SEM analysis showed no significant differences in the
factor structures for males and females for these measures
(Condon & Schroeder, 2003).
We computed standardized scores (z-scores) for the eight
tests and computed average z-scores for each factor. Test
scores were partialled for sex and age in order to eliminate
nuisance variance. The general intelligence g-score for each
subject was the average of their z-scores on the eight tests
(see (Haier et al., 2009) for additional details) with an alpha
reliability of .80. The g and residualized (that is, g-partialled)
z-scores for each factor for these 40 subjects were used to
determine the correlations to fMRI activations and DTI, as
described below. Note that residualized scores for speed of
reasoning, numerical, spatial, and memory represent participants' performance not shared with the general factor score
(g). This distinction is often neglected in neuroimaging
studies of intelligence and cognitive abilities (Colom, 2007).

2.3. fMRI task
Based on a single letter n-back paradigm, a multi-back
paradigm was developed using E-Prime (PST Inc., Pittsburgh,
PA). Four levels of memory load were presented using Ns
varying between 0, 1 , 2 and 3. Each n was presented 3 times
with different randomizations. All letter sequences were

Table 1
Descriptive statistics for samples.
Test

Large sample

Small sample

M

SD

M

SD

Speed of reasoning
Inductive speed
Analytical reasoning

143.14
53.80

23.10
14.49

140.38
60.33

24.51
14.17

Numerical
Number series
Number facility

23.58
94.36

4.55
17.06

24.48
100.43

4.98
19.97

Spatial
Wiggly block
Paper folding

261.03
22.38

98.55
14.08

320.35
28.73

88.60
14.68

Memory
Verbal-assoc. memory
Number memory

20.41
80.55

9.64
28.89

24.40
91.55

10.14
28.17

Note. For the large sample, Ns ranged from 6778 to 6889; for the small sample,
N was 40 for all tests. The reported values are for raw scores unpartialled for sex
or age. The small sample scored significantly higher than the large sample on all
the tests except for Inductive speed and Number series (p b .05).

296

C.Y. Tang et al. / Intelligence 38 (2010) 293–303

Table 2
Correlations among the tests and factors for the combined sample.
Measure

1

2

3

4

5

6

7

8

9

10

11

12

13

1. Inductive speed
2. Analytical reasoning
3. Number series
4. Number facility
5. Wiggly block
6. Paper Folding
7. Verbal-associative memory
8. Number memory
9. Spatial factor
10. Memory factor
11. Numerical factor
12. Speed of reas. factor
13. G factor

–

0.42
–

0.19
0.40
–

0.30
0.44
0.50
–

0.30
0.41
0.34
0.29
–

0.23
0.43
0.44
0.30
0.59
–

0.16
0.27
0.35
0.28
0.17
0.26
–

0.14
0.27
0.39
0.36
0.24
0.31
0.48
–

− 0.20
− 0.12
− 0.18
− 0.40
0.38
0.56
− 0.36
− 0.35
–

− 0.21
− 0.28
− 0.09
− 0.16
− 0.31
− 0.26
0.49
0.55
− 0.45
–

− 0.09
0.16
0.49
0.54
− 0.20
− 0.21
0.05
0.07
− 0.53
− 0.27
–

0.73
0.35
− 0.29
0.08
0.10
− 0.23
− 0.17
− 0.27
− 0.22
− 0.25
− 0.22
–

0.47
0.67
0.69
0.62
0.66
0.77
0.53
0.61
0.00
0.00
0.00
0.00
–

Note. Ns range from 6712 to 6929. All correlations significant at alpha .01 except for the correlations between g and the group factors, which were zero.

randomized and counterbalanced. Each trial was preceded by
a 2 s instruction screen indicating which n-back was to
follow. Each trial lasted for 30 s. In between the trials was a
20 s rest period where the subjects were presented with a
fixation screen. All participants received instructions on the
task before the imaging session. Each BOLD (Blood Oxygenation Level Dependent) scan consisted of twelve trials of
different n-back stimuli.
2.4. Imaging
All imaging was performed on a 3T Allegra MRI scanner
(Siemens, Ehrlangen, Germany). For fMRI, EPI (Echo Planar
Imaging) BOLD scans were acquired using a GE-EPI sequence
with the parameters: TR = 2 s, TE = 27 ms, FOV = 21 cm,
2.5 mm thick, skip=0.5 mm, Matrix size=64×64, 34 slices,
246 measurements with a total scan time of about 8 min. DTI
used a pulsed-gradient spin-echo sequence with EPI-acquisition
(TR=4100 ms, TE=80 ms, FOV=21 cm, matrix=128×128,
34 slices, thickness=3 mm skip 1 mm, b-factor=1250 s/mm2,
12 gradient directions, 5 averages). For incidental pathology
screening we also acquireT2-weighted anatomical scans of the
whole brain using a turbo spin-echo (TSE) pulse sequence (34
axial slices, repetition time [TR]=5380 ms, echo time [TE]=
99 ms, flip angle=170°, field of view [FOV]=210 mm, matrix=512×336, voxel size=0.41×0.41×4 mm). For coregistration and normalization purposes a high resolution T1weighted structural image with good gray/white matter contrast
was acquired using an 3D-MP-RAGE (Magnetization Prepared
Rapid Gradient Echo) sequence with the following protocol:
matrix size = 256 × 256 × 208, FOV = 21CM, TR = 2500 ms,
TE=4.38 ms, TI=1100 ms and an 8° flip angle FLASH acquisition giving a total imaging time of about 10 min.
3. Analyses
3.1. fMRI
Prior to analyses, the anatomical scans were read by staff
radiologists to screen for any incidental clinical findings; none
were found. BOLD data were processed using SPM5 (Wellcome
Department of Cognitive Neurology, Institute of Neurology,
University College London, London, UK). Functional data was
slice-time corrected by interpolation to the middle slice prior

