Trends in Neuroscience and Education 2 (2013) 85–93

Contents lists available at ScienceDirect

Trends in Neuroscience and Education

journal homepage: www.elsevier.com/locate/tine

Research in Perspective

Interventions for improving numerical abilities: Present and future

Roi Cohen Kadosh a,n,1, Ann Dowker a,1, Angela Heine b,1, Liane Kaufmann c,1,
Karin Kucian d,1
a
Department of Experimental Psychology, University of Oxford, 9 South Parks Road, Oxford OX1 3UD, UK
b
Department of Psychology, Freie Universität Berlin, Habelschwerdter Allee 45, 14195 Berlin, Germany
c
Department of Psychiatry and Psychotherapy A, General Hospital, A-6060 Hall in Tyrol, Austria
d
Center for MR-Research, University Children's Hospital Zurich, Steinwiesstrasse 75, 8032 Zurich, Switzerland

art ic l e i nf o a b s t r a c t

Article history: Low numeracy skills have a negative impact on the employment prospects and mental and physical
Received 22 February 2013 health of individuals, and on the economic status of countries. Clearly, this is a high priority area where
Accepted 30 April 2013 efficient strategies for intervention can lead to a better outcome, especially when implemented at an
early age. We discuss here present and future directions for intervention. The development of such
Keywords: interventions has been based on the study of numerical difficulties through methods ranging from
Cognitive training standardized tests to behavioral measures to neuroimaging. The intervention techniques range from
Developmental Dyscalculia group-based interventions targeted at strengths and weaknesses in specific components of arithmetic, to
Neuroimaging educational computer-games, to non-invasive brain-stimulation. We discuss the principles behind each
Learning Disabilities
method, the current evidence, and future directions.
Noninvasive Brain Stimulation
& 2013 Elsevier GmbH. All rights reserved.
Numerical Cognition

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
2. The componential nature of arithmetic: Implications for targeted intervention . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
3. The effect of intervention on atypical neural responsiveness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
4. Neural underpinnings and intervention using EEG . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
5. Neural underpinnings and intervention using fMRI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
6. Using transcranial electrical stimulation to improve cognitive training . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
7. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92

1. Introduction for a discussion on definition see [83]), or mathematical learning

disability (MLD) [4,10,30]. Numerical difficulties are linked to lack
Mathematical achievement is one of the foundations for a of progress in education, increased unemployment, reduced salary
thriving society. Approximately 20% of people have low numeracy and job opportunities, and additional costs in mental and physical
skills [23], and depending partly on diagnostic criteria, 3–13% of health [22,69]. Many of these increased risks operate over and
people are considered to have a more serious specific disability above those associated with social and educational disadvantages
with numbers, a condition called developmental dyscalculia (DD, in general, including those associated with literacy difficulties or
lack of qualifications [31,69]. Furthermore, the effects of numeracy
skills expand beyond the life of the individual and affect society in
n
Correspondence to: Department of Experimental Psychology, University of general [31]. The current review aims to describe the current state-
Oxford, Tinbergen Building, South Parks Road, Oxford OX1 3UD, UK.
of-the-art of interventions to improve numerical skills, based on
Tel.: +44 1865 271385; fax: +44 1865 310447.
E-mail address: [email protected] (R. Cohen Kadosh). cognitive, educational and neuroscientific research evidence on
1
All authors contributed equally to this work. the nature of mathematical cognition and learning.

2211-9493/$ - see front matter & 2013 Elsevier GmbH. All rights reserved.
http://dx.doi.org/10.1016/j.tine.2013.04.001
86 R. Cohen Kadosh et al. / Trends in Neuroscience and Education 2 (2013) 85–93

2. The componential nature of arithmetic: Implications for The componential nature of arithmetic is important in planning
targeted intervention and formulating interventions with children with arithmetical
difficulties. Interventions that focus on the particular components
One way in which neuroscience influences education is through with which an individual child has difficulty are likely to be more
the application of the findings of brain-based research to guide effective than those which assume that all children's arithmetical
approaches to teaching and intervention. Although such applications difficulties are similar.
are still at a relatively early stage, and some are based on ‘neuro- Systematic development of targeted programmes for children
myths’ rather than solid evidence [28], findings from neuroscience with mathematical difficulties began only recently [86,95].
are beginning to inform behavioral and cognitive interventions These programmes are highly intensive, and involve approxi-
[7,29]. We will focus here on the componential nature of arithmetic. mately 30 min of individualized intervention per day. They are
The most striking evidence for the functional separability of different generally targeted at children with severe difficulties: approxi-
components comes from neuropsychological studies of acquired mately the lowest-achieving 5%. However, they exclude many
dyscalculia [11,18,19]. Functional brain imaging techniques provide children with less severe numeracy difficulties that may never-
converging evidence that different components of arithmetic can theless have a serious practical impact on their lives but for whom
involve different brain areas and networks [96]. intensive intervention may not be a practical or cost-effective
possibility. In contrast, Catch Up Numeracy is an intervention
based on the ‘Numeracy Recovery’ scheme [20], which applies to
primary school children with moderate mathematical weaknesses.
It is a less intensive, but still highly targeted, intervention [21,37]
(Box 1).
Box 1–Catch Up Numeracy
The results so far indicate that participants who received the
Children in the project receive interventions from trained Catch Up intervention improved more than twice as much in
teachers or teaching assistants during two 15 min sessions Number Age on a standardized test as expected from passage of
per week, typically for one school term. time, and made significantly higher ratio gains than controls who
The components are as follows: received non-targeted mathematical intervention (Fig. 1). Thus, a
behavioral-targeted intervention program based on cognitive and
(1) Counting verbally (counting verbally from 0 or 1; neuroscientific principles of the targeted cognitive ability can lead
counting on from a given number; counting back from to successful improvement.
a given number). In the next section we will discuss the application of neuroima-
(2) Counting objects (counting objects; order irrelevance; ging to assess the effect of intervention on the neural substrates of
repeated addition of objects; repeated subtraction of atypical numeracy.
objects).
(3) Reading and writing numerals and number words.
(4) Handling tens and units (number comparison; adding
tens and units; subtracting tens and units).
(5) Ordinal numbers (stating the ordinal position – e.g.
second, fourth, etc – of a bead within a bead string).
(6) Word problems.
(7) Translation between different formats (i.e. between
quantities of objects and number words or numerals).
(8) Derived fact strategies (including the use of commu-
tativity of addition and the inversion principle for
addition and subtraction to derive unknown number
facts from a given number fact).
(9) Estimation of set size, and of answers to arithmetic
problems.
(10) Remembered number facts.

