Developmental Cognitive Neuroscience 2S (2012) S152S166

Contents lists available at SciVerse ScienceDirect

Developmental Cognitive Neuroscience

j o u r n a l h o m e p a g e : h t t p : / / w w w.e l s e v i e r.c o m / l o c a t
e/dcn

Weak task-related modulation and stimulus representations during

arithmetic problem solving in children with developmental
dyscalculia
Sarit Ashkenazi a, , Miriam Rosenberg-Lee a , Caitlin Tenison a , Vinod Menon a,b,c,d,
a
Department of Psychiatry and Behavioral Sciences, Stanford University School of Medicine, 401 Quarry Rd, Stanford, CA 94304-5719, USA
b
Department of Neurology and Neurological Sciences, Stanford University School of Medicine, Stanford, CA 94304, USA
c
Neuroscience Program, Stanford University School of Medicine, Stanford, CA 94304, USA
d
Symbolic Systems Program, Stanford University School of Medicine, Stanford, CA 94304, USA

a r t i c l e i n f o a b s t r a c t

Article history:
Developmental dyscalculia (DD) is a disability that impacts math learning and skill acqui-
Received 10 July 2011
sition in school-age children. Here we investigate arithmetic problem solving decits in
Received in revised form
young children with DD using univariate and multivariate analysis of fMRI data. During
28 September 2011
fMRI scanning, 17 children with DD (ages 79, grades 2 and 3) and 17 IQ- and reading
Accepted 28 September 2011
ability-matched typically developing (TD) children performed complex and simple
addition problems which differed only in arithmetic complexity. While the TD group
Keywords:
showed strong modulation of brain responses with increasing arithmetic complexity,
Developmental dyscalculia
children with DD failed to show such modulation. Children with DD showed signicantly
Children
reduced activation compared to TD children in the intraparietal sulcus, superior parietal
fMRI
Intraparietal sulcus
lobule, supramarginal gyrus and bilateral dorsolateral prefrontal cortex in relation to
Arithmetic arithmetic complexity. Crit- ically, multivariate representational similarity revealed that
Prefrontal cortex brain response patterns to complex and simple problems were less differentiated in the
Learning disabilities DD group in bilateral anterior IPS, independent of overall differences in signal level. Taken
together, these results show that children with DD not only under-activate key brain
regions implicated in mathemat- ical cognition, but they also fail to generate distinct
neural responses and representations for different arithmetic problems. Our ndings
provide novel insights into the neural basis of DD.
2011 Elsevier Ltd. All rights reserved.

1. Introduction
Developmental dyscalculia (DD) is a disorder of
Abbreviations: AG, angular gyrus; BA, Brodmann area; DD, develop- numerical and mathematical abilities in children of
mental dyscalculia; fMRI, functional magnetic resonance imaging; IFG, normal intelligence who do not have other cognitive
inferior frontal gyrus; IPS, intraparietal sulcus; ITG, inferior temporal impairments (Butterworth et al., 2011; Rubinsten and
gyrus; LOC, lateral occipital cortex; MFG, middle frontal gyrus; MTG, Henik, 2009; von Aster and Shalev, 2007). Children with
mid- dle temporal gyrus; PFC, prefrontal cortex; PMC, premotor cortex;
DD show poor performance on a broad range of basic
PPC, posterior parietal cortex; ROI, region-of-interest; RSA,
representational similarity analysis; SMG, supramarginal gyrus; SPL, numerical tasks including magnitude judgment
superior parietal lob- ule; TD, typically developing. (Ashkenazi et al., 2009; Geary et al., 2000; Holloway et
Corresponding author. Tel.: +1 650 7360128; fax: +1 650 7211594. al., 2010; Mussolin et al.,
Corresponding author at: Department of Psychiatry and
2010; Piazza et al., 2010; Price et al., 2007) and enumer-
Behavioral Sciences, Stanford University School of Medicine, 401 Quarry
Rd, Stanford, CA 94304-5719, USA. Tel.: +1 650 7360128; fax: +1 650 ation (Geary et al., 1992; Geary and Wiley, 1991; Knootz
7211594. and Berch, 1996; Landerl et al., 2004; Schleifer and Landerl,
E-mail addresses: [email protected] (S. Ashkenazi), 2011). Children with DD also lag behind their typically
[email protected] (V. Menon).
developing (TD) peers in basic arithmetic skills (Geary et
al.,

1878-9293/$ see front matter 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.dcn.2011.09.006
 S.S.Ashkenazi
Ashkenazietetal.al./ Developmental
/ DevelopmentalCognitive
CognitiveNeuroscience
Neuroscience2S2S(2012)
(2012)S152S166
S152S166
15 15

1992; Shalev et al., 2000, 2005). Research over the past left inferior parietal cortex, and multiple frontal lobe
two decades has shown that uent retrieval of arithmetic areas during a number ordering task in a group of sixteen
facts is a key decit in children with DD (Geary, 1993, 810-year-old children with DD (Kucian et al., 2011). Most
1994, of these ndings support the hypothesis that the right IPS
2004; Geary et al., 2007). In TD children, with increased is a common locus of weak numerical representations in
training and practice, typically acquired during the 2nd
and 3rd grades, there is a rapid shift in the distribution of
strategies towards greater use of direct retrieval in solving
simple arithmetic problems (Ashcraft and Fierman, 1982;
Carpenter and Moser, 1984; Geary, 1994; Siegler, 1996).
In contrast, children with DD demonstrate poorer
performance and continue to use less mature strategies
such as nger counting to solve arithmetic problems
(Geary, 1993; Shalev et al., 2000).
Neuroimaging studies have suggested that DD is asso-
ciated with core decits in the representation of
numerical information in the posterior parietal cortex
(PPC). The intraparietal sulcus (IPS) in dorsal aspects of
the PPC has been consistently implicated in numerical
information pro- cessing (Ansari and Dhital, 2006;
Arsalidou and Taylor,
2011; Cohen Kadosh et al., 2008; Delazer et al., 2003;
Menon et al., 2002; Pinel et al., 2001; Rosenberg-Lee et
al.,
2009; Zago et al., 2008) leading to the hypothesis that
impaired numerical representation in the IPS is a core
fea- ture of DD (Butterworth et al., 2011; Rubinsten and
Henik,
2009; Wilson and Dehaene, 2007). However, behavioral
studies have suggested that domain general decits such
as lower visuo-spatial working memory capacity and
weak central executive processes also contribute to
poorer arith- metic skills in children with DD (Geary,
2004; Geary et al.,
2007). Consistent with this view, structural brain imaging
studies in DD have provided evidence for decits not only
in the IPS but also more extended regions within the PPC,
as well as ventral visual stream areas including the
bilateral parahippocampal gyrus and fusiform gyrus
(Rotzer et al.,
2008; Rykhlevskaia et al., 2009).
Only a few studies to date have examined in the
neural basis of DD using functional brain imaging; of
these, ve studies have focused on basic number
processing (Kaufmann et al., 2009a,b; Kucian et al.,
2011; Mussolin et al., 2010; Price et al., 2007) and two
on arithmetic problem solving (Davis et al., 2009;
Kucian et al., 2006). In a group of nine 12-year-old
children with DD, Price and colleagues observed weaker
numerical distance effects in the right IPS, left fusiform
and left medial prefrontal cortex (PFC) during a non-
symbolic number comparison task (Price et al., 2007).
During symbolic number comparison, Mussolin and
colleagues observed weaker numerical dis- tance effects
in the right IPS, left superior parietal lobule (SPL), right
middle frontal gyrus (MFG) and left anterior cingulate
cortex in a group of fteen 10-year-old children with DD
(Mussolin et al., 2010). In contrast, Kaufmann et al.
(2009a,b) found increased bilateral PPC and prefrontal
cor- tex activation during non-symbolic number
comparison task in a group of nine children with DD
from the 2nd to
4th grade (Kaufmann et al., 2009b). Kucian and colleagues
(2011) found reduced activation in the right IPS and SPL,
 S.S.Ashkenazi
Ashkenazietetal.al./ Developmental
/ DevelopmentalCognitive
CognitiveNeuroscience
Neuroscience2S2S(2012)
(2012)S152S166
S152S166
15 15