to motion correction. Functional images were co-registered to
each subject's anatomical scan. Co-registered anatomical
images were then segmented to produce the parameters
used for normalization into a standard anatomical brain
reference template developed by the Montreal Neurological
Institute (MNI). Images were spatially smoothed using a 6 mm
isotropic Gaussian smoothing kernel. Individual contrast
images were produced in the context of the general linear
model using a boxcar function [0-back versus 1,2,3-back]
convolved with a canonical hemodynamic response function.
Next, a one-sample t-test was performed using each subject's
contrast images to determine areas of increased activity during
the active (1,2,3-back) state compared to baseline (0-back)
(Fig. 1). Significant clusters were identified (p b .001, uncorrected; separate analyses of the 1 vs. 0, 2 vs. 0, and 3 vs. 0
conditions did not yield appreciably different results; there
were no sex differences in n-back performance) and coordinates for spherical Regions-of-Interest (ROI) with 3 mm radii
were obtained from the local maxima of these clusters. Percent
activation values for these ROIs were extracted from individual
subjects' normalized scans and transferred to Statistica V7
(Statsoft Inc., Tulsa, OK) for further analysis.
3.2. DTI
Raw DTI data were transferred to an off-line workstation
for post-processing. The Diffusion data set was eddy-current
and motion corrected using an adaptation of the Camino/SPM
package (Cook, 2006). In-house software was developed in
Matlab v2007a (The Mathworks Inc., Natick, MA) for further
processing of the DTI and tractography, a method to extract
and visualize fiber tract bundles based on the geometric
information contained in adjacent voxels obtained by DTI
scans. Fractional Anisotropy (FA) and directionally colorcoded FA maps were also computed. Color coded FA maps
improve inter-rater reliabilities for ROI-based analysis of FA
data by segmenting the different adjacent fiber bundles into
different colors based on their orientation. For fiber tracking a
commonly used “multiple region brute-force” fiber tracking
method was used (Huang, Zhang, van Zijl, & Mori, 2004).
First, fibers were traced using a streamline tractography
algorithm from every voxel throughout the entire volume
that exceeded a minimum fractional anisotropy (Basser,
Pajevic, Pierpaoli, Duda, & Aldroubi, 2000; Conturo et al.,

C.Y. Tang et al. / Intelligence 38 (2010) 293–303

297

Fig. 1. fMRI-SPM group activation clusters superimposed on anatomical MRI (radiological convention), 1 Anterior Cingulate Cortex (ACC), 2 Prefrontal Cortex
(PC) , 3 Parietal Cortex (PC), 4 Internal Capsule (IC), 5 Visual Cortex (VC).

1999; Mori, Crain, Chacko, & van Zijl, 1999). Tracking was
terminated when FA fell below a 0.1 or when the algorithm
encountered a sharp angle change in the principal diffusion
direction between sequential voxels (45°). Second, FA maps
were analyzed using a region of interest (ROI) approach. ROIs
were defined on color-coded tract directional maps by two
independent raters and then averaged. Portions of the following white matter tracts were surveyed using ROIs:
Cingulum Bundle, Internal Capsule, Corpus Callosum, Forceps
Minor, Forceps Major, Inferior- and Superior Longitudinal
Fasciculus and corona radiata (Fig. 2). The ROI voxel locations
were used to extract the FA values, which were then transferred to Statistica V7.1 (Statsoft Inc., Tulsa, OK) and merged
with intelligence factor scores for correlation analysis.
4. Results
4.1. fMRI
As shown in Fig. 1, clusters of activation during the n-back
task were identified (1,2,3 back vs. 0 back; p b .001, uncorrected) in: Anterior Cingulate (ACC), bilateral Prefrontal

Cortex (PFC), bilateral Parietal cortex (PC), bilateral Insular
Cortex (IC) and bilateral visual cortex (VC). As noted, we used
these areas to correlate percent activation with intelligence
factor scores; two regions, although activated during the nback task, did not produce any significant correlations:
namely, Insular Cortex and Visual Cortex.
Significant correlations between percent activation in the
ACC, PFC areas, and PC areas and intelligence factor scores are
shown in Table 3. For the whole sample, activation in the
cluster in the right parietal cortex (PC) was inversely
correlated with general intelligence (r = −.34, p b .03) and
spatial intelligence (r = −.39, p b .01). Spatial intelligence was
also inversely correlated with activation at the right prefrontal cortex (r = −.35, p b .03). For males, activation in the right
prefrontal and right parietal cortices was inversely correlated
with spatial intelligence (r =−.58 and −.45, p b .01 and .04,
respectively). Activation at the right prefrontal cortex was
also inversely correlated with general intelligence in males
(r = −.45, p b .04). Finally, for females, general intelligence
was inversely correlated with activation in left (r = −0.52,
p b 0.02) and right parietal cortex (r = −.50, p b .03). In the
females, numerical intelligence was also inversely related to

Fig. 2. DTI–ROI positions superimposed on directional color-coded fractional anisotropy maps. 1: Superior Corona Radiata, 2: Body Cingulum Bundle, 3: Anterior
Cingulum, 4: Posterior Cingulum, 5: Superior Longitudinal Fasciculus, 6: Body of Corpus Callosum, 7: Genu Corpus Callosum, 8: Splenium Corpus Callosum, 9:
Forceps Minor, 10: Anterior Horn Internal Capsule, 11: Forceps Minor.

