This article appeared in a journal published by Elsevier. The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
Author's personal copy
Neuroscience Letters 479 (2010) 201–205
Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/neulet
Dietary supplementation of creatine monohydrate reduces the human fMRI
Stephen T. Hammett∗, Matthew B. Wall, Thomas C. Edwards, Andrew T. Smith
Department of Psychology, Royal Holloway University of London, Egham Hill, Egham, Surrey TW20 0EX, UK
a r t i c l ei n f o
Received 3 February 2010
Received in revised form 27 April 2010
Accepted 18 May 2010
a b s t r a c t
Creatine monohydrate is an organic acid that plays a key role in ATP re-synthesis. Creatine levels in the
human brain vary considerably and dietary supplementation has been found to enhance cognitive per-
formance in healthy individuals. To explore the possibility that the fMRI Blood Oxygen Level Dependent
(BOLD) response is influenced by creatine levels, BOLD responses to visual stimuli were measured in
visual cortex before and after a week of creatine administration in healthy human volunteers. The mag-
nitude of the BOLD response decreased by 16% following creatine supplementation of a similar dose to
that previously shown to increase cerebral levels of phosphocreatine. We also confirmed that cognitive
performance (memory span) is increased. These changes were not found in a placebo group. Possible
mechanisms of BOLD change are considered. The results offer potential for insight into the coupling
between neural activity and the BOLD response and the more immediate possibility of accounting for
an important source of variability during fMRI analysis in clinical studies and other investigations where
between-subjects variance is an issue.
© 2010 Elsevier Ireland Ltd. All rights reserved.
Creatine monohydrate (Cr) is a naturally occurring organic acid
that acts as a buffer for cytosolic and mitochondrial pools of
adenosine triphosphate (ATP). It increases an organism’s ability
to re-synthesise ATP from ADP by using the energy provided by
the phosphate bond from phosphocreatine (PCr). Oral administra-
tion of Cr has been found to enhance performance on memory and
intelligence tests in both a young healthy vegetarian cohort and in
sleep-deprived subjects and it has been suggested that this effect
is related to an enhancement in cerebral energetics [22,20]. Cr is
known to increase oxidative metabolism in skeletal muscle  and
oral supplementation has been demonstrated to increase cerebral
ATP and PCr concentrations . However, Cr is an ergogenic com-
pound that enhances the ability of all cells to re-synthesise ATP.
enhancement of cerebral metabolism. Nevertheless, the results are
striking and if Cr has a direct effect upon cerebral metabolism then
the implications, particularly for functional Magnetic Resonance
is that neural activity is tightly coupled to the BOLD signal  but
the precise nature of such coupling is uncertain and remains an
area of active research [27,23]. Whilst it is known that the turnover
of creatine kinase increases in visual cortex during visual stimula-
∗Corresponding author. Tel.: +44 1784 443702; fax: +44 1784 443702.
E-mail address: firstname.lastname@example.org (S.T. Hammett).
tion , there is as yet no evidence that Cr levels directly affect the
magnitude of the BOLD response.
To establish whether the magnitude of the BOLD response is
influenced by Cr levels, we have measured responses to visual
stimuli in the primary visual cortex (V1) of 22 healthy human
volunteers using fMRI, before and after oral administration of Cr
or a placebo (11 in the Cr group and 11 in the placebo group).
We chose to measure responses in V1 since it is the only corti-
cal area in which an increase in creatine kinase turnover during
stimulation has been documented  and we chose a dosing
regime that is known to lead to an increase in PCr in occipital
cortex . Despite reports of large increases in cognitive perfor-
mance in healthy young and older adults [22,20], one recent study
failed to find cognitive enhancement . We therefore also mea-
sured cognitive performance to verify previous reports of cognitive
The experiment was a standard fMRI event-related design. Each
trial lasted 2s and consisted of a presentation of a counter-phasing
checkerboard. Inter-trial intervals (ITI) were determined by a Pois-
run included 40 trials of each stimulus contrast and a 10s buffer
For six subjects in each group, retinotopic data were acquired in a
previous session. These subjects completed three runs of 604s in
each session. A further five subjects in each group completed their
retinotopic scans at the beginning of their first session in week 1.
0304-3940/$ – see front matter © 2010 Elsevier Ireland Ltd. All rights reserved.