children with DD, but also point to the involvement of adolescents had greater PPC and prefrontal activity on
multiple, but inconsistently identied, PPC and PFC more difcult arith- metic problems, age-matched
regions. participants with Turner syndrome had comparable
Findings in arithmetic problem solving tasks have levels of activation for both easy and hard problems.
been weaker and even less consistent. Kucian and However, mathematical difculties in children with these
colleagues examined eighteen children with DD from neurodevelopmental disorders are typ- ically accompanied
grades 3 and by profound visuo-spatial and reading disabilities, thereby
6 and found no group differences at the whole-brain limiting their generalizability to pure DD (Ross et al.,
level. However, region-of-interest (ROI) analyses revealed 2002).
reduced brain activity in the right IPS, and inferior and In the current study we used both univariate and
middle frontal gyri. These differences were found for multivariate approaches to examining brain responses
approximate, but not exact, addition problems (Kucian underlying arithmetic processing decits in well-matched
et al., 2006). In contrast, Davis and colleagues (2009) used groups of 79-year-old children with DD and TD chil-
these same tasks and found increased PPC and PFC acti- dren. We focused on differential modulation of neural
vation during both exact and approximation addition in a responses in relation to arithmetic problem complexity.
group of 24 children in the 3rd grade with mathematical Specically, we compared brain responses to two types of
disability (Davis et al., 2009). These studies leave two-operand addition problems: Complex problems, where
unclear the nature of arithmetic processing decits in one operand ranged from 2 to 9, the other from 2 to 5 and
children with DD. De Smedt and colleagues compared Simple problems, where one of the operands was always
brain activation to large and small addition and 1. The two problem types have different arithmetic
subtraction problems in a group of 1012-year-old process- ing demands, but the same sensory and motor
children with typical and low arithmetic achievement (n response selection demands. We examined whether
= 28) (De Smedt et al., 2011). They found that unlike children with DD exhibit modulation of brain responses
typically achieving children, low achieving children in relation to arithmetic problem complexity in the same
failed to show a problem size effect in the right IPS. manner as TD children by using whole-brain analyses, as
Consistent with this result, previous studies of children well as anatom- ically based ROI analyses. We used
and adolescents with neurodevelopmental disor- ders that cytoarchitectonically dened maps of the IPS (Choi et
impair mathematical cognition, such as Fragile-X and al., 2006; Scheperjans et al., 2008) to more precisely
Turner syndromes, have reported aberrant task mod- examine differential brain responses to the Complex and
ulation of multiple PPC and PFC regions during Simple problems in chil- dren with DD (Rosenberg-Lee et
arithmetic problem solving (Kesler et al., 2006; Molko al., 2011a). In addition to voxel-wise analyses, we also
et al., 2003; Rivera et al., 2002). Kesler and colleagues examined multi-voxel acti- vation patterns using
(2006), for exam- ple, found that while TD children and representational similarity analysis
(RSA). RSA is a multivariate approach for investigating the community groups. They had no history of psychiatric and
relationship between stimulus representation and neural neurological illness. All children, except two, were right-
activity (Norman et al., 2006). RSA examines the spatial handed as determined by the Edinburgh handedness test
pattern of multi-voxel brain activity in a specic region (Oldeld, 1971). Intelligence was assessed using the Wech-
of interest across task conditions, independent of overall sler Abbreviated Scale of Intelligence (WASI, Wechsler,
differences in signal level. Here we use RSA to probe the 1999) with an inclusion criterion of full-scale IQ above
similarity of spatial activation patterns between different 80. The Wechsler Individual Achievement Test, Second
types of arithmetic problems, and to provide complemen- Edition (WIAT-II, Wechsler, 2001) was used to determine
tary information about problem representation in children grade specic achievement and dyscalculia status. Seven-
with DD. teen children (11 girls and 6 boys) who scored at or below
Although previous studies of arithmetic (Davis et al., 90 (i.e. the 25 percentile) on the Numerical Operations sub-
2009; Kucian et al., 2006) do not point to a clear prole test of the WIAT-II formed the DD group. The TD group
of activation decits in young children with DD, one consisted of seventeen children (11 girls, 6 boys) individu-
common feature of neurodevelopmental disorders ally matched on age, gender, IQ and reading ability to the
which signi- cantly impair mathematical cognition DD group. TD children were required to score at or above
(Kesler et al., 2006; Molko et al., 2003; Rivera et al., 95 (i.e. the 37 percentile) on the Numerical Operations
2002) is a lack of arith- metic complexity-related subtest (Table 1).
modulation of brain response. We therefore
hypothesized that children with DD would show weaker
arithmetic complexity-related modulation of brain 2.2. Standardized cognitive assessments
response in the PPC, fusiform gyrus and PFC than TD
children. Given inconsistencies in previous studies which 2.2.1. Mathematical abilities
have used only univariate analyses to examine process- Mathematical abilities were assessed using the WIAT-II
ing decits in children with DD, we further hypothesized (Wechsler, 2001). This achievement battery includes
that separate from differences in overall level of nationally standardized measures of K-12 academic skills
activation, they would also show an aberrant pattern of and problem-solving abilities, which are normed by grade
multi-voxel similarity between neural responses to and time of the academic year (Fall, Spring, or Summer).
Complex and Simple arithmetic problems. Together, these The Numerical Operations subtest is a paper-and-pencil
results would provide convergent evidence for weak test that measures number writing and identication,
problem representations in children with DD. rote counting, number production, and simple addition,
subtraction, multiplication, and division problems. For
example, 4 2 = and 37 + 54 (presented vertically) are
2. Methods two problems in the 2nd and 3rd grade range. The Math-
ematical Reasoning subtest is a verbal problem-solving
2.1. Participants test that measures counting, geometric shape identica-
tion, and single- and multi-step word problem-solving
Children in the 2nd and 3rd graders (ages 79) were involving time, money, and measurement with both
recruited from a wide range of schools in the San verbal and visual prompts. The child is required to solve
Francisco Bay Area using mailings to schools, postings at
libraries and

Table 1
Standardized IQ, math and reading achievement and working memory scores for children in the DD and TD groups.

Measure Group p

DD (N = 17) TD (N = 17)

M SD M SD

Males/females 6/11 6/11

Age (months) 98.05 6.72 97.41 6.59 .80

WASI
Verbal 106.64 12.32 112.41 16.48 .26
Performance 107.58 18.2 113.23 12.76 .30
Full Scale 107.7 13.36 114.58 10.73 .11

WIAT-II
Word Reading 108.47 9.23 113.29 10.65 .17
Numerical Operations 85.7 3.49 115.94 15.53 <.01
Mathematical Reasoning 102.47 14.82 115.29 12.76 .01

WMTB-C
Backward Digit Recall 100.17 21.1 99.1 15.89 .87
Digit Recall 109.23 21.32 112.41 15.98 .66
Block Recall 92.41 17.69 103.88 8.65 .02
Counting Recall 93.05 18.67 97.52 20.39 .51