298

C.Y. Tang et al. / Intelligence 38 (2010) 293–303

Table 3
Significant correlations of intelligence factor scores with % fMRI activation.
fMRI

Region†

Subjects
All

Males

Females

Spatial
Memory
Numerical
Reasoning
g
Spatial
Memory
Numerical
Reasoning
g
Spatial
Memory
Numerical
Reasoning
g

Ant. Cing.

L-PFC

R-PFC

L-PC

R-PC

− 0.16
− 0.10
− 0.27
0.14
− 0.24
− 0.17
− 0.38
− 0.07
0.04
− 0.23
− 0.24
0.06
− 0.50⁎
− 0.19
− 0.33

− 0.20
0.96
− 0.00
− 0.16
− 0.09
− 0.42
− 0.07
0.07
− 0.24
− 0.26
0.16
0.34
− 0.01
− 0.14
0.13

− 0.35⁎
− 0.03
− 0.24
− 0.08
− 0.28
− 0.58⁎

− 0.25
− 0.21
− 0.24
− 0.02
− 0.27
− 0.42
− 0.07
0.07
− 0.24
− 0.26
− 0.36
− 0.20
− 0.55⁎
− 0.48⁎
− 0.52⁎

− 0.39⁎
0.20
− 0.25
0.05
− 0.34⁎
− 0.45⁎

− 0.20
− 0.19
− 0.16
− 0.45⁎
− 0.27
− 0.15
− 0.35
− 0.00
− 0.30

− 0.08
− 0.04
0.03
− 0.26
− 0.33
− 0.41
− 0.50⁎
− 0.19
− 0.50⁎

† Ant. Cing. is Anterior Cingulate Gyrus, L-PFC: Left, Right Prefrontal Cortex; L,R PC: Left, Right Parietal Cortex. All significances are at p b 0.05.

activation in the anterior cingulate (r = −.50, p b .03), left
(r = −0.55, p b 0.02) and right parietal cortex (r = −.50,
p b .03). Speed of reasoning was inversely correlated with
activation in the left parietal cortex (r = − .48, p b .04). It is
noteworthy that all the correlations were negative (higher
intelligence scores were associated with less brain activation); no significant correlations were positive. Only the
memory factor showed no correlations with activation. None
of the correlations survived Bonferroni correction for multiple
comparisons, although most imaging studies have samples
too small for this correction given the number of possible
comparisons.
4.2. DTI
We computed brute-force streamline tractography (Carpenter et al., 2008; Huang et al., 2004) to identify the
connection networks between the selected areas shown in
Fig. 1. Significant activated clusters in the prefrontal cortex and
parietal cortex were used as seed regions for fiber tracking
through the genu and splenium of the corpus callosum; these
formed the forceps minor and forceps major respectively
Segments of the anterior cingulated where the cingulum was
clearly visible were used as a seed region for tracking of the
cingulum bundle. All fMRI activation clusters were dilated by
one voxel in all directions before being used as seed region
making them closer to nearby white matter tracts for better
tracking. We have integrated the fMRI activated regions
together with the interconnecting white matter tracts in
Fig. 3. This figure illustrates the relationship of the functional
and structural anatomy and the correlations that we have
detected.
We then investigated FA correlates of the intelligence
factors. FA in ROIs in interhemispheric fibers showed correlations with intelligence factors as detailed in Table 4 and shown
in Fig. 3. FA in the genu of the corpus callosum was positively
correlated with the memory factor scores for the whole sample
(r = .34, p b .03) as well as for females (r = .56, p b .01). FA in the
genu of the corpus callosum was also inversely related to
spatial intelligence in males (r = −.51, p b .02). FA in the left
Forceps Major was inversely correlated with general intelli-

gence in males (r = −.53, p b .01) and positively in females
(r = .48, p b .04). FA in the right Forceps Major was also
inversely related to general intelligence in males (r = −.45,
p b .04). Finally, FA in the Splenium of Corpus Callosum was
inversely related to numerical intelligence in males (r = −.46,
p b .04). None of these correlations survived Bonferroni
correction.
ROIs in ipsilateral fibers also showed correlations with intelligence factors as detailed in Table 5 and shown in Fig. 3. FA in
the left Cingulum Bundle was positively related to speed
of reasoning for the whole sample (r=.36, pb .02) and for
males (r=.58, pb .01). FA in the right Cingulum Bundle was also
correlated with speed of reasoning in males (r=.53, pb .01). FA
in the right Superior Corona Radiata was positively correlated
with spatial intelligence in the whole sample (r=.32, pb .04).
Finally, FA in the left Superior Longitudinal Fasciculus was
inversely related to memory in males (r=−.47, pb .03). None
of these correlations survived Bonferroni correction.
5. Discussion
This study examined intelligence factors with g removed
to determine if the parieto-frontal integration theory of intelligence (P-FIT) characterizes specific cognitive abilities
beyond the pervasive influence of the g-factor representing
general intelligence. This was examined separately in males
and females and uniquely combined fMRI and DTI imaging to
study the neuroanatomy of intelligence. Since the sample sizes
are relatively small and none of the significant findings survived correction for multiple comparisons, all results should be
regarded as illustrative and interpreted with caution. Most
neuroimaging studies have small samples and limited statistical power, so reporting uncorrected results allows for potential replication across studies.
5.1. Relationship to P-FIT and other n-back studies
Functional correlations between the BOLD signal obtained
from the working memory task (n-back) and intelligence
factors were mostly detected in the right prefrontal and
bilateral parietal cortices. All these correlations were negative