Each child is assessed individually by a trained teacher/

teaching assistant using ‘Catch Up Numeracy formative
assessments’ which the member of staff then uses to
complete the ‘Catch Up Numeracy learner profile’. This
personalized profile is used to determine the entry level for
Fig. 1. The effect of the Catch Up intervention program. In this study [37], 395
each of ten Catch Up Numeracy components and the children received the Catch Up intervention program, which was based on the
appropriate focus for numeracy teaching. Children are ‘Numeracy Recovery’ scheme [20] in collaboration with Catch Ups (a not-for-profit
provided with mathematical games and activities targeted UK charity). There were two smaller control groups: (1) Matched-time individua-
to their specific levels in specific activities. Where possible, lized mathematics intervention group (n¼ 50). This group involved reviewing work
these games and activities involve the use of materials that done in the school lessons and was not specifically targeted to assessed individual
are commonly available in schools. strengths and weaknesses. (2) No intervention group, except for the usual school
Each 15-min teaching session includes (i) a review and instruction (n¼48). All children were given a number screening test before and
introduction to remind the child of what was achieved in the after the intervention. At the start, participants’ mean age was 104.97 months
(SD¼ 13.6). Their mean mathematics age (mathematical achievements based on
previous session and to outline the focus of the current
their age) was 96.46 months (SD ¼ 14.66). The groups did not differ in chronological
session; (ii) a numeracy activity; and (iii) a linked recording
age, or in mathematics age (ps 40.35). The children who received the Catch Up
activity where the child records the results of the activity in intervention made significantly higher ratio gains (months gained in mathematics
oral, written, and/or concrete fashion, and where the child age divided by the number of months between initial and final testing) than either
receives focused teaching related to their performance in the of the other groups as indicated by a significant effect of group (F(2490) ¼14.67;
activity and to any observed error. po 0.001), and post-hoc tests. A detailed account of the program and of an
evaluation of its effectiveness is given by [37].
 R. Cohen Kadosh et al. / Trends in Neuroscience and Education 2 (2013) 85–93 87

Fig. 2. Summary of deficient brain function (pink circles), gray matter (green squares), white matter (yellow stars), and brain metabolism (blue triangle) in children with DD.
Reported deficits include a variety of brain regions, however, there seems to be consistent evidence that DD is associated with deficits in the parietal lobes (marked in white)
which host core regions for numerical understanding. (Brain templates by P.J. Lynch and C.C. Jaffe). Reproduced from [55]. Brain Correlates of Numerical Disabilities.
In R. Cohen Kadosh & A. Dowker (Eds.), The Oxford Handbook of Numerical Cognition: Oxford University Press, with permission from Oxford University Press.

3. The effect of intervention on atypical neural responsiveness Using ERP measures and standardized diagnostic measures to
assess the effects of remedial training for elementary school
Children with MLD or DD are exhibiting behavioral impair- children with MLD, a recent intervention study focused on
ments as well as atypical brain activity and anatomy [55] (Fig. 2). training-related changes in groups of second and third graders
In this section we will discuss how intervention administered in a [93]. The training was based on a highly effective intervention in
game-like fashion (Box 2) can affect behavior as well as brain third-grade children with DD [49]. The intervention program is
functions. We will offer examples both from electroencephalogra- theory-based, organized into semi-hierarchical modules, and
phy (EEG) and functional magnetic resonance imaging (fMRI), focuses on the explicit teaching of basic numerical skills (e.g.,
which provides good temporal and spatial resolution as to where semantic number knowledge) and arithmetic conceptual knowl-
activation occurs in the brain, respectively (Box 3). edge (e.g., understanding of arithmetic operations and principles).
Over a 9-months period training sessions were offered once a
week for groups of 2–6 children with each session lasting about
4. Neural underpinnings and intervention using EEG 90 min.
Analysis of the data from groups of children with MLD who
Electrophysiological investigations into basic numerical abil- either took part in the numeracy intervention (intervention group)
ities typically focus on late parietal positivities (positive-going or underwent a reading and spelling training (low achieving
deflections in the P2 and P3 time window) that are assumed to be controls), and a third group of age-matched typical achievers,
reflections of quantity-processing functions in infants, children, revealed changes that reflect gains on typically achieving peers in
and adults [17,39,40,43]. For instance, during numerical compar- diagnostic measures [93], as well as electrophysiological and
ison tasks larger amplitudes of the late parietal event-related behavioral parameters for the intervention group (Fig. 3). Adopting
potentials (ERPs) were found in response to large compared to a well-documented experimental design for the EEG part of the
small numerical distances (e.g., comparing the numbers 2 and study [5], the children were presented with symbolic and non-
8 vs. the numbers 2 and 3), both in adults [70,88] and in younger symbolic approximate addition tasks before (t1) and after (t2) the
populations ([34,78,84]; but cf. [38,61]). intervention phase. At t1, the typical achievers showed signifi-
Previous studies have reported atypical distance-related mod- cantly larger amplitudes of the critical ERP components than both
ulations of these late positive-going ERP components in adoles- groups of low achieving children, which is consistent with the
cents with DD [78] and in children with MLD [35] during results of previous electrophysiological studies [35]. However, the
numerical comparison tasks, when compared to age-matched group of children who took part in the numeracy intervention
typical achievers. The amplitudes of the late posterior positivities program showed a marked shift in the amplitudes of the late
are commonly assumed to be related to neural activity primarily in positive-going waveform (Fig. 3, [94]). This suggests that the
inferior parietal regions (e.g., [17,34,78]), which have been sug- intervention did not only affect children's behavioral performances
gested to play a causal role in numeracy [15]. as assessed by standardized diagnostic tools [93], but changes on
88 R. Cohen Kadosh et al. / Trends in Neuroscience and Education 2 (2013) 85–93