WASI = Wechsler Abbreviated Scales of Intelligence; WIAT-II = Wechsler Individual Achievement Test Second Edition; WMTB-C = Working Memory Test
Battery for Children.
problems with whole numbers, fractions or decimals, 2.3.2. fMRI data acquisition
interpret graphs, identify mathematical patterns, and Images were acquired on a 3T GE Signa scanner (Gen-
solve problems of statistics and probability. For example, eral Electric, Milwaukee, WI) using a custom-built head coil
a dime is presented and the child is states: How many at the Stanford University Lucas Center. Head movement
pennies does it take to equal the value of one dime? A was minimized during the scan by a comfortable custom-
probability problem asks: If you ipped a coin ten times, built restraint. A total of 29 axial slices (4.0 mm thickness,
how many times would the coin be most likely to land 0.5 mm skip) parallel to the AC-PC line and covering the
on heads? whole brain was imaged with a temporal resolution of
2 s using a T2* weighted gradient echo spiral in-out pulse
sequence (Glover and Lai, 1998) with the following param-
2.2.2. Reading abilities eters: TR = 2 s, TE = 30 ms, ip angle = 80 , 1 interleave.
The WIAT-II was also used to assess reading abilities. The
The Word Reading subtest involves reading individual eld of view was 20 cm, and the matrix size was 64 64,
words presented visually to the child, and was used for providing an in-plane spatial resolution of 3.125 mm. To
matching the DD and TD groups on reading ability. reduce blurring and signal loss from eld in homogeneity,
an automated high-order shimming method based on spi-
ral acquisitions was used before acquiring functional MRI
2.2.3. Working memory scans (Kim et al., 2002).
Four subtests of the Working Memory Test Battery for
Children (Pickering and Gathercole, 2001) were used to 2.3.3. fMRI preprocessing
assess working memory abilities. The Central Executive fMRI data were analyzed using SPM8 (http://www.l.
was assessed using the Counting Recall and Backward ion.ucl.ac.uk/spm/). The rst 5 volumes were not analyzed
Digit Recall subtests. Phonological capacity was assessed to allow for T1 equilibration. A linear shim correction
using the Digit Recall subtest and visuo-spatial was applied separately for each slice during reconstruc-
capacity was assessed using the Block Recall test, as tion (Glover and Lai, 1998). ArtRepair software was used
described elsewhere (Meyer et al., 2010). to correct for excessive participant movement (Mazaika
et al., 2007). Images were realigned to correct for move-
ment, smoothed with a 4 mm FWHM Gaussian kernel
2.3. Brain imaging and motion adjusted. Deviant volumes resulting from
sharp movement or spikes in the global signal were
2.3.1. Experimental procedures then interpolated using the two adjacent scans. No more
The functional magnetic resonance imaging (fMRI) than 20% of the volumes were interpolated. Participants
experiment was an event-related design with two task with head movement exceeding 5 mm in any of the
con- ditions: Complex addition (26 trials) and Simple x, y, and z directions were excluded from the study.
addition (26 trials). In the Complex addition task, There were no differences between translational move-
participants were presented with an equation involving ment parameters in the two groups in the x (DD = 0.64
two addends and asked to indicate, via a button box, mm, TD = 0.52 mm; p = .66), y (DD = 1.18 mm, TD = 0.88
whether the answer presented was correct or incorrect mm; p = .54) or z (DD = 1.8 mm, TD = 1.48 mm; p = .45)
(e.g. 3 + 4 = 8). One operand ranged from 2 to 9, the directions. Finally, images were corrected for errors in
other from 2 to 5 (tie problems, such as 5 + 5 = 10, were slice-timing, spatially transformed for registration to
excluded), and answers were correct in half of the trials. standard MNI space, and smoothed again at 4.5 mm
Incorrect answers deviated by 1 or 2 from the correct
FWHM Gaussian kernel. The two step sequence of rst
sum (Ashcraft and Battaglia,
smoothing with a
1978). The Simple addition task was identical except one
4 mm FWHM Gaussian kernel and later with 4.5 mm
of the addends was always 1 (e.g. 3 + 1 = 4). Behavioral
FWHM Gaussian kernel approximates a total smoothing
research in adults suggests that N + 1 addition is solved
of 6 mm, because total smoothing is equivalent to the
by incremental counting (Campbell and Metcalfe, 2007).
square root of the sum of the squares of the individual
Our use of this task was based on pilot studies, which
smoothing steps.
suggested that children are consistently faster on these
problems compared to the Complex addition problems.
2.3.4. Individual subject and group analyses
Critically, because stimuli in the Simple task have the
Task-related brain activation was identied using the
same format as the Complex task, it provides a high-
general linear model implemented in SPM8. In the indi-
level control for sensory and number processing, as well
vidual subject analyses, interpolated volumes agged at
as motor response selection. A verication, rather than
the preprocessing stage were de-weighted. Each trial was
verbal production, for- mat was used in the scanner
modeled using a boxcar function convolved with the
because overt verbal responses can result in signicant
canonical hemodynamic response function and a temporal
head movement in children result- ing in unusable fMRI
dispersion derivative to account for voxel-wise latency
data. Stimuli were displayed for 5 s with an inter-trial
dif- ferences in hemodynamic response. Low-frequency
interval of 500 ms and a jitter period which varied
drifts at each voxel were removed using a high-pass
between 0 and 3500 ms with an average of
lter (0.5 cycles/min). Serial correlations were accounted
1846 ms. During the inter-trial interval and the jitter
for by mod- eling the fMRI time series as a rst-degree
period a xation cross-appeared on the screen. The total
autoregressive process. Voxel-wise t-statistics maps of the
length of the experimental run was 6 min and 22 s.
following con- trasts were generated for each
participant: (1) Complex
additionSimple addition, (2) Complex additionRest, and 3. Results
(3) Simple additionRest. Two analyses were conducted,
one with correct and incorrect trials together, and the 3.1. Neuropsychological measures
sec- ond with correct trials only. To examine the neural
basis of weaker performance in children with DD both Table 1 shows the neuropsychological proles of the DD
correct and incorrect trials were included in the rst and TD groups. The two groups were individually matched
analysis. For the second analysis, only correct trials were on age, gender, IQ and reading ability and so did not differ
included in the contrast images. in any of these measures. All statistical tests were two-
For group analysis, contrast images corresponding to tailed. The groups differed signicantly on the Numerical
the Complex additionSimple addition were analyzed Operation subset of the WIAT-II, with the DD group scor-
using a random effects analysis. Two group-level anal- ing lower than the TD group (t(32) = 7.8, p < .01). The
yses were conducted: (i) one-way t-tests were rst DD group also scored signicantly lower than the TDs on
used to identify areas of signicant activation (Com- the Mathematical Reasoning subtest (t(32) = 2.7, p < .
plex additionSimple addition) and deactivation (Simple 01). DD children had lower scores on the Block Recall
additionComplex addition) for each group separately, subtest of the WMTBC (t(32) = 2.1, p < .05) but not on
and (ii) between group t-tests were used to directly any of the other three working memory subtests: Digit
compare activation between TD children and children Recall, Count- ing Recall and Backward Digit Recall.
with DD. Sig- nicant activation clusters were determined,
after applying a gray matter mask, using a height 3.2. Brain imaging
threshold of p < .01, with family-wise error (FWE)
correction for multiple compar- isons at p < .01 3.2.1. Behavioral differences between TD and DD groups
determined using Monte Carlo simulations (Nichols and 3.2.1.1. Accuracy. Mean accuracy was calculated for each
Hayasaka, 2003). Monte Carlo simulations were participant and entered into a two-way analysis of
implemented in MatLab using methods similar to variance (ANOVA), with Group (DD, TD) as a between-
AFNIs AlphaSim program (Forman et al., 1995; Ward, participants factor and Complexity (Complex, Simple)
2000). In each iteration of the Monte Carlo procedure, a as a within- participants factor. A signicant main effect
3D image with the same resolution and dimensions as the of Complexity was observed (F(1, 32) = 51.3, p < .01),
fMRI scan was randomly generated and smoothed with a participants had higher accuracy in the Simple
6 mm FWHM Gaussian kernel. For consistency with the compared to the Complex task. There was no main
inclu- sive mask used to report the results of the general effect of Group (p = .56) but the interaction between
linear model analysis, a gray matter mask was also Group and Complexity was marginally signicant (F(1, 32)
applied to this image. The maximum cluster size at a = 3.7, p = .06).
given height thresh- old was recorded for each interaction, Differences between accuracy for Complex and Simple
and 10,000 iterations were performed. The distribution of problems was greater in the DD group (67% vs. 88%), com-
maximum cluster size across these 10,000 iterations was pared to the TD group (75% vs. 88%) (t(32) = 1.43, p < .
used to determine the FWE corrected extent threshold. 16) (Fig. 1A).
At a height threshold of p < .01, less than 1% of the
iterations had maximum cluster size greater than 128 3.2.1.2. RT. Median reaction time (RT) on correct trials was
voxels. calculated for each participant and a two-way ANOVA
with factors Group and Complexity was conducted. The
analy- sis revealed a signicant main effect of
2.3.5. ROI analyses Complexity (F(1,
Anatomically based ROI analyses were conducted to 32) = 157.2, p < .01), such that participants were faster for
fur- ther examine the prole of activation differences Simple problems (2080 571 ms) compared to Complex
between the DD and TD groups. We used the observer- problems (3160 365 ms). There was no main effect of
independent cytoarchitectonically dened maps of the Group (F(1, 32) = 0.1, p = .71) or interaction between Group
IPS, which pro- vide an anatomically precise and and Complexity (F(1, 32) = 0.2, p = .64) (Fig. 1B).
consistent basis for examining fMRI responses across
participant groups. The IPS consists of two anterior 3.2.2. Brain activation in DD and TD groups
subdivisions, hIP1 and hIP2 (Choi et al., 2006) and a 3.2.2.1. TD. Complex > Simple. We rst compared brain
posterior subdivision, hIP3 (Scheperjans et al., 2008). responses to Complex and Simple problems separately in
Additional control analyses were performed using the each group. In the TD group, Complex problems elicited
primary visual cortex areas V1 and V2 (Amunts et al., signicantly greater activation in the PPC, specically,
2000). in the bilateral IPS (BA 7), bilateral SPL (BA 7), bilateral
precuneus (BA 7/19), and right supramarginal gyrus
2.3.6. Representational similarity analysis (SMG, BA 40). The TD group also showed greater activity
RSA assesses the similarity of spatial activity patterns in the ventral visual stream, specically, bilateral fusiform
produced by two experimental conditions within an ROI gyrus (BA 37), lateral occipital cortex (LOC), visual
(Kriegeskorte et al., 2008). RSA were conducted by com- cortex (BA
puting the cross-correlation between individuals t-maps 17/18), and in the left inferior temporal gyrus (ITG, BA
for the Complex and Simple tasks within the cytoarchitec- 37) and middle temporal gyrus (MTG, BA 37).
tonically dened ROIs. The correlation coefcients Furthermore, Complex problems elicited signicantly
between conditions were transformed to a normal greater activation in the frontal cortex, specically, in the
distribution using the Fishers r-to-Z transform. bilateral MFG (BA
8/9/10), left inferior frontal gyrus (IFG, BA 47), bilateral
 Fig. 1. (A and B) Behavioral performance in the DD and TD groups. Both groups showed strong differences between addition problem type, with Simple
problems being performed faster and more accurately than Complex problems. There were no group differences in either accuracy or RT. However, the
interaction between problem type and group was marginally signicant (p = .06) indicating weaker performacne for Complex problems in the DD group.