C.Y. Tang et al. / Intelligence 38 (2010) 293–303

299

Fig. 3. 3D illustration of the significantly correlated fMRI activation clusters* and their interconnecting fiber tracts° for females (top) and males (bottom).
Significant fMRI correlations are labeled with yellow legends, all fMRI correlations are negative with performance scores; Significant white matter tracts
correlations are labeled with white legends where + indicate a positive correlation with better connectivity and − indicates a negative correlation with better
connectivity for the indicated performance scores. (*R,L-PC: right,left parietal cortex; R,L-PFC: right, left prefrontal cortex; ACC: anterior cingulate cortex; ° F-Maj:
forceps major; F-min: forceps minor; GCC: genu of corpus callosum; SCC: splenium of corpus callosum; CB: cingulum bundle, L-SLF: Superior Longitudinal
Fasciculus, C-Rad: Corona Radiata; Only regions or tracts for which significances were found are rendered).

indicating that subjects with high intelligence factor scores
had less activation during the n-back memory task, in support
of the efficiency model of brain function (Haier, 1993; Haier
et al., 1988; Neubauer & Fink, 2009). Gray et al (2003)
reported that, on more demanding n-back conditions, partic-

ipants with higher intelligence scores were more accurate
and showed greater activity in several frontal and parietal
regions. It should be noted that the focus of their analysis was
based on an event related design but their report also
included the results using a block design which showed a

300

C.Y. Tang et al. / Intelligence 38 (2010) 293–303

Table 4
DTI correlations with intelligence factor scores for interhemispheric fibers.
DTI

White matter tracts†

Subjects
All

Males

Females

Spatial
Memory
Numerical
Reasoning
g
Spatial
Memory
Numerical
Reasoning
g
Spatial
Memory
Numerical
Reasoning
g

L Fcp Maj

R Fcp Maj

GCC

SCC

L Fcp Min

R Fcp Min

− 0.11
0.12
0.20
− 0.20
0.10
− 0.08
− 0.10
0.16
0.04
− 0.53⁎
− 0.24
0.24
0.19
− 0.30
0.49⁎

0.01
0.19
− 0.03
− 0.15
0.03
− 0.05
− 0.09
− 0.14
0.22
− 0.45⁎
0.00
0.38
− 0.04
− 0.36
0.14

− 0.28
0.34⁎

− 0.07
0.14
− 0.31
0.22
− 0.15
− 0.00
0.14
− 0.46⁎
0.19
− 0.11
− 0.15
0.15
− 0.21
0.27
− 0.17

0.06
− 10
0.07
0.11
− 0.14
0.28
− 0.10
r− 0.01
− 0.12
0.11
− 0.15
− 0.14
− 0.11
0.45
− 0.32

− 0.06
0.01
− 0.26
− 0.21
0.14
0.03
0.07
0.30
− 0.30
0.10
− 0.26
− 0.18
0.21
0.16
0.13

− 0.21
0.11
− 0.05
− 0.51⁎
0.20
− 0.13
0.33
− 0.17
0.05
0.56⁎
− 0.26
0.30
0.08

† L,R Fcp Maj: Left, Right Forceps Major; GCC: Genu of Corpus Callosum; SCC: Splenium of Corpus Callosum; L,R Fcp Min: Left, Right Forceps Minor. All significances
are at p b 0.05.

trend of lower activity with higher scores on a single test of
fluid intelligence. Waiter et al. (2009) did not find significant
correlations between individual differences in brain activity
during an n-back task and intelligence scores. Activation
levels are known to fluctuate across working memory loads in
an inverted-U shape response (Callicott et al., 1999). The
position of the inverted-U can shift depending on the working
memory capacity of the group or individual (Callicott et al.,
2003). Waiter et al. (2009) used a simple version of the nback task with only 0- and 2-back levels . The limited levels
and range of task difficulties in their study may have created
confounds related to shifts in the inverted-U curve. In addition, their study focused on elderly (mid to late 60-years-old)
subjects, a population that shows a decrease in working
memory capacity and related neurophysiology (Mattay et al.,
2006) as well as a wide variability among individuals in the
extent, rate and pattern of age-related changes that are exhibited at both neural and behavioral levels (Hedden &
Gabrieli, 2004). For the current study, a greater range of
working memory task was used on a younger population,
which could help avoid inconsistencies due to shifts of the

inverted-U. This method does not completely avoid the
potential anomalies as equal sampling of both sides of the
inverted-U is not guaranteed but is less risky than sampling
only one level of working memory difficulty. The optimal
method would include enough levels and range in the task
difficulty to determine the point of peak activation for each
subject.
Our n-back results are in agreement with the single
published study aimed at quantifying the neuro-anatomical
overlap between the general factor of intelligence (g) and
working memory capacity (Colom, Jung, & Haier, 2007). That
study showed that a common neuro-anatomic framework for
these constructs implicates mainly frontal gray matter regions
belonging to the right superior frontal gyrus, the left middle
frontal gyrus, and the right inferior parietal lobule. These
findings (a) were thought to support the role of a discrete
parieto-frontal network, as proposed by the P-FIT model, and (b)
were consistent with Cowan's (2005) theory which distinguished a capacity limit (related to parietal regions) and the
control of attention (related to frontal areas). It was suggested
that capacity limits and attention control relate to the

Table 5
DTI FA correlations with intelligence factor scores for ipsilateral fibers.
DTI

White matter tracts†

Subjects
All

Males

Females

Spatial
Memory
Numerical
Reasoning
g
Spatial
Memory
Numerical
Reasoning
g
Spatial
Memory
Numerical
Reasoning
g

L Cing.

R Cing.