Box 2–Computer-based Interventions Box 3–Brain Imaging Methods

As the development of each child’s numerical abilities follows Cognitive neuroscience combines strategies of cognitive
different trajectories and is intertwined with the development psychology with different methods to examine brain struc-
of other cognitive domains, a high grade of individualization ture or brain function. Thanks to these modern brain imaging
is needed. Adaptive educational computer-based training can techniques, we are able to generate high resolution anato-
contribute to these requirements. Computer-based interven- mical images of our brains, examine fiber tracts, gain
tion can be designed to adapt for cognitive or performance metabolic insights, or observe brain activation while we are
profiles and provides intensive training in a stimulating performing a task.
environment. In combination with the fact that the computer Magnetic resonance imaging (MRI)
is an emotionally neutral medium, it may also foster MRI produces brain images non-invasively by a powerful
motivation and enhance positive self-concepts as every child magnet and radio-frequency. Different MRI acquisition meth-
gains feelings of success [3,79]. Moreover, computers are an ods provide information about various aspects of our brains.
attractive medium for children and seem to be effective when The recording of high resolution anatomical brain images
trainings are sensibly constructed [25,58]. However, it has to allows to differentiate between gray and white brain matter
be kept in mind that computers cannot replace teachers or and to investigate focal differences in morphometry. Alter-
therapists, but interactive games can form helpful tools for natively, DTI enables the measurement of the integrity of
successful remediation. fiber connections between different brain regions and MRS
Regarding the math intervention, only a few computer- measures brain chemistry to study changes of various brain
based trainings have been evaluated scientifically: metabolites. Finally, fMRI uses the change in oxygen levels of
The training called “Number Race” is based on principles the blood in active brain areas to create images of brain
for remediation of DD and focuses on quantity representation regions that are active during a specific task.
and the association between number and space [91]. Evalua- Positron emission tomography (PET)
tion indicated a significant improvement in basic numerical PET-imaging enables the visualization of biochemical and
cognition, but the effect did not generalize to counting or physiological functions of the brain. A radioactive tracer is
arithmetic ([72,90,92]). injected into the blood system. Areas of high radioactivity
“Elfe and Mathis I” is a recently developed computer- indicate high amounts of radioactive-labelled oxygen, and
based program which trains basic numeric capabilities, therefore are associated with brain activity; similar to the
arithmetic and geometry [59]. The program is aligned to the principle of fMRI, it is assumed that active regions are flooded
school curriculum and its evaluation demonstrated a higher with oxygenated blood.
increase in mathematical competence in the training group Near infrared spectroscopy (NIRS)
compared to matched controls. Near infrared light is shined through the head, travels
Another computer-assisted instruction (CAI) to enhance through the outer layers of the brain, and is measured by a
number combination skills has been presented by Fuchs et al. nearby receiver as it leaves the head. By measuring the
[27]. The training was effective in improving addition but not quantity of returning photons, one can infer the spectral
subtraction, and no transfer to arithmetic story problems absorption of the underlying tissue and make some conclu-
occurred. sions about its average oxygenation and deoxygenation.
Finally, the training, “Rescue Calcularis,” discussed in the Therefore, NIRS can be used for non-invasive assessment of
text, has been further developed. The new extended version brain function by detecting changes in oxygen concentrations
is called “Calcularis” and includes a variety of games in the blood which are associated with neural activity.
designed in line with current neurocognitive concepts of Electroencephalography (EEG)/ Magnetoencephalography
mathematical development, insights on DD and general (MEG)
learning principles [46,47]. The innovation of Calcularis is Electrophysiological methods register the electrical activity
the use of an adaptive control algorithm which enables of neurons non-invasively. Depolarisations of synchronously
individual adjustment on the difficulty level as well as the active neurons create electrical and magnetic fields that can
choice of appropriate games. Evaluation showed that chil- be recorded at the scalp. While EEG measures the changes of
dren benefited from the training regarding number repre- the electric field with electrodes placed on the scalp, MEG
sentation, and addition and subtraction skills [89]. records magnetic field changes by an arrangement of super-
conductive coils. In contrast to fMRI and PET, which provide
high spatial resolution but lower temporal resolution,
electrophysiological methods measure across larger regions
of the brain but can detect changes of brain activation in the
the behavioral level were accompanied by differences in brain millisecond-range.
functioning as assessed by EEG-measures. However, since ERPs
lack the necessary spatial resolution, it is unclear whether the
effects can be related to improvement in brain regions that were
initially impaired, or whether other brain regions have been
recruited to compensate for atypical brain organization. intraparietal sulcus (IPS) and the superior and inferior parietal
lobule, which are known to be core regions for numerical and
mathematical processing. However, aside from parietal areas,
5. Neural underpinnings and intervention using fMRI other cortical and subcortical regions that contribute to numerical
cognition can also be associated with mathematical difficulties.
In the last few years a clearer picture has emerged of functional Such results include reduced brain activation found by fMRI
processes in the typical adult and child brain during number [56,65,71] or EEG [35,78] and atypical brain metabolism by
processing and calculation, by means of contemporary brain magnetic resonance spectroscopy (MRS) [60], as well as reduced
imaging techniques (Box 3). However, only a small number of gray matter volume or deficient fiber connections measured by
imaging studies have addressed the question of neural correlates morphometric MRI [74] and diffusion tensor imaging (DTI) [75].
of atypical development in DD. Nevertheless, a recent meta- Moreover, compensatory mechanisms have been observed in DD
analysis has emphasized the neural aspects of DD [51]. Convergent children; these are usually characterized by stronger recruitment
evidence suggests that differences are found primarily in the of supporting areas associated with working memory, attention,
 R. Cohen Kadosh et al. / Trends in Neuroscience and Education 2 (2013) 85–93 89