Table 2
Brain areas that showed signicant differences between Complex and Simple arithmetic problems in the DD and TD groups.

Region BA # of voxels Peak Z-score Peak MNI coordinates

x y z

TD group
Complex > Simple
L IPS 7 440 3.22 24 82 48
L Precuneus 19 3.04 20 84 48
L SPL 7 2.79 28 70 50
R Precuneus 7 573 3.22 2 68 56
L Precuneus 7 2.87 4 66 44
R IPS/SPL 7 481 3.15 36 62 58
R SMG 40 3.08 44 50 58
L/R Fusiform gyrus, L 17, 18 5426 4.76 22 94 18
Cerebellum, L ITG 37
L Visual cortex 18 4.66 22 102 4
L LOC 19 3.81 36 92 14
R MFG 9 7331 4.46 50 28 26
R Frontopolar cortex 11 3.98 26 60 4
R MFG 8 3.81 8 32 44
R PMC 6 3.81 54 12 44
L MFG 9 3061 3.77 54 22 32
L MFG 10 3.68 46 52 4
L MFG 47 3.47 36 42 4
L PMC 6 134 3.08 28 2 60
L Insula 13 176 2.90 32 18 2
L Mid-cingulate gyrus 23 402 2.90 2 10 34
L Posterior-cingulate gyrus 31 2.74 2 30 32
R Anterior-cingulate 24 2.58 4 6 30
Brain Stem 182 3.44 0 14 8
Simple > Complex
None

DD group
Complex > Simple
None
Simple > Complex
L vmPFC 11 350 3.01 8 38 6
R vmPFC 10 2.66 6 44 6
R Caudate 143 3.28 2 12 6
R MTG 37 132 2.72 60 60 18

IPS = intraparietal sulcus; ITG = inferior temporal gyrus; LOC = lateral occipital cortex; MFG = middle frontal gyrus; MTG = middle temporal gyrus;
PMC = premotor cortex; SMG = supramarginal gyrus; SPL = superior parietal lobule; vmPFC = ventromedial prefrontal cortex.
Fig. 2. Arithmetic complexity effects in the TD and DD groups. Brain response related to arithmetic complexity obtained by contrasting Complex and
Simple addition problems. Coronal slices show signicant activation (Complex > Simple, red scale) and signicant deactivation (Simple > Complex, blue
scale) in each group. (A) TD children showed greater complexity-related activation in multiple dorsal and ventral stream areas as well as the prefrontal
cortex, including left intraparietal sulcus (IPS), right IPS and superior parietal lobule (SPL), bilateral precuneus, bilateral dorsolateral prefrontal cortex
(middle frontal gyrus), bilateral insula, left inferior and middle temporal gyrus (MTG) and bilateral fusiform gyrus. (B) In contrast, no brain areas
showed greater complexity-related activation in children with DD. Instead, they showed greater activation to Simple problems in left MTG and in the
ventromedial prefrontal cortex (vmPFC).
premotor cortex (PMC, BA 6) and right frontopolar cortex Complex problems compared to Simple problems (Fig. 2B
(BA 11). The TD group also showed greater activity in the and Table 2).
anterior, mid and posterior cingulate (BA 23/24/31), left Simple > Complex. The DD group showed greater acti-
insula (BA 13) (Fig. 2A and Table 2). vation in bilateral ventromedial prefrontal cortex (BA
Simple > Complex. In the TD group, there were no 11/10), caudate nucleus and right MTG (BA 37) (Fig. 2B and
areas where activation was greater for Simple problems Table 2).
com- pared to Complex problems (Fig. 2A and Table 2).
3.2.3. Differences in brain activation between DD and TD
groups
3.2.2.2. DD. Complex > Simple. In the DD group, there Brain responses to Complex and Simple problems
were no brain areas that showed greater activation for were contrasted in each child and then compared
between the
Fig. 3. Comparison of dorsal and ventral visual stream responses in the DD and TD groups. (A) Surface rendering of brain areas that showed reduced
complexity-related responses in the DD, compared to the TD, group. (B) The DD group showed reduced activation in the right superior parietal lobule
(SPL,
44, 60 58), right supramarginal gyrus (SMG, 60, 52, 34), and the right intraparietal sulcus (IPS, 32, 74, 54). (C) The DD group also had reduced
responses
in the lateral occipital (LOC, 54, 74, 6), right fusiform gyrus (50, 62, 20), and left middle temporal gyrus (MTG, 60, 28, 14). Plots of signal levels
in regional peaks shows that the TD group showed incresed activtity on Complex, compared to Simple, problems. In contrast, children with DD had
simlar activation levels for Simple and Complex problems.

two groups. Relative to the TD group, the DD group Simple) as a within-participants factor (Table S1 in the
showed signicantly reduced activation in several right Sup- plementary Materials).
PPC areas including IPS and SPL (BA 7), angular gyrus (AG, In order to examine whether these activation decits
BA 39), SMG (BA 40) and bilateral precuneus (BA 7). The arose from lower visuo-spatial working memory in the
DD group also showed reduced activation in several TD group, we conducted additional analyses using an
ventral and visual areas, including left and right ANOCVA model with block recall scores as a covariate.
fusiform gyrus, LOC (BA This analy- sis replicated our previous ndings without
37), left MTG and ITG and superior temporal gyrus (BA the covariate, except for the mid-cingulate cortex (BA 31)
20/21/22/37) (Fig. 3 and Table 3). The DD group also had which was no longer signicant (Table S2 in the
less activation in bilateral PFC including MFG (BA 8/9/10), Supplementary Materi- als).
right premotor cortex (PMC, BA 6/8), right IFG (BA 47) The above analyses were conducted using both correct
and mid-cingulate cortex (BA 23). Analysis of the prole and incorrect trials, in order to examine the neural basis
of responses in these regions revealed that TD children of weaker performance in children with DD. Because brain
had strong activation on Complex problems but near and cognitive processes for incorrect trials may differ
baseline activation on Simple problems. On the other from those involved in processing correct trials, we
hand, children with DD either had low activation levels conducted additional analyses using only correct trials.
for both prob- lem types or greater activity for Simple The results of this analysis largely replicated ndings
problems (Fig. 4 and Table 3). Similar results were from our pre- vious analysis. At the stringent threshold
obtained when contrast images for Complex-rest and used in this study (p < .01, with FWE correction for
Simple-rest conditions were entered into a two-way multiple comparisons at p < .01) signicant group
ANOVA with Group (DD, TD) as a between-participants differences were found in the left
factor, and Complexity (Complex,
Table 3
Brain areas that showed signicant differences in activation between the DD and TD groups.

Region BA # of voxels Peak Z-score Peak MNI coordinates

x y z

TD > DD
R SPL 7 174 3.19 44 60 58
R IPS 7 2.95 32 74 54
L/R Precuneus 7 417 3.12 2 64 62
R LOC 37 623 3.45 56 72 4
R SMG 40 3.01 60 60 36
R AG 39 2.94 58 62 38
R STG 22 2.83 60 58 18
L Cerebellum 368 3.5 50 64 26
L Fusiform gyrus 37 3.47 58 60 18
L MTG 37 2.99 58 70 2
L MTG 21 222 3.40 58 36 16
L ITG 20 2.86 62 24 18
R Fusiform gyrus 37 169 2.9 50 62 20
ITG 20 52 48 26
R MFG, R PMC 6, 8 505 3.17 56 16 40
R IFG 47 943 3.16 30 26 20
R Superior Temporal gyrus 38 3.04 36 20 38
R Frontopolar cortex 10 3.03 36 52 10
L MFG 9 554 2.99 52 30 30
L MFG 8 2.94 42 36 38
L MFG 10 2.76 38 48 18
L/R Mid-cingulate gyrus 23 134 2.69 0 32 38
Brain Stem 146 3.02 2 16 12