LS Cor R

RS Cor R

L Sup L F

R Sup L F

− 0.08
− 0.25
− 0.05
0.36⁎
− 0.23
− 0.22
− 0.36
− 0.11
0.58⁎

− 0.21
− 0.11
− 0.07
0.34
− 0.09
− 0.38
− 0.30
0.07
0.53⁎
− 0.03
0.04
0.11
− 0.30
0.25
− 0.26

0.32⁎
0.03
− 0.18
− 0.08
− 0.08
0.40
− 0.21
− 0.14
− 0.00
− 0.08
0.25
0.15
− 0.21
− 0.11
− 0.08

− 0.13
− 0.09
0.15
0.05
0.04
− 0.05
− 0.47⁎

− 0.22
0.12
− 0.17
− 0.04
0.13
− 0.32

0.08
− 0.70
− 0.06
− 0.05
− 0.00
− 0.09
− 0.07
− 0.24
0.31
− 0.05
0.37
− 0.09
0.12
− 0.42
0.05

− 0.18
0.19
− 0.10
− 0.06
− 0.01
− 0.17
0.27
− 0.16
0.02
0.07
− 0.21
0.06
− 0.06
0.21
− 0.07

0.10
0.39
− 0.13
− 0.28
0.27
0.16
− 0.25
0.12

† L,R Cing: Left, Right Cingulum Bundle; L, R Cor R; Left, Right Superior Corona Radiata; L, R Sup L F: Left, Right Superior Longitudinal Fasciculus. All significances are
at p b 0.05.

C.Y. Tang et al. / Intelligence 38 (2010) 293–303

commonality between intelligence and working memory. We
also note that we found no correlations between our memory
factor and any fMRI activations, possibly because the factor was
derived as a broader assessment of memory than the more
focused processes required for the n-back task, although this is
not determined.
5.2. Sex differences
Because of the small sample sizes, the sex differences
observed in the present study should be interpreted with
particular caution. Therefore, we do not interpret individual
correlations, but note that all the fMRI correlations with
intelligence factors were negative, consistent with the brain
efficiency hypothesis for both males and females. We tested
the male/female difference of each fMRI/factor correlation
using z-transform. Only the difference for Numerical/Left
Parietal Cortex was significant (p b .045, 2-tailed); there was a
trend for Spatial/Left Prefrontal Cortex (p b .077).
Similarly, the DTI results show sex differences but the same
cautions apply, especially because these analyses were exploratory, so we will discuss specific correlations only to make
general illustrative points about each factor. Using z-transform
comparisons for DTI interhemisphere/factor correlations, the
male/female difference was significant for g/Left Forceps Major
(p b .001, 2-tailed), and there were trends (p b .05 to .09) for
Speed of Reasoning and for g in the Right Forceps Major; for the
Spatial Factor and the Genu of the Corpus Callosum; and, for
Speed of Reasoning and the Left Forceps Minor. For DTI
ipsilateral/factor correlations, the male/female difference was
significant (2-tailed) for Speed of Reasoning/Left Superior
Corona Radiata (p b .026) and for Memory/Left Superior Longitudinal Fasciculus (p b .022); Speed of Reasoning/Left Superior
Longitudinal Fasciculus showed a trend (p b .052). These
differences, for both fMRI and DTI correlations, support the
view that separate analyses by sex are justified, even when
there is no task performance difference.
5.2.1. Spatial factor
In our study, spatial factor scores were correlated with
fMRI activation in the right prefrontal and posterior parietal
cortex in the whole group. However, when analyzed separately for males and females, only the males showed
significant correlations. Although there is a general notion
that spatial ability is a right hemisphere function, recent
studies have found that sex matters. A meta-analysis (Vogel,
Bowers, & Vogel, 2003) showed that females are much less
lateralized than their male counterparts in terms of brain
function and spatial tasks. In that study females showed no
hemispheric preference while males showed a right hemisphere advantage. It was hypothesized that women use both
verbal abilities, a left hemisphere function, as well as spatial
abilities, a right hemisphere function, to accomplish spatial
tasks. This might be contributing to the variances in our
dataset. The lack of any significant correlations in our dataset
for spatial factors in the female group might be additional
support for these findings.
Contrary to our expectation of positive correlations, the DTI
results show that the FA of the genu of the corpus callosum was
negatively correlated with spatial factor scores (r = −51,
p b 0.02) in males. This is the section of the corpus callosum

301

that connects parts of the prefrontal cortex. FA is believed to be
an indirect measure of myelination and the purpose of myelin
is to allow rapid and efficient transmission of signals along the
axons (Ritchie, 1984; Yakovlev & Lecours, 1967). The negative
correlation means that stronger connections between the two
hemispheres decrease performance. Increased anisotropy
(better connectivity) in the corpus callosum can be interpreted
as an interference factor for brains using only one region or
one hemisphere for a certain task. Enhanced interhemispheric
connectivity increases unnecessary information flow from the
contralateral side of the brain, possibly consistent with efficient
information flow. We also note that although both the right
prefrontal and right parietal lobes had significant fMRI signal
correlation with spatial factor scores in males, the prefrontal
cortex showed stronger correlations than the parietal (r = −
0.58 vs. r = −0.45). This is consistent with the DTI finding in the
genu, which showed decreased spatial performance scores
with increased FA in the genu in the male group.
5.2.2. Numerical factor
Females had significant negative correlations with the
numerical factor in terms of activation in the anterior cingulate
as well as both sides of the parietal cortex. Other studies
(Dehaene, Spelke, Pinel, Stanescu, & Tsivkin, 1999) have shown
that numerical cognitive abilities recruit both sides of the
parietal cortex. To our knowledge, no other studies have shown
any gender effect for this function. There were no significant
fMRI correlations with the numerical factor for the male group.
5.2.3. Speed of reasoning factor
The integrity of the cingulum bundle, a white matter tract
inside the cingulate gyrus, as quantified using FA, was positively correlated with the speed of reasoning factor scores in
males. This is not an interhemispheric connection and thus it
is expected that better ipsilateral connections are positively
correlated with performance. The cingulate gyrus is an area in
the brain that is involved with early learning and problem
solving, anticipation of tasks, motivation, and modulation of
emotional responses (Posner & Raichle, 1998).
5.2.4. Memory factor
We note that for the group as a whole there was a
significant correlation (r = 0.34) between the memory factor
and the interhemispheric white matter tracts in the genu of
the corpus callosum (GCC), but when separated by gender we
find that the effect was mainly due to the female group
(r = 0.56) with the males no longer significant. This significant positive correlation is additional support for the notion
that females rely on both hemispheres and benefit from
better interhemispheric connections. No significant correlations between functional activation and the memory factor
were detected. It is conceivable that the GCC connects larger
and more diffuse brain regions used for many memory
processes and that our working memory task uses a more
narrow set of processes.
5.2.5. g-factor
The g-factor was correlated with activation in the right
prefrontal area in males, but was correlated to both left and
right sides of the parietal cortices in females. While the FA of
the forceps major was negatively correlated with g in males,