Box 4–Types and Mechanisms of Transcranial Electrical

Stimulation

Transcranial direct current stimulation (TDCS) involves the

application of a constant electrical current. Studies on
animals and humans have found that the induced changes
in tissue excitability vary with current polarity. Anodal
stimulation pushes neural resting membrane potentials
closer to the activation threshold and therefore increases
tissue excitability, while the reverse polarity, cathodal
stimulation, inhibits cell firing and decreases excitability
[26,68]. Most of the studies so far found that anodal
stimulation improved human performance, while cathodal
stimulation impaired human performance [14,44].
The long-lasting effects of TDCS are protein synthesis-
dependent and are accompanied by several mechanisms
including the modifications of intracellular cyclic adenosine
monophosphate (cAMP) and calcium levels [33], brain-
derived neurotrophic factor [26], and activation of adenosine
A1 receptors [63] and therefore share some features with
long-term potentiation and long-term depression [12,66].
FMRI experiments in humans have found that TDCS can alter
local and remote brain activation [36,52]. MRS studies found
change the local concentration of GABA and glutamate [81],
which are critically involved in learning and memory [80].
Transcranial random noise stimulation (TRNS) typically
involves the generation of random ‘samples’ of alternating
electrical current at a rate of several hundred times
per second. These samples are randomly assigned current
amplitudes, which are normally distributed around a direct-
current component of 0. The random fluctuation of these
sample currents between positive and negative amplitudes
generates the electrical ‘noise’ that cortical regions of interest
are exposed to. The technique is preferred over TDCS for its
higher cutaneous perception threshold [1], making it easier to
maintain experimental blinds, and for its oscillatory rather
than direct current, which ensures that application is
independent of polarity (i.e. anodal and cathodal) [13].
Although the mechanisms underlying TRNS are less well-
studied than TDCS, and have been attributed both to
stochastic resonance or the induction of sodium ion influxes
[85], this technique has been shown to enhance cortical
excitability. The effect of TRNS has been suggested to be
Fig. 3. Panel (a) gives an overview of stimulus-locked ERPs at recording site P8 for facilitatory at both electrodes. Moreover, compared to anodal
the three groups of children [intervention group: blue; typical achievers: black; low TDCS, high-frequency TRNS (100–640 Hz) yields more power-
achieving controls: gray]. Pre- [solid lines] and post-results [dashed lines] are ful results [24].
plotted for the mean grand averages over both experimental conditions, i.e., the
nonsymbolic and symbolic approximate calculation tasks. The relevant ERP
component is highlighted, and the topography of mean difference potentials is
shown for the critical time window and contrast (i.e. typical achievers minus
intervention group at t1). Panel (b) specifies diagnostic [bluish bars, with higher
values indicating better performance levels; [48]], behavioral [greenish bars; error
rates in %] and ERP parameters [reddish bars; mean amplitudes in μV, 300–500 ms
after stimulus onset] for both points in time [t1, t2]. Asterisks denote Bonferroni-
corrected levels of statistical significance of differences between pre- and post-
testing results. (For interpretation of the references to color in this figure legend,
the reader is referred to the web version of this article.)

monitoring, updating or finger representation [50,57]. Such an

increased need for additional supportive functions might be
explained by underdevelopment of number representations, and/
or a failure in automatization of access to these representations.
Aberrant brain activation, structure or metabolism in children with
DD has not yet been integrated into the diagnose, but studies in the
field of ADHD and reading pointed to the promising potential of
combining behavioral measures with neuroimaging markers to Fig. 4. Combination of wireless TES and cognitive training using a video-game. In
improve diagnostic accuracy or to predict further outcome [8,9,76]. this example, a participant receives stimulation to the dorsolateral prefrontal
cortex while being trained on fractions. In this training the fractions need to be
However, the human brain is a highly plastic organ and ade- mapped on a horizontal line (cf. [54]) by moving her body to the respective location
quate stimulation is able to induce structural as well as func- between two anchors (zero and one). Body movements are detected using a
tional changes. A 5-weeks computer-based intervention “Rescue motion-detector sensor [64].
 90
Table 1
Overview of the different interventions described in the current review, including sample size, age, length of the intervention, and effect sizes. Note that in the case of effect size it is not accurate to compare the different
interventions, as the different interventions involved different populations, age, effects, and the intervention length varied. The reader is referred to Ise, et al., [41] for a meta-analysis which includes other types of training not
discussed here. N/A notes the inability to conclude whether the intervention enabled the individuals to improve their performance and thus catch up with their peers.

Type of Sampe Age (in Length of Country Randomization Single- Double- Control Transfer Catching up the Effect size (Cohen d’)
intervention size years) intervention blind blind group effect difference with
peers

Catch Up Numeracy n¼ 440 6–10 4 months UK No Yes No Yes Yes Yes Number Age gain: d¼ 0.47 (Intervention vs. Matched Time Control);
educational (into d ¼0.55 (Intervention vs No-Intervention Control)

R. Cohen Kadosh et al. / Trends in Neuroscience and Education 2 (2013) 85–93

intervention for 3 groups)
children [21,37]
Remedial training n¼ 64 7–11 9 months Germany No Yes No Yes Yes Arithmetic skills (HRT 1-4 ; [32]): d ¼ 0.84; Visuo-spatial skills (HRT 1-4 ;
for children with (into [32]): d ¼0.63
dyscalculia [93] 5 groups)
Rescue Calcularis n¼ 32 8–11 5 weeks Switzerland Yes No No Yes, but only Yes in linearity and positive effects of training (pre vs. post) on: Number line task:
[54] (into for children variability of dyscalculics d ¼1.08, controls d ¼ 1.15; Addition & Substraction:
3 groups) with arabic digit dyscalculics d ¼0.36, controls d ¼ 0.47
dyscalculia representation
Calcularis [89] n¼ 32 8–11 6 weeks Switzerland Yes No No Yes Yes N/A positive effects of Calcularis vs. control group: Addition d ¼0.31;
(into Subtraction d ¼ 0.39; Number line task 0-10: d¼ 0.28; 0-100: d ¼0.18;
2 groups) 0-1000: 0.15; Estimation d¼ 0.29; Subitizing d¼ 0.08; Heidelberger
Rechentest [32]: Addition d ¼ 0.16; Subtraction d¼ 0.52
Number Race n¼ 9 7–9 10 weeks France No No No No No N/A positive effects of training (pre vs. post) on: large addition problems
[91,90] (into d ¼0.33; negative effects of training (pre vs. post) on: small addition
1 group) problems d ¼ 1.59
Elfe and Mathis I n¼ 130 7–9 10 weeks Germany No No No Yes N/A N/A Math skills: d ¼ 0.59 1st graders; d¼ 0.62 2nd graders
[59] (into
4 groups)
CAI [27] n¼ 33 1st 18 weeks USA Yes No No Yes No N/A positive effects of CAI vs. spelling training: Addition d ¼ 0.49;
(into graders Subtraction d ¼ 0.02; negative effects of CAI vs. spelling training: Story
2 groups) problem d ¼0.06
TES [16] n¼ 15 20–22 6 days UK Yes Yes No Yes No N/A Numerical automaticity: d ¼ 1.09
(into
3 groups)
TES [42] n¼ 19 20–31 6 days UK Yes Yes No Yes No N/A Learning rate: d ¼ 0.85; Numerical automaticity: d ¼ 0.55
(into
3 groups)
TES [77] n¼ 51 18–28 5 days UK Yes Yes Yes Yes Yes N/A Learning rate: d ¼ 0.89 (drill learning), 0.77 (calculation learning);
(into
4 groups)
 R. Cohen Kadosh et al. / Trends in Neuroscience and Education 2 (2013) 85–93 91