DD > TD
None

AG = angular gyrus; IFG = inferior frontal gyrus; IPS = intraparietal sulcus; ITG = inferior temporal gyrus; LOC = lateral occipital cortex; MFG = middle
frontal gyrus; MTG = middle temporal gyrus; PMC = premotor cortex; SMG = supramarginal gyrus; SPL = superior parietal lobule; STG = superior temporal
gyrus.
fusiform gyrus (BA 37), left IFG (BA 47) and right MFG Simple problems. Anatomically dened subdivisions of the
(BA6/8/9) (Table S3 in the Supplementary Materials). At a IPS were examined separately. We used only responses
threshold of p < .05, with FWE correction for multiple to correct trials and excluded one child with DD who had
com- parisons at p < .05, all the areas reported above were overall accuracy of less than 55%.
found to be signicant, except for the left MTG. Mean similarity between Complex and Simple prob-
lems was computed for each IPS subdivision, and for
3.2.4. Group differences in cytoarchitectonically dened each participant, and normalized using Fishers r-to-Z
IPS ROIs transform. The resulting Z-scores were subjected to a
To further investigate potential PPC decits in chil- three- way ANOVA with Group as a between-participants
dren with DD, we examined task-related responses in factor, and Laterality (Left, Right), and Subdivision (hIP1,
anatomically dened ROIs in the IPS. Mean t-scores for hIP2, hIP3) as within-participants factors. A signicant
each participant were calculated for each problem type, interac- tion between Group and Subdivision was
in each subdivision of the IPS, and entered into a four- observed (F(2,
way ANOVA with between-participants factor Group (DD, 62) = 6.1, p < .01), with the DD group having higher repre-
TD), and within-subject factors of Laterality (Left, Right), sentational similarity (r = .89; Z = 1.41) than the TD group
Subdivision (hIP1, hIP2, hIP3), and Complexity (Complex, (r = .81; Z = 1.09) in hIP2 (F(1, 31) = 4.5, p < .05)
Simple). There was a signicant interaction between between Complex and Simple problems. No group
Group and Complexity (F(1, 32) = 5.2, p < .05). In differences were observed in hIP1 (F(1, 31) = 0.7, p = .
children with DD, activation was comparable for 66) and hIP3 (F(1,
Complex and Simple problems in all the IPS subdivisions 31) = 1.8, p = .18). There were no other signicant interac-
whereas the TD group showed greater activation to the tions or main effects (Fig. 6).
Complex problems in all the subdivisions (Fig. 5). There
was no main effect of Group or other interactions with 4. Discussion
Group. To examine the specicity of our ndings
additional control analyses were performed using the In this study we examined brain activation and neu-
primary visual cortex areas V1 and V2. There were no ral representations during arithmetic problem solving in
group differences or interactions with group in V1 or V2 (p children with DD using both univariate and multivariate
> .22) (Supplementary Materials). approaches. Importantly, IQ, reading ability and working
memory capacity of the DD group were all within the nor-
3.2.5. Representational similarity mal range; visuo-spatial working memory capacity was
RSA was used to examine the similarity between the only general cognitive measure where the TD group
multi-voxel activation patterns elicited by Complex and had greater abilities. This well-matched group of children
therefore allowed us to investigate the neural correlates of
Fig. 4. Comparison of prefrontal cortex responses in the DD and TD groups. (A) Surface rendering of prefrotnal areas that showed reduced complexity-
related brain responses in the DD, compared to the TD, group. (B) The DD group showed lower activation in the left middle frontal gyrus (MFG, 52, 30,
30), right premotor cortex (PMC, 56, 16, 40) and adjoining MFG, and in the right inferior frontal gyrus (IFG, 30, 28, 20) and adjoining MFG. Plots of signal
levels in regional peaks shows that the TD group had increased activtity on the Complex, compared to Simple, problems. In contrast, children with DD
had similar activation levels for Simple and Complex problems.
DD without comorbid cognitive decits. Using this sample, were signicantly faster on Simple, compared to Com-
we examined behavioral and brain responses to Complex plex, arithmetic problems. There were no group differences
and Simple problems which differed in arithmetic com- in RT, but for accuracy the interaction between problem
plexity but not in sensory processing, decision making type and group was marginally signicant, such that chil-
or motor responses. Children with DD, like TD children, dren with DD performed worse on Complex problems. In