302

C.Y. Tang et al. / Intelligence 38 (2010) 293–303

the females had parts of the forceps major positively
correlated with g. Again, this negative correlation may be an
indicator of interference from contralateral side of the brain
in males who rely mostly on the right side of the brain. The
positive FA correlations in the forceps major in females are
also compatible with the bilateral fMRI correlations in the
parietal regions connected by the forceps major. Also, our
finding that part of the FA of the forceps major (an extension
of the splenium) is positively correlated with g (r = 0.49,
p b 0.034) in females is consistent with some evidence that
the splenium may be larger in females (Dubb, Gur, Avants, &
Gee, 2003). Overall, our DTI results are also consistent with a
report on DTI/IQ correlations in a cohort of subjects aged 5–18
where correlations were mainly negative in older males and
positive in older females (Schmithorst et al., 2008). There are
also other higher-order latent variable models that show sex
differences that could be explored with imaging (Johnson &
Bouchard, 2007a,b; Keith, Reynolds, Patel, & Ridley, 2008).
In summary, in this study, we have shown that: (a)
individual differences in activation during the n-back task are
correlated with the general intelligence factor (g), as well as
to distilled estimates (removing g) of speed of reasoning,
numerical ability, and spatial ability, but not to memory, (b)
the correlations are mainly bilateral for females and unilateral
for males, and (c) differences in the integrity of the axonal
connections are also related to the functional findings:
integrity of interhemispheric connections is positively related
to several intelligence factors in females but is negatively
correlated in males. It should be noted that none of these
correlations survived correction for multiple comparisons, so
replication in larger samples is necessary. Nevertheless, these
results show encouraging trends to support further studies
regarding the relationship between working memory and
intelligence factors. They support the growing recognition
that brain structure and function underlie individual differences in general intelligence and other specific cognitive
abilities. These results also underscore the importance of
analyzing neuroimaging data separately for males and females.
Acknowledgements
Funding for imaging and for R. Haier was provided by the
Johnson O'Connor Research Support Corporation. R. Colom
was funded by the grant SEJ-2006-07890 from the “Ministerio de Educación y Cultura” (MEC) [Ministry of Education
and Culture, Spain].
References
Basser, P. J. (1997). New histological and physiological stains derived from
diffusion-tensor MR images. Annals of the New York Academy of Sciences,
820, 123−138.
Basser, P. J., Pajevic, S., Pierpaoli, C., Duda, J., & Aldroubi, A. (2000). In vivo
fiber tractography using DT-MRI data. Magnetic Resonance in Medicine,
44(4), 625−632.
Cabeza, R., & Nyberg, L. (2000). Imaging cognition II: An empirical review of
275 PET and fMRI studies. Journal of Cognitive Neuroscience, 12(1), 1−47.
Callicott, J. H., Mattay, V. S., Bertolino, A., Finn, K., Coppola, R., Frank, J. A., et al.
(1999). Physiological characteristics of capacity constraints in working
memory as revealed by functional MRI. Cerebral Cortex, 9(1), 20−26.
Callicott, J. H., Mattay, V. S., Verchinski, B. A., Marenco, S., Egan, M. F., &
Weinberger, D. R. (2003). Complexity of prefrontal cortical dysfunction
in schizophrenia: More than up or down. American Journal of Psychiatry,
160(12), 2209−2215.