Author's personal copy
S.T. Hammett et al. / Neuroscience Letters 479 (2010) 201–205
These subjects completed two runs of 604s in each session. With
this exception, all scanning sessions were identical.
MRI images were acquired with a 3-Tesla Siemens Magnetom
Trio scanner with an 8-channel array head coil. Anatomical (T1-
weighted) images were obtained at the start of each scanning
session (MPRAGE, 160 axial slices, in-plane resolution 256×256,
1mm isotropic voxels, TR=1830ms, TE=4.43ms, flip angle=11◦,
bandwidth=130Hz/pixel), followed by functional data acquisition
with a gradient echo, echoplanar sequence (TR=2s, 28 contiguous
axial slices, interleaved acquisition order, 3mm isotropic voxels,
in-plane resolution of 64×64 voxels, flip angle=90◦, TE=30ms,
bandwidth=1396Hz/pixel). Functional scanning runs consisted of
formed with BrainVoyager QX (version 1.9; Brain Innovation, Inc.,
The Netherlands). Functional data were pre-processed to correct
for head-motion and slice-timing, and filtered with a temporal
high-pass filter (0.014Hz). No spatial smoothing was performed
on the functional data. Functional images from both sessions were
co-registered to a high-quality anatomical image (MPRAGE) of the
same subject, acquired in the first session. A standard GLM anal-
ysis was performed with regressors related to the two stimulus
priate model. The phase of the fitted model at each voxel was
visual areas. An area corresponding to V1 was defined on the 2D
3D space. ROI-based GLM analyses were subsequently conducted
for each subject yielding signal amplitude (beta) values for each
time-point, group and contrast condition. These were converted
to % signal change using the values of constant (session-related)
regressors as denominators.
Stimuli were back-projected onto a screen mounted in the rear
of the scanner bore by a computer-controlled LCD projector. Sub-
on the head coil that provided a horizontal view of approximately
30◦. The stimuli consisted of low (10%) and high (nominally 100%)
contrast counter-phase checkerboards whose contrast reversed at
8Hz. The check size subtended approximately 2◦and the entire
stimulus subtended approximately 20◦×20◦. On half the trials
the checkerboard was presented at high contrast and half of the
trials were presented at low contrast. The central 2◦and back-
ground of the stimulus was medium grey and contained a small
were acquired using a counterphase (8Hz) checkerboard “wedge”
stimulus (a 24◦sector) of radius 12◦. Check size was scaled in
wedge rotated clockwise at a rate of 64s/cycle, eight cycles were
Creatine supplementation (Sci-Mx: Gloucestershire, UK) was
provided at a dose of 20g/day for five days, followed by two addi-
tional days at a dose of 5g/day. Subjects returned to the lab once
one that was taken immediately, and one that they were instructed
to take later the same day. Subjects were provided with additional
doses to cover weekends, and any other days when it was impossi-
the dosing regimen. The placebo group was given the same dosage
of maltodextrin. All subjects were informed that they were receiv-
and scanning sessions began. We used maltodextrin, a polysaccha-
then lowers insulin levels. Therefore, even following a large phys-
iological dose, preprandial levels would be restored rapidly. The
doses administered in this experiment were small in comparison
to typical dietary intake (e.g. around 20% of the glucose content of a
ful degradation in Cr concentration known to occur after 5g (the
any acute modulation of the BOLD response. The mean and median
age of the Cr group was 30.18 and 27 years (SD=8.37) respec-
tively and the mean and median age of the placebo group was 25
years (SD=4.82). There was no significant difference between the
age of the groups (t=1.779, df=20, p=0.091). All were students or
employees at Royal Holloway University of London who reported
that they had not used Cr supplementation within the last three
years. Experiments were conducted in accordance with the Decla-
ration of Helsinki, approved by a local ethics committee at Royal
Holloway, and written informed consent was obtained. Standard
MRI screening procedures were followed.
lowing Cr supplementation we also measured performance on the
Backwards Digit Span (BDS)  and Raven’s Advanced Progres-
sive Matrices (RAPM)  prior to each scan. The BDS comprises
a set of number sequences of increasing length with two different
sequences of each length. Subjects were required to repeat each
sequence backwards. The test was terminated when the subject
failed to repeat two sequences of the same length. Different num-
required to complete as many items of the RAPM as possible in
5min. Since the RAPM tests are ordered in terms of difficulty, odd-
numbered and even-numbered tests were administered on weeks
1 and 2 respectively. Whilst we did not counterbalance adminis-
tration of odd and even-numbered tests, the differential difficulty
between contiguous tests is small and our analyses revealed that
there was no significant difference in performance in the placebo
groups in week 1 (t=0.187, df=20, p=0.854).