Fig. 5. ROI analysis in bilateral cytoarchitectonically dened subdivisions of the IPS. In all three subdivisions (hIP2, hIP1, hIP3) of the bilateral IPS, TD
children showed greater activation for Complex, compared to Simple, problems. In contrast, children with DD showed similar activation levels for
Complex and Simple problems in all three subdivisions of the left and right IPS; *p < .05.
Fig. 6. Representational similarity analysis in cytoarchitectonically dened subdivisions of the IPS. Children with DD showed greater similarity of multi-
voxel brain responses to Complex and Simple problems in the anterior most subdivision (hIP2: 44, 40, 48) of the left and right IPS; *p < .05.
spite of modest differences in performance on two- differences were all localized to the right PPC. The IPS,
operand addition problems during the fMRI session, SPL and SMG are known to be involved in different cog-
children with DD showed signicant decits in nitive processes during numerical problem solving (Cohen
activation in multiple brain systems that are known to Kadosh et al., 2008; Dehaene et al., 2003; Menon, 2010).
be important for math- ematical cognition. Furthermore, While the IPS is sensitive to quantity representation and
independent of overall decits in brain activation, RSA the semantic aspects of number processing (Ansari and
showed poor differentia- tion between Complex and Dhital, 2006; Cohen Kadosh et al., 2008; Delazer et al.,
Simple problems in children with DD. Thus, both 2003; Menon et al., 2002; Pinel et al., 2001; Rosenberg-
univariate and multivariate analy- ses revealed that Lee et al., 2009; Zago et al., 2008), adjoining areas of the
children with DD have weak modulation and problem SPL are thought to be involved in covert visuo-spatial
representation associated with arithmetic complexity. attention processes (Andres et al., 2011; Wu et al., 2009)
and attentional decits have been reported in individu-
4.1. Children with DD show weaker activation in als with DD (Ashkenazi and Henik, 2010a,b; Ashkenazi
multiple cortical regions et al., 2009). The right IPS, SPL and SMG are all also known
to be involved in visuo-spatial working memory in chil-
Children with DD had weaker activity in multiple cor- dren (Kwon et al., 2002). Consistent with this
tical regions. The loci of decits in the DD group can be observation, cognitive testing in our DD sample
categorized into three major brain systems that are revealed signicant decits in visuo-spatial working
impor- tant for numerical cognition the dorsal visual memory capacity even though the central executive,
stream, the ventral visual stream and the PFC (Arsalidou phonological capacity and other cognitive abilities were
and Taylor, well matched between the groups.
2011; Rivera et al., 2005; Rosenberg-Lee et al., 2011a). In In the ventral visual stream we found differences in the
the dorsal visual stream, DD children had decits in both right and left fusiform gyrus and the LOC. Structural
lateral PPC, including the right IPS, SPL and AG, and decits have been found in the right fusiform gyrus in
medial PPC, including the precuneus. In the ventral visual children with DD (Rykhlevskaia et al., 2009). Moreover,
stream decits were observed bilaterally in the LOC and white matter pro- jections linking the right fusiform gyrus
fusiform gyri, in the left MTG and ITG. The distributed to the right PPC are also weaker in children with DD
nature of these task-related reductions suggests that (Rykhlevskaia et al., 2009). Taken together, these results
impairments during arithmetic problem solving in the DD suggest that weak functional interactions along this and
group extend beyond the IPS. These decits were other ventral-dorsal pathways may contribute to DD.
observed for both correct and incorrect trials taken Children with DD also showed prominent decits in
together and when we restricted the analysis to just three distinct clusters within the PFC. These three clus-
the correct trials, providing multiple converging ters were localized to the right PMC, the right dorsolateral
evidence for aberrant processing of arithmetic problems PFC extending to the ventrolateral PFC, and the left dor-
in children with DD. solateral PFC. Reduced right dorsolateral PFC activation in
Although the IPS has been the focus of most children with DD has been reported during both symbolic
neuroimag- ing studies of DD our ndings point to number comparison (Mussolin et al., 2010) and approxi-
multiple foci of PPC decits in young children with DD. mate arithmetic tasks (Kucian et al., 2006), but not during
Specically, we found differences not only in the IPS, but non-symbolic magnitude comparison (Price et al., 2007).
also the SPL, AG and the SMG. Consistent with stronger
gray matter decits in the right hemisphere
(Rykhlevskaia et al., 2009), functional
Our ndings suggest that multiple dorsolateral and ven- observed in all three systems highlighted above: the
trolateral PFC regions in both hemispheres contribute to dorsal and ventral visual streams and the PFC (Figs. 3 and
arithmetic processing decits in children with DD. Cen- 4). For Complex problems, children with DD also showed
tral executive and working memory decits are known weaker activation than TD children. This pattern of brain
to be important factors contributing to poor problem response suggests that children with DD overcompensate
solving and fact retrieval in children with DD (Geary, when solv- ing Simple problems and fail to modulate
2004; Geary et al., 2007; Rotzer et al., 2009). Although, neural activity in response to increasing arithmetic
children in the DD group in our study were well matched complexity.
on a number of cognitive measures, including central
executive measures, they showed signicant decits in 4.3. IPS as a locus of weak neural representations in
visuo-spatial working memory. However, additional children with DD
analysis showed that decits in PFC activation were
prominent even after controlling for differences in visuo- The IPS has been an area of particular interest in DD,
spatial working memory. These results suggest that since it plays an important role in both basic number
children with DD do not adequately engage PFC regions processing and arithmetic problem solving (Ansari and
known to be important for domain general functions, Dhital, 2006; Cohen Kadosh et al., 2008; Dehaene et al.,
such as working memory, even when their cog- nitive 2003; Delazer et al., 2003; Menon et al., 2002; Pinel et
and intellectual capacities are not decient (Kwon et al., al.,
2002; Rivera et al., 2005). 2001; Rosenberg-Lee et al., 2009; Zago et al., 2008). Neu-
Finally, a surprising nding of this study were the pro- roanatomical studies have found that children with DD
cessing decits in the left MTG. No such decits were show reduced gray matter volume in the IPS and adjoin-
found in the homologous right hemisphere regions. The ing PPC (Rotzer et al., 2008; Rykhlevskaia et al., 2009).
MTG has generally not been reported as a source of However, as previously noted, functional imaging studies
functional or structural decits in DD. Yet there is a of arithmetic processing in children with DD have yielded
growing recogni- tion of the role of the left MTG in the mixed ndings, with some studies reporting increased
verbal retrieval of semantic information (Booth et al., and others decreased IPS responses in comparison to the
2002; Fiebach et al., TD group. To resolve this discrepancy we conducted both
2002). In the context of numerical cogntion, it is note- univariate and multivariate RSA analyses to examine
worthy, that the left MTG shows both numerical distance task- related modulation of responses in the IPS.
(Pinel et al., 2001) and developmental changes in size At the whole-brain level, signicant differences were
con- gruity effects (Wood et al., 2009). Recently, it was observed in the right IPS and multiple adjoining regions
reported that a left MTG region identied by comparing of the right PPC. Convergent results were obtained from
words to symbols strings also displayed greater a more rened neuroanatomical analysis of IPS responses
activation for mul- tiplication problems than subtraction using the three cytoarchitectonically dened subdivisions
(Prado et al., 2011). The authors attributed these (hIP2, hIP1 and hIP3) that span its anteriorposterior axis
differences to the greater use of verbally mediated (Caspers et al., 2006; Choi et al., 2006; Rosenberg-Lee et
retrieval for multiplication than sub- traction. These left al.,
MTG regions implicated in numerical and arithmetic 2011b; Wu et al., 2009). Children with DD showed signi-
cogntion tend to be more dorsal to the area identied in cant task-related differences in the right IPS; additionally,
the current study. However, developmental increases in we found that all three subdivisions of both the right and
activitation for an arithmetic task were found in regions left IPS showed a signicant group by complexity interac-
spanning the left inferior and middle tempo- ral gyrus tion such that children with DD failed to up-regulate
(Rivera et al., 2005). Further research is needed to brain responses to the Complex problems. In contrast, in
examine whether the left lateralized temporal decits each of these regions, TD children showed the expected
observed in children with DD are directly related to poor pattern of signicant complexity-related modulation.
arithmetic fact retrieval in this population. These three IPS regions are not only cytoarchitectonically
distinct, but they also differ in their patterns of intrinsic
4.2. Children with DD show weak differentiation functional con- nectivity and structural connectivity with
between problem types other brain areas (Uddin et al., 2010), pointing to multiple
circuits and func- tional processes that are dysfunctional
A striking novel nding of our current study is that in children with DD.
chil- dren with DD show less differentiated neural To further examine the underlying neural represen-
responses to Complex and Simple problems. In the PPC tations, we used RSA to compare spatial patterns of IPS