Carpenter, D. M., Tang, C. Y., Friedman, J. I., Hof, P. R., Stewart, D. G.,
Buchsbaum, M. S., et al. (2008). Temporal characteristics of tract-specific
anisotropy abnormalities in schizophrenia. NeuroReport, 19(14),
1369−1372.
Chiang, M. C., Barysheva, M., Shattuck, D. W., Lee, A. D., Madsen, S. K.,
Avedissian, C., et al. (2009). Genetics of brain fiber architecture and
intellectual performance. Journal of Neuroscience, 29(7), 2212−2224.
Colom, R. (2007). Intelligence? What intelligence? Behavioral and Brain
Sciences, 30, 155−156.
Colom, R., Abad, F., Quiroga, M. A., Shih, P. C., & Flores-Mendoza, C. (2008).
Working memory and intelligence are highly related constructs, but
why? Intelligence, 36, 584−606.
Colom, R., Abad, F., Rebollo, I., & Shih, P. C. (2005). Memory span and general
intelligence: A latent variable approach. Intelligence, 33, 623−642.
Colom, R., Haier, R. J., Head, K., Alvarez-Linera, J., Quiroga, M. A., Shih, P. C.,
et al. (2009). Gray matter correlates of fluid, crystallized, and spatial
intelligence: Testing the P-FIT model. Intelligence, 37(2), 124−135.
Colom, R., Jung, R. E., & Haier, R. J. (2007). General intelligence and memory
span: Evidence for a common neuroanatomic framework. Cognitive
Neuropsychology, 24(8), 867−878.
Colom, R., Rebollo, I., Palacios, A., Juan-Espinosa, M., & Kyllonen, P. C. (2004).
Working memory is (almost) perfectly predicted by g. Intelligence, 32,
277−296.
Condon, C. A., & Schroeder, D. H. (2003). Establishing the factor structure of the
Johnson O'Connor Research Foundation's test battery. Paper presented at
the annual meeting of the International Society for Intelligence Research,
Newport Beach, CA.
Condon, C. A., & Schroeder, D. H. (2005). Quality of school and standard test
battery performance (Statistical Bulletin 2005-3). Chicago: Johnson O'
Connor Research Foundation.
Conturo, T. E., Lori, N. F., Cull, T. S., Akbudak, E., Snyder, A. Z., Shimony, J. S.,
et al. (1999, August). Tracking neuronal fiber pathways in the living
human brain. Proceedings of the National Academy of Science, 96,
10422−10427.
Cook, P. A. (2006). Camino: Open-source diffusion-MRI reconstruction and
processing 14th Scientific Meeting of the International Magnetic Resonance
in Medicine (pp. 2759). Washington: Seattle.
Cowan, N. (2005). Working memory capacity. New York: Psychology Press.
Dehaene, S., Spelke, E., Pinel, P., Stanescu, R., & Tsivkin, S. (1999). Sources of
mathematical thinking: Behavioral and brain-imaging evidence. Science,
284(5416), 970−974.
Dubb, A., Gur, R., Avants, B., & Gee, J. (2003). Characterization of sexual
dimorphism in the human corpus callosum. Neuroimage, 20(1), 512−519.
Engle, R. W. (2002). Working memory capacity as executive attention.
Current Directions in Psychological Science, 11, 19−23.
Fjell, A. M., Westlye, L. T., Greve, D. N., Fischl, B., Benner, T., van der Kouwe, A. J.,
et al. (2008). The relationship between diffusion tensor imaging and
volumetry as measures of white matter properties. Neuroimage, 42(4),
1654−1668.
Grabner, R. H., Fink, A., Stipacek, A., Neuper, C., & Neubauer, A. C. (2004).
Intelligence and working memory systems: Evidence of neural efficiency
in alpha band ERD. Brain Research. Cognitive Brain Research, 20(2),
212−225.
Gray, J. R., Chabris, C. F., & Braver, T. S. (2003). Neural mechanisms of general
fluid intelligence. Nature Neuroscience, 6(3), 316−322.
Haier, R. J. (2009). Neuro-intelligence, neuro-metrics and the next phase of
brain imaging studies. Intelligence, 37(2), 121−123.
Haier, R. J. (Ed.). (1993). Cerebral glucose metabolism and intelligence.
Norwood, NJ: Ablex. ed.
Haier, R. J., & Benbow, C. P. (1995). Sex differences and lateralization in
temporal lobe glucose metabolism during mathematical reasoning.
Developmental Neuropsychology, 11, 405−414.
Haier, R. J., Colom, R., Schroeder, D. H., Condon, C. A., Tang, C., Eaves, E., et al.
(2009). Gray matter and intelligence factors: Is there a neuro-g?
Intelligence, 37(2), 136−144.
Haier, R. J., Jung, R. E., Yeo, R. A., Head, K., & Alkire, M. T. (2005). The
neuroanatomy of general intelligence: Sex matters. Neuroimage, 25(1),
320−327.
Haier, R. J., Siegel, B. V., Nuechterlein, K. H., Hazlett, E., Wu, J. C., Paek, J., et al.
(1988). Cortical glucose metabolic-rate correlates of abstract reasoning
and attention studied with positron emission tomography. Intelligence,
12(2), 199−217.
Halpern, D. F. (2000). Sex differences in cognitive abilities, 3rd ed. Mahwah,
NJ: Erlbaum.
Hedden, T., & Gabrieli, J. D. (2004). Insights into the ageing mind: A
view from cognitive neuroscience. Nature Reviews. Neuroscience, 5(2),
87−96.
Huang, H., Zhang, J., van Zijl, P. C., & Mori, S. (2004). Analysis of noise effects
on DTI-based tractography using the brute-force and multi-ROI
approach. Magnetic Resonance in Medicine, 52(3), 559−565.