There was a clear reduction in BOLD amplitude following Cr
supplementation that was not seen in the placebo condition. The
results are summarized in Fig. 1 (left panel). They are expressed in
terms of the log2ratio of BOLD response obtained after Cr/placebo
to that obtained, in exactly the same region of interest, in the initial
scan. Thus, a value of zero represents no change, positive numbers
indicate an increase and negative numbers a decrease. The very
large reduction in BOLD response following Cr supplementation
for one subject in one (low contrast) condition may be consid-
ered an outlier (z=−4.22). However, the subject’s response in this
condition was consistent across hemispheres and is graphed for
completeness. Removal of this subject’s data does not affect the
significance of our main finding of reduced BOLD signal following
There was no significant difference between Placebo and Cr
BOLD responses to visual stimuli in V1 before Cr/placebo sup-
plementation for either low (t=0.458, df=20, p=0.652) or high
contrast (t=0.204, df=20, p=0.84). There was no significant differ-
ence in the effect of Cr with respect to stimulus contrast (t=0.5643,
df=20). However, the mean BOLD response was 16.4% lower fol-
lowing Cr supplementation and 9.6% higher following placebo
a significant interaction effect (F(1, 42)=4.81, p=0.034) between
compound (Cr/placebo) and week (pre-/post-supplementation)
and individual t tests revealed a significant reduction in
BOLD response following Cr supplementation (t=2.791, df=21,
p=0.0109) but no significant change in BOLD response following
placebo supplementation (t=0.659, df=21, p=0.517). One-sample
t tests, averaged across contrast conditions, similarly revealed a
significant reduction in BOLD relative to no change for the Cr
group (t=2.551, df=21, p=0.0186) but no significant difference for
Author's personal copy
S.T. Hammett et al. / Neuroscience Letters 479 (2010) 201–205
Fig. 1. Left panel: The ratio of the BOLD response before and after Cr/placebo. The ratio for each subject, averaged across the two hemispheres, is represented by a circle
(Cr) or triangle (placebo). Open symbols represent high contrast and closed symbols represent low contrast stimuli. Long horizontal lines represent the mean, error bars
represent±1 S.E.M. (n=11). A value of zero (left y axis) indicates no change after supplementation. Right panel: BOLD response to visual stimuli before (left) and after (right)
administration of Cr (top panels) or placebo (bottom panels) for a typical subject from each group (the one whose individual results are closest to the group mean). Activation
is shown as an overlay coloured according to the t value obtained for each voxel. Activation maps were thresholded at p=0.001 (uncorrected for multiple comparisons).
the placebo group (t=1.038, df=21, p=0.3105). The difference in
response between week 1 and week 2 is also significantly different
the potential oulier (see Fig. 1) revealed the same pattern of results
(t=2.192, df=41, p=0.0341, two-tailed). The variance in response
difference between week 1 and week 2 was, however, signifcantly
lower for the Cr group (F=4.165, df=21,20, p=0.0023) once the
potential outlier was removed. In order to estabish whether Cr
also had an effect on the dynamics of the haemodynamic response
we also conducted a 3×2 mixed design ANOVA on the time-to-
peak of the BOLD signal (determined by the maximum value for
each condition and each subject of the averaged event-related
time courses). Week and contrast were evaluated as within groups
variables and compound (Cr vs placebo) as a between groups vari-
able. The results indicate no significant main effect of week (F(1,
20)=1.045, p>0.05) on the time-to-peak of the BOLD signal.
The effect of Cr is also illustrated in Fig. 1 (right panel) for a
tal slice through the visual cortex. The decrease in BOLD response
following creatine but not placebo is clearly visible.