and the PFC, for example, TD children showed the activity to Complex and Simple problems in the two
expected pattern of increases in activation for Complex groups. We found that children with DD have less
problems compared to Simple problems. While children distinct neural representations for the two problem
with DD showed signi- cant activation in the same types than TD chil- dren in both the left and right IPS.
fronto-parietal regions as TD children, they did not Interestingly, this effect was localized to the anterior
demonstrate greater activation to Complex, compared to most subdivision of the IPS, specically hIP2. Connectivity
the Simple problems. Furthermore, in contrast to TD analysis has shown that hIP2 has strong connections with
children, who showed greater activa- tion for Complex the dorsolateral PFC, premotor cortex and the frontal eye
versus Simple problems, children with DD showed eld (Uddin et al., 2010), sug- gesting that processing
greater activation for Simple versus Complex problems in decits in this region may result in weaker task-
several regions. This reversed pattern of the expected dependent top-down inuences in children with DD.
complexity-related modulation of responses was
Consistent with this view, PFC responses were
signicantly undifferentiated in these children. Because
RSA examines spatial correlations in brain activity, these of cognitive and performance differences. We found that
results are independent of overall differences in magni- children with DD not only exhibit aberrant activity in key
tude of signal level and provide new information about brain regions implicated in mathematical cognition but
the lack of distinct problem representations in children they also fail to modulate task-relevant neural responses
with DD. Taken together, both univariate and multivariate and representations of distinct arithmetic problems. Both
analyses suggest that weak problem representations and univariate and multivariate analysis suggest that the
task-related modulation of neural activity in the IPS is an IPS is a major locus of poor problem representation and
important hallmark of DD. arithmetic complexity-related modulation in children
with DD. Our study offers novel insights into the neural
4.4. Limitations correlates of DD by providing for the rst time, evidence
of representational differences in key brain areas, and
Multiple naming conventions and diagnostic criteria highlights the utility of multivariate approaches to under-
have been used by researchers in recent years to charac- standing aberrant stimulus representations underlying
terize and classify children with mathematical disability complex learning disabilities. Our approach is likely to be
(Rubinsten and Henik, 2009), including DD (Ashkenazi applicable to other developmental learning disorders and
et al., 2009; Kaufmann et al., 2009b; Rykhlevskaia et al., has important practical implications for members of the
2009), mathematical learning disabilities (Geary et al., educational neuroscience community.
2007; Mazzocco et al., 2011; Rousselle and Nol, 2008)
with cutoffs at 46th percentile and below (Geary et al.,
Acknowledgments
1992), 25th percentile and below (Davis et al., 2009), 15th
percentile and below (Rousselle and Nol, 2008), 10th
This research was supported by grants from the
percentile and below (Mazzocco et al., 2011), and 7th per-
National Institutes of Health (HD047520, HD059205,
centile and below (Price et al., 2007). In the present study,
MH084164) and the National Science Foundation
we used a 25th percentile cutoff for mathematical disabil-
(BCS/DRL 0750340).
ity, a more liberal criterion than those recommended in
previous behavioral studies (Mazzocco et al., 2011). An
important issue here is that at the low end of math abil- Appendix A. Supplementary data
ities, high rates of co-morbidity with dyslexia or attention
decit/hyperactivity disorder have been reported Supplementary data associated with this article can be
(Hanich, found, in the online version, at doi:10.1016/j.dcn.2011.
2001; Rubinsten and Henik, 2009; von Aster and Shalev, 09.006.
2007). Using a stricter denition of DD therefore runs
the risk of increasing co-morbid disabilities, and failing References
to match on non-numerical abilities, such as IQ, reading
and working memory (Rubinsten and Henik, 2009). Impor- Amunts, K., Malikovic, A., Mohlberg, H., Schormann, T., Zilles, K., 2000.
Brodmanns areas 17 and 18 brought into stereotaxic space-where
tantly, in the present study, we used an extensive battery and how variable? Neuroimage 11, 6684.
of neuropsychological measures to not only demonstrate Andres, M., Pelgrims, B., Michaux, N., Olivier, E., Pesenti, M., 2011. Role
that the two groups are matched on IQ, reading abilities, of distinct parietal areas in arithmetic: an fMRI-guided TMS study.
Neuroimage 54, 30483056.
attention and working memory, but also to ensure that Ansari, D., Dhital, B., 2006. Age-related changes in the activation of the
all measures of non-numerical abilities are in the normal intraparietal sulcus during nonsymbolic magnitude processing: an
range, and therefore unimpaired, in the DD group. The event-related functional magnetic resonance imaging study. J. Cogn.
Neurosci. 18, 18201828.
tradeoff here is clearly that as one imposes stricter
Arsalidou, M., Taylor, M.J., 2011. Is 2 + 2 = 4? Meta-analyses of brain areas
selection criteria based on numerical measures, needed for numbers and calculations. Neuroimage 54, 23822393.
comorbid disabil- ities in other domains, such as reading Ashcraft, M.H., Battaglia, J., 1978. Cognitive arithmetic evidence for
disabilities, also become more prominent a situation we retrieval and decision-processes in mental addition. J. Exp. Psychol.:
Hum. Learn. Memory 4, 527538.
tried to avoid. Nevertheless, given the current lack of Ashcraft, M.H., Fierman, B.A., 1982. Mental addition in third, fourth, and
consensus in den- ing DD, the interpretation of the sixth graders. J. Exp. Child Psychol. 33, 18.
present results and other related neuroimaging studies of Ashkenazi, S., Henik, A., 2010a. Attentional networks in developmental
dyscalculia. Behav. Brain Funct. 6, 112.
children with math dis- abilities, should still be Ashkenazi, S., Henik, A., 2010b. A disassociation between physical and
tempered by the disparate nature of the selection and mental number bisection in developmental dyscalculia. Neuropsy-
exclusion criteria employed. chologia 48, 28612868.
Ashkenazi, S., Rubinsten, O., Henik, A., 2009. Attention, automaticity, and
developmental dyscalculia. Neuropsychology 23, 535540.
5. Conclusions Booth, J.R., Burman, D.D., Meyer, J.R., Gitelman, D.R., Parrish, T.B.,
Mesulam, M.M., 2002. Modality independence of word
comprehension. Hum. Brain Mapp. 16, 251261.
DD is a disability that impacts mathematical learning
Butterworth, B., Varma, S., Laurillard, D., 2011. Dyscalculia: from brain to
and skill acquisition in school-age children, and arithmetic education. Science 332, 10491053.
fact retrieval is one of the most pronounced decits in Campbell, J.I.D., Metcalfe, A.W.S., 2007. Arithmetic rules and numeral for-
mat. J. Cogn. Psychol. 19, 335355.
chil- dren with DD. Our study overcomes limitations of
Carpenter, T.P., Moser, J.M., 1984. The acquisition of addition and sub-
previous neuroimaging studies by employing a narrow traction concepts in grades one through three. J. Res. Math. Educ. 15,
age range, using a standardized test to identify children 179202.
with DD, and excluding children with comorbid cognitive Caspers, S., Geyer, S., Schleicher, A., Mohlberg, H., Amunts, K., Zilles, K.,
2006. The human inferior parietal cortex: cytoarchitectonic parcella-
disabilities. With this sample, we were able to assess tion and interindividual variability. Neuroimage 33, 430448.
univariate and multivariate differences in brain
response independent
Choi, H.J., Zilles, K., Mohlberg, H., Schleicher, A., Fink, G.R., Armstrong, E., Kwon, H., Reiss, A.L., Menon, V., 2002. Neural basis of protracted devel-
Amunts, K., 2006. Cytoarchitectonic identication and probabilistic opmental changes in visuo-spatial working memory. Proc. Natl. Acad.
mapping of two distinct areas within the anterior ventral bank of Sci. U.S.A. 99, 1333613341.
the human intraparietal sulcus. J. Comp. Neurol. 495, 5369. Landerl, K., Bevan, A., Butterworth, B., 2004. Developmental dyscalcu-
Cohen Kadosh, R., Lammertyn, J., Izard, V., 2008. Are numbers special? An lia and basic numerical capacities: a study of 89-year-old students.
overview of chronometric, neuroimaging, developmental and com- Cognition 93, 99125.
parative studies of magnitude representation. Prog. Neurobiol. 84, Mazaika, P., Whiteld-Gabrieli, S., Reiss, A., Glover, G., 2007. Artifact
132147. repair for fMRI data from high motion clinical subjects. In: 13th
Davis, N., Cannistraci, C.J., Rogers, B.P., Gatenby, J.C., Fuchs, L.S., Anderson, Annual Meet- ing of the Organization for Human Brain Mapping,
A.W., Gore, J.C., 2009. Aberrant functional activation in school age Chicago, IL.
chil- dren at-risk for mathematical disability: a functional imaging Mazzocco, M.M.M., Feigenson, L., Halberda, J., 2011. Impaired acuity of
study of simple arithmetic skill. Neuropsychologia 47, 24702479. the approximate number system underlies mathematical learning
Dehaene, S., Piazza, M., Pinel, P., Cohen, L., 2003. Three parietal circuits disability (Dyscalculia). Child Dev. 82, 12241237.
for number processing. Cogn. Neuropsychol. 20, 487506. Menon, V., 2010. Developmental cognitive neuroscience of arith-
Delazer, M., Domahs, F., Bartha, L., Brenneis, C., Lochy, A., Trieb, T., Benke, metic: implications for learning and education. ZDM 42,
T., 2003. Learning complex arithmetican fMRI study. Brain Res. Cogn. 515525.
Brain Res. 18, 7688. Menon, V., Mackenzie, K., Rivera, S.M., Reiss, A.L., 2002. Prefrontal cortex
De Smedt, B., Holloway, I.D., Ansari, D., 2011. Effects of problem size and involvement in processing incorrect arithmetic equations: evidence
arithmetic operation on brain activation during calculation in children from event-related fMRI. Hum. Brain Mapp. 16, 119130.