Calcularis” was developed with the aim of improving number Box 5–Outstanding Questions
representations and strengthening the links between numbers
and spatial processes on the internal mental number line [54]. 1. Might the intense intervention and great emphasis on
Results have indicated that children with and without DD improving a given cognitive ability have a positive effect
improved their spatial number representations and arithmetical on other mental faculties as well? It seems plausible that a
abilities. This highlights the importance of a precise mapping of, positive learning experience has the potential to improve
and automated access to, the mental number line for adequate general attitudes towards learning by enhanced confi-
development of calculation skills. dence and motivation. However, could an intense inter-
Additionally, the training resulted in a modulation of brain vention have also a negative effect on a non-trained
functions. FMRI depicted a reduction in the recruitment of relevant ability? The latter might occur due to a shift of metabolic
brain regions after the training, including mainly frontal areas, consumption and neurochemical modulation caused by
bilateral IPS and the left fusiform gyrus. A decrease of brain the intervention, which changes the respective involve-
activation in these regions and particularly of the frontal lobe is ment of different brain areas.
assumed to reflect automatization of cognitive processes neces- 2. What are the long-term effects of the intervention
sary for mathematical reasoning [96]. In a follow-up examination programs? Do the students maintain the level displayed
5 weeks after training, a significant increase of activity in parietal at post-intervention assessments, do they improve even
areas was found in children with DD. Since the IPS is known to further improve, or do they show a decline in perfor-
play a pivotal role in number representation, these results sug- mance? May the degree of such a decline, if one occurred,
gested that time for consolidation after training was needed to be affected by the type of intervention (e.g., computer-
establish neuronal representation [54]. based vs. personal tutorials), or is it more linked to
In conclusion, domain-specific game-like interventions are individual characteristics? Will TES be able to elongate
associated with neuroplasticity in functional circuitry that is and maintain the positive effect of intervention?
impaired in children with DD and MLD, and furthermore, they 3. What are the cognitive and biological mechanisms that
can transform brain activation that is atypical in respect to time make computer-based cognitive training a successful tool
and localization, into typical brain activation. for intervention? For example, might the attractiveness,
engagement and reward-based nature of this training act
on the dopaminergic system that is involved in plasticity
([62].
6. Using transcranial electrical stimulation to improve 4. What is the temporal dynamic between behavioral and
cognitive training brain changes due to intervention? Do behavioral changes
precede changes in physiology or the other way around?
So far, we have discussed the effect of intervention on behavior 5. Intervention efficacy: Which socio-emotional, cognitive,
and brain functions. Intervention, by itself, aims to affect brain neural or genetic modulating factors may affect interven-
mechanisms by influencing cognitive functions, leading to a tion efficacy? How is the efficacy of cognitive intervention
virtuous circle whereby these changes in brain functions also and TES in children and adults influenced by factors such
impact subsequent cognitive functions. However, transcranial as age, individual differences in cognitive abilities [87], or
electrical stimulation (TES) can have a more direct influence on level of education [6], and specific genes [2].
brain functions and neuroplasticity [14,53] (Box 4). 6. Which intervention methods (cognitive, neuronal stimula-
TES delivers weak electrical currents (e.g., 1–2 mA) via electro- tion, etc.) and which combination of methods are most apt
des, most frequently at the size of 25–35 cm2, which are placed on to exert positive intervention effects? Moreover, some-
the scalp above the brain area that the experimenter is interested times interventions improve performance on a specific
in affecting. When the current is applied over a short duration task, but do not transfer to similar tasks [72]. Can these
(∼20 min), it passes painlessly through the scalp and skull and methods or their combination increase the likelihood for a
alters spontaneous neural activity [26,67]. transfer effect? Is there a systematic relation between
The recent results obtained from TES experiments offer pro- intervention efficacy, neural changes and severity of
mising possibilities for both the cognitive enhancement of normal mathematical difficulty?
abilities and treatment of impairments in different domains
including attention, working memory, numeracy, language, and
executive functions (for reviews see, [14,44,53]).
In the numerical domain, TES positively impacted basic numer- the possible mental cost of cognitive enhancement [42] in this
ical skills, arithmetic training, symbolic learning, and automaticity population.
[16,42,77]. Notably, some of these studies have found long-lasting
behavioral effects [16], including transfer effect to non-learned
material, and long-lasting efficiency in brain functions in the 7. Summary
stimulated brain region [77] that span 6 months.
Results so far have indicated that stimulation needs to be paired In this review, we discussed recent approaches to intervention
with cognitive training intervention and that the timing of stimula- such as targeted intervention and computer-based intervention, as
tion with respect to task performance has important effects [82]. In well as the effect of intervention on brain functions, and the
this regard, when the aim is to improve learning, TES during possibility in the future in enhancing cognitive training interven-
intervention yields the most robust results [73,82]. tion using TES. These approaches and their possible combinations
TES is a portable, painless, non-invasive and inexpensive (Fig. 4) serve as an excellent example for the fruitful synergy
method. These characteristics increase the likelihood of future among the fields of psychology, neuroscience, and education;
use of TES in different populations outside of the laboratory, in together, these disciplines can contribute to optimal designs for
clinics or in educational institutions [53]. However, currently there intervention targeting neurocognitive mechanisms, and can
is only a limited amount of work with pediatric populations [53], furthermore evaluate the efficacy of such interventions at the
which leaves questions as to its safety and efficacy, as well as to behavioral and brain levels. As with any new development, some
92 R. Cohen Kadosh et al. / Trends in Neuroscience and Education 2 (2013) 85–93