C.Y. Tang et al. / Intelligence 38 (2010) 293–303
Jausovec, N., & Jausovec, K. (2008). Spatial rotation and recognizing
emotions: Gender related differences in brain activity. Intelligence, 36(5),
383−393.
Johnson, W., & Bouchard, T. J. (2007). Sex differences in mental abilities: g
masks the dimensions on which they lie. Intelligence, 35(1), 23−39.
Johnson, W., & Bouchard, T. J. (2007). Sex differences in mental ability: A
proposed means to link them to brain structure and function. Intelligence,
35(3), 197−209.
Johnson, W., Jung, R. E., Colom, R., & Haier, R. J. (2008). Cognitive abilities
independent of IQ correlate with regional brain structure. Intelligence,
36(1), 18−28.
Jung, R. E., & Haier, R. J. (2007). The Parieto-Frontal Integration Theory (PFIT) of intelligence: Converging neuroimaging evidence.Behavioral and
Brain Sciences, 30(2), 135−154 discussion 154–187.
Kane, M. J., Hambrick, D. Z., & Conway, A. R. (2005). Working memory
capacity and fluid intelligence are strongly related constructs: Comment
on Ackerman, Beier, and Boyle (2005).Psychological Bulletin, 131(1),
66−71 author reply 72-65.
Keith, T. Z., Reynolds, M. R., Patel, P. G., & Ridley, K. P. (2008). Sex differences
in latent cognitive abilities ages 6 to 59: Evidence from the Woodcock–
Johnson III tests of cognitive abilities. Intelligence, 36(6), 502−525.
Lee, K. H., Choi, Y. Y., Gray, J. R., Cho, S. H., Chae, J. H., Lee, S., et al. (2006).
Neural correlates of superior intelligence: Stronger recruitment of
posterior parietal cortex. Neuroimage, 29(2), 578−586.
Luders, E., Narr, K. L., Bilder, R. M., Szeszko, P. R., Gurbani, M. N., Hamilton, L.,
et al. (2008). Mapping the relationship between cortical convolution and
intelligence: Effects of gender. Cerebral Cortex, 18(9), 2019−2026.
Luders, E., Narr, K. L., Thompson, P. M., Rex, D. E., Jancke, L., Steinmetz, H.,
et al. (2004). Gender differences in cortical complexity. Nature
Neuroscience, 7(8), 799−800.
Mattay, V. S., Fera, F., Tessitore, A., Hariri, A. R., Berman, K. F., Das, S., et al.
(2006). Neurophysiological correlates of age-related changes in working
memory capacity. Neuroscience Letters, 392(1–2), 32−37.
Mori, S., Crain, B. J., Chacko, V. P., & van Zijl, P. C. (1999). Three-dimensional
tracking of axonal projections in the brain by magnetic resonance
imaging. Annals of Neurology, 45(2), 265−269.
Naghavi, H. R., & Nyberg, L. (2007). Integrative action in the fronto-parietal
network: A cure for a scattered mind. Behavioral and Brain Sciences, 30,
161−162.
Neubauer, A. C., & Fink, A. (2009). Intelligence and neural efficiency.
Neuroscience and Biobehavioral Reviews, 33(7), 1004−1023.
Neubauer, A. C., Fink, A., & Schrausser, D. G. (2002). Intelligence and neural
efficiency: The influence of task content and sex on the brain–IQ
relationship. Intelligence, 30(6), 515−536.

303

Oberauer, K., Schulze, R., Wilhelm, O., & Suss, H. M. (2005). Working memory
and intelligence—Their correlation and their relation: Comment on
Ackerman, Beier, and Boyle (2005).Psychological Bulletin, 131(1), 61−65
author reply 72-65.
Posner, M. I., & Raichle, M. E. (1998, February). The neuroimaging or human
brain function. Proceedings of the National Academy of Sciences of the
United States of America Colloquium Paper, 95, 763−764.
Prabhakaran, V., Smith, J. A. L., Desmond, J. E., Glover, G. H., & Gabrieli, J. D. E.
(1997). Neural substrates of fluid reasoning: An fMRI study of
neocortical activation during performance of the Raven's Progressive
Matrices Test. Cognitive Psychology, 33(1), 43−63.
Rabinowicz, T., Dean, D. E., Petetot, J. M. C., & de Courten-Myers, G. M. (1999).
Gender differences in the human cerebral cortex: More neurons in males;
More processes in females. Journal of Child Neurology, 14(2), 98−107.
Ritchie, J. M. (1984). Physiological basis of conduction in myelinated nerve
fibers. myelin (pp. 117−146). New York: Plenum.
Schmithorst, V. J., & Holland, S. K. (2007). Sex differences in the development of
neuroanatomical functional connectivity underlying intelligence found
using Bayesian connectivity analysis. Neuroimage, 35(1), 406−419.
Schmithorst, V. J., Holland, S. K., & Dardzinski, B. J. (2008). Developmental
differences in white matter architecture between boys and girls. Human
Brain Mapping, 29(6), 696−710.
Sowell, E. R., Peterson, B. S., Kan, E., Woods, R. P., Yoshii, J., Bansal, R., et al. (2007).
Sex differences in cortical thickness mapped in 176 healthy individuals
between 7 and 87 years of age. Cerebral Cortex, 17(7), 1550−1560.
Toga, A. W., & Thompson, P. M. (2005). Genetics of brain structure and
intelligence. Annual Review of Neuroscience, 28, 1−23.
Vogel, J. J., Bowers, C. A., & Vogel, D. S. (2003). Cerebral lateralization of
spatial abilities: A meta-analysis. Brain and Cognition, 52(2), 197−204.
Wager, T. D., Jonides, J., & Reading, S. (2004). Neuroimaging studies of
shifting attention: A meta-analysis. Neuroimage, 22(4), 1679−1693.
Wager, T. D., & Smith, E. E. (2003). Neuroimaging studies of working
memory: A meta-analysis. Cognitive, Affective & Behavioral Neuroscience,
3(4), 255−274.
Waiter, G. D., Deary, I. J., Staff, R. T., Murray, A. D., Fox, H. C., Starr, J. M., et al.
(2009). Exploring possible neural mechanisms of intelligence differences
using processing speed and working memory tasks: An fMRI study.
Intelligence, 37, 199−206.
Yakovlev, P. I., & Lecours, A. R. (1967). The myelogenetic cycles of regional
maturation of the brain. In A. Minkowski (Ed.), Regional development of
the brain in early life (pp. 3−70). : Oxford: Blackwell Scientific.
Yu, C., Li, J., Liu, Y., Qin, W., Li, Y., Shu, N., et al. (2008). White matter tract
integrity and intelligence in patients with mental retardation and
healthy adults. Neuroimage, 40(4), 1533−1541.