Following Cr, mean Backward Digit Span (BDS) increased sig-
nificantly by 26.9% (t=3.39, df=10, p=0.0069) whilst there was
no significant change in BDS following placebo (t=0.43, df=10,
p 0.6761). Performance on the RAPM increased non-significantly
by 9.6% following Cr (t=1.882, df=10, p=0.0745) and reduced
non-significantly by 4.5% (t=0.7733, df=10, p=0.4572) following
placebo. A Group×Week ANOVA revealed a main effect of week
(F(1, 20)=5.75, p=0.026, two-tailed) and a significant interaction
for BDS performance. No significant effects were found for RAPM
Our principal result demonstrates that the BOLD response in
V1, generally considered to be coupled to metabolic demand and
neural activity, is reduced following ingestion of Cr for one week.
Our results also show an increase in backward memory span con-
sistent with previous reports using similar doses [22,20] and a
non-significant increase in RAPM performance that may possibly
be due to insufficient power since a significant increase in RAPM
post-Cr but used a dose that was far smaller than previous studies
and was typically around 10% of the dose used here. It is possible
that these large differences in dose account for the discrepancy.
Why should the BOLD response be reduced following Cr supple-
mentation? One possibility is that the effect is related to elevation
of cerebral PCr which allows for more efficient synthesis of ATP.
Creatine is known to cross the blood-brain barrier and oral sup-
plementation at a very similar dose and duration to that which we
have used increases ATP and PCr concentrations within the brain
. Thus, the energetic advantage conferred by elevated PCr may
lead to the substantial reduction in the BOLD response we find.
However, the signalling pathways that underlie neuro-metabolic
and neuro-vascular coupling are not well understood. Tradition-
ally, the increase in CBF (and thus BOLD response) attendant upon
local cortical activity has been thought to reflect a response to O2
depletion and CO2increases, whilst the cerebral metabolic rate of
glucose and O2(CMRglu and CMR02, respectively) have been con-
bral measurements of these parameters in human primary visual
cortex indicate that whilst CBF and CMRglu increase by circa 50%
during visual stimulation, O2consumption is raised by around 5%
[12,18]. Thus the majority of the extra O2available during stimula-
tion is not usually uptaken.
There is a number of plausible mechanisms by which increased
Cr levels may reduce the BOLD response. Three possibilities are (1)
Cr elevation reduces metabolic demand (by elevating PCr) and the
stimulus-related change in CBF is consequently reduced. However,
there is considerable evidence to suggest that such a straightfor-
(2) Cr influences the mechanisms of the vascular response with no
attendant change in metabolic demand. This is akin to suggesting
atively direct vascular changes  and there is good evidence that
the BOLD response is modulated by basal state . Thus we cannot
rule out the possibility that our results reflect an increase in basal
Author's personal copy
S.T. Hammett et al. / Neuroscience Letters 479 (2010) 201–205
perfusion and no, or a proportionately smaller, change in BOLD
response. However, studies of the relation between basal state and
BOLD magnitude have used hypercapnic and other manipulations
that exclusively modulate perfusion. Cr’s ubiquity and physiolog-
ical properties render it unlikely to have an exclusively vascular
effect. (3) Cr enhances the normally relatively low uptake of avail-
able O2(thus reducing the ratio of oxy- to deoxy-haemoglobin). Cr
may lead to an increase in CMRO2by providing a more direct and
abundant energetic pool for oxidative glycolysis, reducing O2lev-
els and therefore the BOLD response. If this is the case, our results
represent an increase in aerobic metabolic activity coupled with a
decrease in the BOLD response and imply that the amount of freely
available ATP in the brain is not tightly coupled with the haemo-
dynamic response. This interpretation of the BOLD reduction is
plausible since (1) Cr is known to increase oxidation in skeletal
muscle  and (2) there is considerable evidence for a decoupling
of CBF and metabolic rate [1,2]. However, other plausible possibil-
ities exist such as a more direct neuromodulatory effect, since Cr
has recently been implicated in modulation of both glutamatergic
and GABAergic function [26,17], and further work will be required
to resolve the issue.