of the interventions are still at an early stage. E.g., some studies [18] Delazer M. Neuropsychological findings on conceptual knowledge of arith-
might have involved relatively small, non-random samples, or did metic. In: Baroody AJ, Dowker A, editors. The development of arithmetic
concepts and skills. Mahwah, N.J: Erlbaum; 2003. p. 385–407.
not include control groups (see Table 1 for a summary of the [19] Demeyere N, Rotshtein P, Humphreys GW. The neuroanatomy of visual
studies in this review). However, as we described here there is enumeration: differentiating necessary neural correlates for subitizing versus
increasing evidence for the effectiveness, in the short- and even counting in a neuropsychological voxel-based morphometry study. Journal of
Cognitive Neuroscience 2012;24:948–64.
long-term, of some interventions, including transfer effect to non- [20] Dowker A. Individual differences in arithmetic. Oxford: Psychology Press;
trained material (Table 1) that is sometimes lacking in interven- 2005.
tions (Box 5). While much work is still needed and outstanding [21] Dowker A, Sigley II G. Targeted interventions for children with arithmetical
difficulties. British Journal of Educational Psychology Monograph Series II
questions need to be answered (Box 5), the current review
2010;7:65–81.
provides an example of the potential for improving and optimizing [22] Duncan GJ, Dowsett CJ, Claessens A, Magnuson K, Huston AC, Klebanov P, et al.
intervention for learning difficulties. School readiness and later achievement. Developmental Psychology
2007;43:1428–46.
[23] EACEA/Eurydice. Mathematics Education in Europe: Common Challenges and
National Policies; 2011.
Acknowledgments [24] Fertonani A, Pirulli C, Miniussi C. Random noise stimulation improves
neuroplasticity in perceptual learning. The Journal of Neuroscience 2011;31
(43):15416–23.
We would like to thank Jackie Thompson for helpful comments [25] Fletcher-Flinn CM, Gravatt B. The efficacy of Computer Assisted Instruciton
and Chung Yen Looi and Xinyang Hong for Fig. 4. AD thanks her (CAI): a meta-analysis. Journal of Educational Computing Research 1995;12
(3):219–41.
collaborators at Catch Up, including Graham Sigley. RCK is sup-
[26] Fritsch B, Reis J, Martinowich K, Schambra HM, Ji Y, Cohen LG, et al. Direct
ported by the Wellcome Trust (WT88378), and filed a patent for an current stimulation promotes BDNF-dependent synaptic plasticity: potential
apparatus for improving and/or maintaining numerical ability. implications for motor learning. Neuron 2010;66(2):198–204.
AD's work was part-funded by the Caxton Trust and the Esmee [27] Fuchs LS, Fuchs D, Hamlet CL, Powell SR, Capizzi AM, Seethaler PM. The effects
of computer-assisted instruction on number combination skill in at-risk first
Fairbairn Trust. The information in Box 3 has been adapted from graders. Journal of Learning Disabilities 2006;39(5):467–75.
[45]. The sole role of the above mentioned funding body has been [28] Geake J. Neuromythologies in education. Educational Research 2008;50:
funding. 123–33.
[29] Goswami U. Neuroscience and education: from research to practice? Nature
Reviews Neuroscience 2006;7:406–11.
[30] Gross-Tsur V, Manor O, Shalev RS. Developmental dyscalculia: prevalence
References and demographic features. Developmental Medicine and Child Neurology
1996;38:25–33.
[1] Ambrus GG, Paulus W, Antal A. Cutaneous perception thresholds of electrical [31] Gross J, Hudson C, Price D. The long term costs of numeracy difficulties: Every
stimulation methods: comparison of tDCS and tRNS. Clinical Neurophysiology Child a Chance Trust (KPMG); 2009.
2010;121(11):1908–14. [32] Haffner J, Baro K, Parzer P, Resch, F. Der Heidelberger Rechentest, Erfassung
[2] Antal A, Chaieb L, Moliadze V, Monte-Silva K, Poreisz C, Thirugnanasambandam mathematischer Basiskompetenzen im Grundschulalter (HRT 1-4). Göttingen:
N, et al. Brain-derived neurotrophic factor (BDNF) gene polymorphisms shape Hogrefe; 2005.
cortical plasticity in humans. Brain Stimulation 2010;3(4):230–7. [33] Hattori Y, Moriwaki A, Hori Y. Biphasic effects of polarizing current on
[3] Ashcraft M, Faust M. Mathematics anxiety and mental arithmetic performance: adenosine-sensitive generation of cyclic AMP in rat cerebral cortex. Neu-
an exploratory investigation. Cognition and Emotion 1994;8:97–125. roscience letters 1990;116(3):320–4.
[4] Barbaresi WJ, Katusic SK, Colligan RC, Weaver AL, Jacobsen SJ. Math learning [34] Heine A, Tamm S, Wißmann J, Jacobs AM. Electrophysiological correlates of
disorder: incidence in a population-based birth cohort, 1976–82, Rochester, non-symbolic numerical magnitude processing in children: Joining the dots.
Minn. Ambulatory Pediatrics: the Official Journal of the Ambulatory Pediatric Neuropsychologia 2011;49(12):3238–46.
Association 2005;5(5):281–9. [35] Heine A, Wißmann J, Tamm S, De Smedt B, Schneider M, Stern E, et al. An