with varying levels of arithmetical uency. Neuroimage 57, 771781. Meyer, M.L., Salimpoor, V.N., Wu, S.S., Geary, D.C., Menon, V., 2010. Dif-
Fiebach, C.J., Friederici, A.D., Muller, K., von Cramon, D.Y., 2002. fMRI evi- ferential contribution of specic working memory components to
dence for dual routes to the mental lexicon in visual word recognition. mathematics achievement in 2nd and 3rd graders. Learn. Individual
J. Cogn. Neurosci. 14, 1123. Differ. 20, 101109.
Forman, S.D., Cohen, J.D., Fitzgerald, M., Eddy, W.F., Mintun, M.A., Noll, Molko, N., Cachia, A., Riviere, D., Mangin, J.F., Bruandet, M., Le Bihan, D.,
D.C., 1995. Improved assessment of signicant activation in Cohen, L., Dehaene, S., 2003. Functional and structural alterations of
functional magnetic resonance imaging (fMRI): use of a cluster-size the intraparietal sulcus in a developmental dyscalculia of genetic ori-
threshold. Magn. Reson. Med. 33, 636647. gin. Neuron 40, 847858.
Geary, D.C., 1993. Mathematical disabilities: cognitive, neuropsychologi- Mussolin, C., De Volder, A., Grandin, C., Schlogel, X., Nassogne, M.C.,
cal, and genetic components. Psychol. Bull. 114, 345362. Nol, M.-P., 2010. Neural correlates of symbolic number com-
Geary, D.C., 1994. Childrens Mathematical Development: Research and parison in developmental dyscalculia. J. Cogn. Neurosci. 22, 860
Practical Applications. American Psychological Association, Washing- 874.
ton, DC, US. Nichols, T., Hayasaka, S., 2003. Controlling the familywise error rate in
Geary, D.C., 2004. Mathematics and learning disabilities. J. Learn. Disabil. functional neuroimaging: a comparative review. Stat. Methods Med.
37, 415. Res. 12, 419446.
Geary, D.C., Bow-Thomas, C.C., Yao, Y., 1992. Counting knowledge and Norman, K.A., Polyn, S.M., Detre, G.J., Haxby, J.V., 2006. Beyond mind-
skill in cognitive addition: a comparison of normal and reading: multi-voxel pattern analysis of fMRI data. Trends Cogn. Sci.
mathematically disabled children. J. Exp. Child Psychol. 54, 372391. 10, 424430.
Geary, D.C., Hamson, C.O., Hoard, M.K., 2000. Numerical and arithmeti- Oldeld, R.C., 1971. The assessment and analysis of handedness: the
cal cognition: a longitudinal study of process and concept decits in Edin- burgh inventory. Neuropsychologia 9, 97113.
children with learning disability. J. Exp. Child Psychol. 77, 236263. Piazza, M., Facoetti, A., Trussardi, A.N., Berteletti, I., Conte, S., Lucangeli,
Geary, D.C., Hoard, M.K., Byrd-Craven, J., Nugent, L., Numtee, C., 2007. D., Dehaene, S., Zorzi, M., 2010. Developmental trajectory of num-
Cog- nitive mechanisms underlying achievement decits in children ber acuity reveals a severe impairment in developmental
with mathematical learning disability. Child Dev. 78, 13431359. dyscalculia. Cognition 116, 3341.
Geary, D.C., Wiley, J.G., 1991. Cognitive addition: strategy choice and Pickering, S., Gathercole, S., 2001. Working Memory Test Battery for Chil-
speed-of-processing differences in young and elderly adults. dren. The Psychological Corporation, London.
Psychol. Aging 6, 474483. Pinel, P., Dehaene, S., Riviere, D., LeBihan, D., 2001. Modulation of pari-
Glover, G.H., Lai, S., 1998. Self-navigated spiral fMRI: interleaved versus etal activation by semantic distance in a number comparison task.
single-shot. Magn. Reson. Med. 39, 361368. Neuroimage 14, 10131026.
Hanich, L., 2001. Performance across different areas of mathematical Prado, J., Mutreja, R., Zhang, H., Mehta, R., Desroches, A.S., Minas, J.E.,
cognition in children with learning difculties. J. Educ. Psychol. 93, Booth, J.R., 2011. Distinct representations of subtraction and multi-
615626. plication in the neural systems for numerosity and language. Hum.
Holloway, I.D., Price, G.R., Ansari, D., 2010. Common and segregated Brain Mapp. 32, 19321947.
neural pathways for the processing of symbolic and nonsymbolic Price, G.R., Holloway, I., Rasanen, P., Vesterinen, M., Ansari, D., 2007.
numerical magnitude: an fMRI study. Neuroimage 49, 10061017. Impaired parietal magnitude processing in developmental dyscalcu-
Kaufmann, L., Vogel, S.E., Starke, M., Kremser, C., Schocke, M., 2009a. lia. Curr. Biol. 17, R1042R1043.
Numerical and non-numerical ordinality processing in children with Rivera, S.M., Menon, V., White, C.D., Glaser, B., Reiss, A.L., 2002. Functional
and without developmental dyscalculia: evidence from fMRI. Cogn. brain activation during arithmetic processing in females with fragile
Dev. 24, 486494. X syndrome is related to FMR1 protein expression. Hum. Brain
Kaufmann, L., Vogel, S.E., Starke, M., Kremser, C., Schocke, M., Wood, G., Mapp.
2009b. Developmental dyscalculia: compensatory mechanisms in 16, 206218.
left intraparietal regions in response to nonsymbolic magnitudes. Rivera, S.M., Reiss, A.L., Eckert, M.A., Menon, V., 2005. Developmen-
Behav. Brain Funct. 5, 35. tal changes in mental arithmetic: evidence for increased functional
Kesler, S.R., Menon, V., Reiss, A.L., 2006. Neuro-functional differences specialization in the left inferior parietal cortex. Cereb. Cortex 15,
asso- ciated with arithmetic processing in Turner syndrome. Cereb. 17791790.
Cortex Rosenberg-Lee, M., Barth, M., Menon, V., 2011a. What difference does a
16, 849856. year of schooling make? Maturation of brain response and connectiv-
Kim, D.H., Adalsteinsson, E., Glover, G.H., Spielman, D.M., 2002. Regular- ity between 2nd and 3rd grades during arithmetic problem solving.
ized higher-order in vivo shimming. Magn. Reson. Med. 48, 715722. Neuroimage 57, 796808.
Knootz, K.L., Berch, D.B., 1996. Identifying simple numerical stimuli: Rosenberg-Lee, M., Chang, T.T., Young, C.B., Wu, S., Menon, V., 2011b. Func-
processing inefciencies exhibited by arithmetic learning disabled tional dissociations between four basic arithmetic operations in the
children. In: Mathematical Cognition, p. 23. human posterior parietal cortex: a cytoarchitectonic mapping study.
Kriegeskorte, N., Mur, M., Bandettini, P., 2008. Representational similarity Neuropsychologia 49, 25922608.
analysis connecting the branches of systems neuroscience. Front. Rosenberg-Lee, M., Lovett, M.C., Anderson, J.R., 2009. Neural correlates
Syst. Neurosci. 2, 4. of arithmetic calculation strategies. Cogn. Affect. Behav. Neurosci. 9,
Kucian, K., Grond, U., Rotzer, S., Henzi, B., Schonmann, C., Plangger, F., 270285.
Galli, M., Martin, E., von Aster, M., 2011. Mental number line Ross, J.L., Stefanatos, G.A., Kushner, H., Zinn, A., Bondy, C., Roeltgen, D.,
training in children with developmental dyscalculia. Neuroimage 57, 2002. Persistent cognitive decits in adult women with Turner syn-
782795. drome. Neurology 58, 218225.
Kucian, K., Loenneker, T., Dietrich, T., Dosch, M., Martin, E., von Aster, Rotzer, S., Kucian, K., Martin, E., von Aster, M., Klaver, P., Loenneker, T.,
M., 2006. Impaired neural networks for approximate calculation in 2008. Optimized voxel-based morphometry in children with devel-
dyscalculic children: a functional MRI study. Behav. Brain Funct. 2, opmental dyscalculia. Neuroimage 39, 417422.
31.
Rotzer, S., Loenneker, T., Kucian, K., Martin, E., Klaver, P., von angular gyrus and intraparietal sulcus: evidence from functional and
Aster, M., 2009. Dysfunctional neural network of spatial working structural connectivity. Cereb. Cortex 20, 26362646.
memory contributes to developmental dyscalculia. Neuropsychologia von Aster, M., Shalev, R., 2007. Number development and developmental
47, 28592865. dyscalculia. Dev. Med. Child Neurol. 49, 868873.
Rousselle, L., Nol, M.-P., 2008. Mental arithmetic in children with math- Ward, B.D., 2000. Simultaneous Inference for FMRI Data. AFNI 3dDecon-
ematics learning disabilities. J. Learn. Disabilities 41, 498513. volve Documentation. Medical College of Wisconsin.
Rubinsten, O., Henik, A., 2009. Developmental dyscalculia: heterogeneity Wechsler, D., 1999. Wechsler Abbreviated Scale of Intelligence. The Psy-
might not mean different mechanisms. Trends Cogn. Sci. 13, 9299. chological Corporation, SanAntonio, TX.
Rykhlevskaia, E., Uddin, L.Q., Kondos, L., Menon, V., 2009. Neuroanatomi- Wechsler, D., 2001. The Wechsler Individual Achievement Test Second
cal correlates of developmental dyscalculia: combined evidence from Edition (WIAT-II). The Psychological Corporation.
morphometry and tractography. Front. Hum. Neurosci. 3, 51. Wilson, A.J., Dehaene, S., 2007. Number sense and developmental dyscal-
Scheperjans, F., Eickhoff, S.B., Homke, L., Mohlberg, H., Hermann, K., culia. In: Coch, D., Dawson, G., Fischer, K.W. (Eds.), Human Behavior,
Amunts, K., Zilles, K., 2008. Probabilistic maps, morphometry, and Learning, and the Developing Brain: Atypical Development.
variability of cytoarchitectonic areas in the human superior parietal Guilford Press, New York, NY, US, pp. 212238.
cortex. Cereb. Cortex 18, 21412157. Wood, G., Ischebeck, A., Koppelstaetter, F., Gotwald, T., Kaufmann, L., 2009.
Schleifer, P., Landerl, K., 2011. Subitizing and counting in typical and Developmental trajectories of magnitude processing and interference
atyp- ical development. Dev. Sci. 14, 280291. control: an FMRI study. Cereb. Cortex 19, 27552765.
Shalev, R., Auerbach, J., Manor, O., Gross-Tsur, V., 2000. Developmental Wu, S.S., Chang, T.T., Majid, A., Caspers, S., Eickhoff, S.B., Menon, V., 2009.
dyscalculia: prevalence and prognosis. Eur. Child Adolesc. Psychiatry Functional heterogeneity of inferior parietal cortex during mathe-
9, S58S64. matical cognition assessed with cytoarchitectonic probability maps.
Shalev, R., Manor, O., Gross-Tsur, V., 2005. Developmental dyscalculia: a Cereb. Cortex 19, 29302945.
prospective six-year follow-up. Dev. Med. Child Neurol. 47, 121125. Zago, L., Petit, L., Turbelin, M.-R., Andersson, F., Vigneau, M., Tzourio-
Siegler, R.S., 1996. Emerging Minds: The Process of Change in Childrens Mazoyer, N., 2008. How verbal and spatial manipulation networks
Thinking. Oxford University Press, New York, NY, US. contribute to calculation: an fMRI study. Neuropsychologia 46,
Uddin, L.Q., Supekar, K., Amin, H., Rykhlevskaia, E., Nguyen, D.A., Gre- 24032414.
icius, M.D., Menon, V., 2010. Dissociable connectivity within human