Regardless of the underlying cause, the substantial reduction in
the BOLD response following Cr supplementation may have gen-
eral consequences for the interpretation of fMRI data. Endogenous
levels of PCr and ATP vary across the brain and oral Cr supplemen-
tation has been shown to yield region-dependent elevation in Cr
levels [21,10]. The coupling of BOLD to Cr supplementation that we
find suggests that meaningful comparison of responses in differ-
ent areas of the brain requires consideration of the relative levels
of PCr across regions. Indeed, it has recently been reported that
similar changes in energy metabolism in the lentiform nuclei and
visual cortex result in substantially lower BOLD response in the
former sub-cortical region  and that levels of PCr in the stria-
tum are substantially lower than in cortical areas . Moreover,
dietary dependent  and independent  variations in Cr lev-
els may well be a major source of between-subjects variance in
fMRI. If so, careful control could potentially yield real benefits in
terms of across-subject statistical signal detection, particularly in
studies that compare BOLD responses in different populations. Dif-
ferences in the shape of the BOLD signal across subjects have often
been observed  and, given that Cr acts as both a spatial and
temporal buffer for ATP, it is conceivable that dietary intake of Cr
might contribute to such differences in temporal dynamics as well
as response amplitude.
The Cr-induced reduction in BOLD may also bear upon clinical
intervention and interpretation in Alzheimer’s disease (AD). It has
previously been suggested that Cr may hold promise in the treat-
ment of AD  and our current results are certainly encouraging
in this respect. The effects of Cr administration we find in healthy
humans – reduced BOLD response and enhanced memory – are the
reverse of those found in pre-symptomatic humans at high risk
of developing AD by virtue of carrying the APOE e4 allele [5,4].
Whether Cr supplementation will have any significant therapeu-
tic effect in Alzheimer’s is unknown. However the current results,
potential therapeutic agent, and investigation of this therapeutic
potential and its ability to inform clinical interpretation is clearly a
In summary, we have shown that Cr supplementation reduces
the fMRI BOLD response by 16% whilst increasing memory span by
in cerebral glucose oxidation or basal perfusion. Whilst it is gener-
ally assumed that the larger the BOLD response, the greater the
underlying neural activity, our reasoning suggests that a reduction
in BOLD may occur in response to an increase in oxidative glycol-
ysis. If so, the amount of freely available ATP in the brain may not
be tightly coupled with the haemodynamic response. Regardless of
the precise mechanism of action, variation in brain Cr may explain
some of the between-subjects variance in BOLD responses across a
for investigating the nature of the BOLD signal.
The authors thank Emma McHarg for extensive help with sta-
tistical analyses and many helpful comments and thank Pavlos
Alifragis, Jonas Larsson, Krish Singh and Robin Williams for many
helpful discussions. Matthew Wall is currently at the Institute of
Neurology, UCL, and GlaxoSmithkline Clinical Imaging Centre, UK.
differences in the coupling of cerebral blood flow and oxygen metabolism
changes in response to activation: implications for BOLD-fMRI, Neuroimage
39 (2008) 1510–1521.
 D. Attwell, C. Iadecola, The neural basis of functional brain imaging signals,
Trends Neurosci. 25 (2002) 621–625.
 D. Barr, L. Wilder, S. Caudill, A. Gonzalez, L. Needham, J. Pirkle, Urinary
creatinine concentrations in the U.S. population: implications for urinary
biologic monitoring measurements, Environ. Health Perspect. 113 (2005)
 M.W. Bondi, W.S. Houston, L.T. Eyler, G.G. Brown, fMRI evidence of com-
pensatory mechanisms in older adults at genetic risk for Alzheimer disease,
Neurology 64 (2005) 501–508.
J.C. Mazziotta, G.W. Small, Patterns of brain activation in people at risk for
Alzheimer’s disease, N. Engl. J. Med. 343 (2000) 450–456.
 G.G. Brown, L.T. Eyler Zorrilla, B. Georgy, S.S. Kindermann, E.C. Wong, R.B.
Buxton, BOLD and perfusion response to finger-thumb apposition after aceta-
zolamide administration: differential relationship to global perfusion, J. Cereb.
Blood Flow Metab. 23 (2003) 829–837.
 T.S. Burklen, U. Schlattner, R. Homayouni, K. Gough, M. Rak, A. Szeghalmi, T.
Wallimann, The creatine kinase/creatine connection to Alzheimer’s disease:
CK-inactivation, APP-CK complexes and focal creatine deposits, J. Biomed.