[5] Barth H, La Mont K, Lipton J, Dehaene S, Kanwisher N, Spelke E. Non-symbolic electrophysiological investigation of non-symbolic magnitude processing:
arithmetic in adults and young children. Cognition 2006;98(3):199–222. numerical distance effects in children with and without mathematical learn-
[6] Berryhill ME, Jones KT. tDCS selectively improves working memory in older ing disablities. Cortex 2012. http://dx.doi.org/10.1016/j.cortex.2012.11.009
adults with more education. Neuroscience Letters 2012;521(2):148–51. [Epub ahead of print].
[7] Blakemore SJ, Frith U. The learning brain: lessons for education. Oxford: [36] Holland R, Leff Alex P, Josephs O, Galea Joseph M, Desikan M, Price Cathy J,
Blackwell; 2005. et al. Speech facilitation by left inferior frontal cortex stimulation. Current
[8] Brem S, Bach S, Lyytinen H, Maurer U, Richardson U, & Brandeis D. (). Biology 2011;21(16):1403–7.
Electrophysiological correlate of pre-literate print sensitivity—a predictor for [37] Holmes W, Dowker A. Catch up numeracy: a targeted intervention for children
reading outcome? Paper presented at the Organization of the Human Brain who are low-attaining in mathematics. Research in Mathematics Education
Mapping; 2012. Retrieved from: 〈http://ww4.aievolution.com/hbm1201/index. 2013 in press.
cfm?do=abs.viewAbs&abs=6203〉. [38] Hyde DC, Spelke ES. All numbers are not equal: an electrophysiological
[9] Brown MR, Sidhu GS, Greiner R, Asgarian N, Bastani M, Silverstone PH, et al. investigation of small and large number representations. Journal of Cognitive
ADHD-200 Global Competition: diagnosing ADHD using personal characteristic Neuroscience 2009;21(6):1039–53.
data can outperform resting state fMRI measurements. Frontiers in Systems [39] Hyde DC, Spelke ES. Neural signatures of number processing in human
Neuroscience 2012;6:69. infants: evidence for two core systems underlying numerical cognition.
[10] Butterworth B. Foundational numerical capacities and the origins of dyscal- Developmental Science 2011;14(2):360–71.
culia. Trends in Cognitive Sciences 2010;14(12):534–41. [40] Hyde DC, Spelke ES. Spatio-temporal dynamics of numerical processing:
[11] Cappelletti M, Butterworth B, Kopelman M. Numeracy skills in patients with an ERP source localization study. Human Brain Mapping 2012;33
degenerative disorders and focal brain lesions: a neuropsychological investi- (9):2189–203.
gation. Neuropsychology 2012;26:1–19. [41] Ise E, Dolle K, Pixner S, Schulte-Körne G. Effektive Förderung rechenschwacher
[12] Castillo PE, Chiu CQ, Carroll RC. Long-term plasticity at inhibitory synapses. Kinder: Eine Metaanalyse. Kindheit und Entwicklung 2012;21(3):181–92.
Current Opinion in Neurobiology 2011;21(2):328–38. [42] Iuculano T, Cohen Kadosh R. The mental cost of cognitive enhancement. The
[13] Chaieb L, Kovacs G, Cziraki C, Greenlee M, Paulus W, Antal A. Short-duration Journal of Neuroscience 2013;33(10):4482–6.
transcranial random noise stimulation induces blood oxygenation level [43] Izard V, Dehaene-Lambertz G, Dehaene S. Distinct cerebral pathways for
dependent response attenuation in the human motor cortex. Experimental object identity and number in human infants. PLoS Biology 2008;6:e11.
Brain Research 2009;198(4):439–44. [44] Jacobson L, Koslowsky M, Lavidor M. tDCS polarity effects in motor and
[14] Cohen Kadosh R. Using transcranial electrical stimulation to enhance cognitive cognitive domains: a meta-analytical review. Experimental Brain Research
functions in the typical and atypical brain. Translational Neuroscience 2013;4 2012;216(1):1–10.
(1):20–33. [45] Käser T, Baschera G-M, Kohn J, Kucian K, Richtmann V, Grond U, et al.
[15] Cohen Kadosh R, Cohen Kadosh K, Schuhmann T, Kaas A, Goebel R, Henik A, Calcularis—a computer-based training program for enhancing numerical
et al. Virtual dyscalculia induced by parietal-lobe TMS impairs automatic cognition. Learning and Instruction 2013 in preparation.
magnitude processing. Current Biology 2007;17:689–93. [46] Käser T, Busetto A, Baschera G-M, Kohn J, Kucian K, von Aster M, et al.
[16] Cohen Kadosh R, Soskic S, Iuculano T, Kanai R, Walsh V. Modulating neuronal Modelling and optimizing the process of learning mathematics. Intelligent
activity produces specific and long lasting changes in numerical competence. Tutoring Systems 2012;7315:389–98.
Current Biology 2010;20:2016–20. [47] Käser T, Kucian K, Ringwald M, Baschera G-M, von Aster M, Gross M. Therapy
[17] Dehaene S. The organization of brain activations in number comparison: software for enhancing numerical cognition. In: Özyurt J, Anschütz A, Bernholt
event-related potentials and the additive-factors method. Journal of Cognitive S, Lenk J, editors. Interdisciplinary perspectives on cognition, education and
Neuroscience 1996;8:47–68. the brain - Hanse-Studies, Vol. 7. Oldenburg: BIS-Verlag; 2011. p. 207–16.
 R. Cohen Kadosh et al. / Trends in Neuroscience and Education 2 (2013) 85–93 93