Biotechnol. 2006 (2006) 35936.
and AMPK phosphorylation and reduces lactate production in L6 rat skeletal
muscle cells, J. Physiol. 555 (2004) 409–421.
 W. Chen, X.-H. Zhu, G. Adriany, K. Ugurbil, Increase of creatine kinase activ-
ity in the visual cortex of human brain during visual stimulation: a 32P NMR
magnetization transfer study, MRM 38 (1997) 551–557.
 P. Dechent, P.J.W. Pouwels, B. Wilken, F. Hanefeld, J. Frahm, Increase of total
creatine in human brain after oral supplementation of creatine monohydrate,
Am. J. Physiol. (2008) 698–704.
Normal reference values for creatine, creatinine, and carnitine are lower in
vegetarians, Clin. Chem. 35 (1989) 1802–1803.
 P. Fox, M. Raichle, M. Mintun, C. Dence, Nonoxidative glucose consumption
during focal physiologic neural activity, Science 241 (1988) 462–464.
 J. Goense, N. Logothetis, Neurophysiology of the BOLD fMRI signal in awake
monkeys, Curr. Biol. 18 (2008) 631–640.
 G.R.J. Gordon, H.B. Choi, R.L. Rungta, G.C.R. Ellis-Davies, B.A. MacVicar, Brain
metabolism dictates the polarity of astrocyte control over arterioles, Nature
456 (2008) 745–749.
 G. Hagberg, G. Zito, F. Patria, J. Sanes, Improved detection of event-related
functional MRI signals using probability functions, Neuroimage 14 (2001)
yses, Neuroimage 21 (2004) 1639–1651.
 Y. Koga, H. Takahashi, D. Oikawa, T. Tachibana, D.M. Denbow, M. Furuse, Brain
creatine functions to attenuate acute stress responses through GABAnergic
system in chicks, J. Neurosci. 132 (2005) 65–71.
 P. Madsen, S. Hasselbalch, L. Hagemann, K. Olsen, J. Bulow, S. Holm,
G. Wildschiodtz, O. Paulson, N. Lassen, Persistent resetting of the cere-
bral oxygen/glucose uptake ratio by brain activation: Evidence obtained
with the Kety–Schmidt technique, J. Cereb. Blood Flow Metab. 15 (1995)
 W. McCall, A. Persky, Pharmacokinetics of creatine, in: G. Salomons, M. Wyss
(Eds.), Creatine and creatine kinase in health and disease, Springer, Dordrecht,
son, N. Draper, Effect of creatine supplementation and sleep deprivation, with
mild exercise, on cognitive and psychomotor performance, mood state, and
plasma concentrations of catecholamines and cortisol, Psychopharmacology
185 (2006) 93–103.
Author's personal copy Download full-text
S.T. Hammett et al. / Neuroscience Letters 479 (2010) 201–205
humans, Am. J. Physiol. Regul. Integr. Comp. Physiol. 292 (2007) R1745–1750.
 C. Rae, A.L. Digney, S.R. McEwan, T.C. Bates, Oral creatine monohydrate
supplementation improves brain performance: a double-blind, placebo-
controlled, cross-over trial, Proc. R. Soc. Lond. Ser. B: Biol. Sci. 270 (2003)
 A. Rauch, G. Rainer, N.K. Logothetis, The effect of a serotonin-induced dissoci-
ation between spiking and perisynaptic activity on BOLD functional MRI, Proc.
Natl. Acad. Sci. U. S. A. 105 (2008).
 J. Raven, J. Raven, J. Court, Manual for Raven’s Progressive Matrices and Vocab-
ulary Scales, Harcourt Assessment, San Antonio, 1998.
 E.S. Rawson, H.R. Lieberman, T.M. Walsh, S.M. Zuber, J.M. Harhart, T.C.
Matthews, Creatine supplementation does not improve cognitive function in
young adults, Physiol. Behav. 95 (2008) 130–134.
role of NMDA receptors, Neurochem. Int. 53 (2008) 33–37.
 A. Viswanathan, R.D. Freeman, Neurometabolic coupling in cerebral cor-
tex reflects synaptic more than spiking activity, Nat. Neurosci. 10 (2007)
 D. Weschler, Adult Intelligence Scale Manual, Psychological Corporation, New