[48] Kaufmann L, Graf M, Krinzinger H, Delazer M, Willmes K. Test zur Erfassung [72] Räsänen P, Salminen J, Wilson AJ, Aunio P, Dehaene S. Computer-assisted
numerisch-rechnerischer Kompetenzen vom Kindergarten bis zur 3. Klasse intervention for children with low numeracy skills. Cognitive Development
(TEDI-MATH). Bern: Huber; 2008. 2009;24(4):450–72.
[49] Kaufmann L, Handl P, Thöny B. Evaluation of a numeracy intervention [73] Reis J, Fritsch B. Modulation of motor performance and motor learning by
program focusing on basic numerical knowledge and conceptual knowledge: transcranial direct current stimulation. Current Opinion in Neurology 2011;24
a pilot study. Journal of Learning Disabilities 2003;36(6):564–73. (6):590–6 510.1097/WCO.1090b1013e32834c32833db32830.
[50] Kaufmann L, Vogel S, Starke M, Kremser C, Schocke M, Wood G. Develop- [74] Rotzer S, Kucian K, Martin E, von Aster M, Klaver P, Loenneker T. Optimized
mental dyscalculia: compensatory mechanisms in left intraparietal regions in voxel-based morphometry in children with developmental dyscalculia. Neuro-
response to nonsymbolic magnitudes. Behavioral and Brain Functions 2009;5 image 2008;39(1):417–22.
(1):35. [75] Rykhlevskaia E, Uddin LQ, Kondos L, Menon V. Neuroanatomical correlates of
[51] Kaufmann L, Wood G, Rubinsten O, Henik A. Meta-analyses of develop- developmental dyscalculia: combined evidence from morphometry and
mental fMRI studies investigating typical and atypical trajectories of number tractography. Frontiers in Human Neuroscience 2009;3:51.
processing and calculation. Developmental Neuropsychology 2011;36(6): [76] Sidhu GS, Asgarian N, Greiner R, Brown MR. Kernel Principal Component
763–87. Analysis for dimensionality reduction in fMRI-based diagnosis of ADHD.
[52] Keeser D, Meindl T, Bor J, Palm U, Pogarell O, Mulert C, et al. Prefrontal Frontiers in Systems Neuroscience 2012;6:74.
transcranial direct current stimulation changes connectivity of resting- [77] Snowball A, Tachtsidis I, Popescu T, Thompson J, Delazer M, Zamarian L, et al.
state networks during fMRI. The Journal of Neuroscience 2011;31(43): Long-term enhancement of brain function and cognition using cognitive
15284–93. training and brain stimulation. Current Biology 2013;23:987–92.
[53] Krause B, Cohen Kadosh R. Can transcranial electrical stimulation improve [78] Soltész F, Szűcs D, Dékány J, Márkus A, Csépe V. A combined event-related
learning difficulties in atypical brain development? A future possibility for potential and neuropsychological investigation of developmental dyscalculia.
cognitive training Developmental Cognitive Neuroscience 2013 in press. Neuroscience Letters 2007;417(2):181–6.
[54] Kucian K, Grond U, Rotzer S, Henzi B, Schönmann C, Plangger F, et al. Mental [79] Spitzer M. Ja, ich kann! Selbstbild, Selbstbejahung und nachhaltige Leis-
number line training in children with developmental dyscalculia. NeuroImage tungsfähigkeit. Nervenheilkunde 2009;28:425–30.
2011;57:782–95. [80] Stagg CJ, Bachtiar V, Johansen-Berg H. The Role of GABA in Human Motor
[55] Kucian K, Kaufmann L, Von Aster M. Brain correlates of numerical disabilities. Learning. Current Biology 2011;21:480–4.
In: Cohen Kadosh R, Dowker A, editors. The Oxford handbook of numerical [81] Stagg CJ, Best JG, Stephenson MC, O'Shea J, Wylezinska M, Kincses ZT, et al.
cognition. Oxford University Press; 2013 in press. Polarity-sensitive modulation of cortical neurotransmitters by transcranial
[56] Kucian K, Loenneker T, Dietrich T, Dosch M, Martin E, von Aster M. Impaired stimulation. Journal of Neuroscience 2009;29:5202–9.
neural networks for approximate calculation in dyscalculic children: a func- [82] Stagg CJ, Jayaram G, Pastor D, Kincses ZT, Matthews PM, Johansen-Berg H.
Polarity and timing-dependent effects of transcranial direct current stimula-
tional MRI study. Behavioral and Brain Functions 2006;2:31.
tion in explicit motor learning. Neuropsychologia 2011;49(5):800–4.
[57] Kucian K, Loenneker T, Martin E, von Aster M. Non-symbolic numerical
[83] Szucs D, Goswami U. Developmental dyscalculia: fresh perspectives. Trends in
distance effect in children with and without developmental dyscalculia: a
Neuroscience and Education 2013 this issue.
parametric FMRI study. Developmental neuropsychology 2011;36(6):741–62.
[84] Temple E, Posner MI. Brain mechanisms of quantity are similar in 5-year-old
[58] Kulik JA. Meta-analytic studies of findings on computer-based instruction. In:
children and adults. Proceedings of the National Academy of Sciences of the
Baker EL, O'Neill HF, editors. Technology assessment in education and training.
United States of America 1998;95:7836–41.
Hillsdale, N. J.: LEA Publishers; 1994.
[85] Terney D, Chaieb L, Moliadze V, Antal A, Paulus W. Increasing human brain
[59] Lenhard A, Lenhard W, Schug M, Kowalski A. Computer based math training
excitability by transcranial high-frequency random noise stimulation. The
with “Rechenspiele mit Elfe und Mathis I”: Presentation and evaluation of a
Journal of Neuroscience 2008;28:14147–55.
computer program for first to third graders. Zeitschrift für Entwicklungspsy-
[86] Torgerson CJ, Wiggins A, Torgerson DJ, Ainsworth H, Barmby P, Hewitt C., et al.
chologie und Pädagogische Psychologie 2011;43(2):79–88.
Every Child Counts: The Independent Evaluation Executive Summary; 2011.
[60] Levy LM, Reis IL, Grafman J. Metabolic abnormalities detected by 1H-MRS in
[87] Tseng P, Hsu T-Y, Chang C-F, Tzeng OJL, Hung DL, Muggleton NG, et al.
dyscalculia and dysgraphia. Neurology 1999;53(3):639–41.
Unleashing potential: transcranial direct current stimulation over the right
[61] Libertus ME, Woldorff MG, Brannon EM. Electrophysiological evidence for
posterior parietal cortex improves change detection in low-performing
notation independence in numerical processing. Behavioral and Brain Func-
individuals. The Journal of Neuroscience 2012;32(31):10554–61.
tions 2007;3:1–15. [88] Turconi E, Jemel B, Rossion B, Seron X. Electrophysiological evidence for
[62] Lisman J, Grace AA, Duzel E. A neoHebbian framework for episodic memory; differential processing of numerical quantity and order in humans. Cognitive
role of dopamine-dependent late LTP. Trends in Neurosciences 2011;34 Brain Research 2004;21:22–38.
(10):536–47. [89] von Aster M, Käser T, Kucian K, Gross M. Calcularis—Rechenschwäche
[63] Márquez-Ruiz J, Leal-Campanario R, Sánchez-Campusano R, Molaee-Ardekani mit dem Computer begegnen. Schweizerische Zeitschrift für Heilpädagogik
B, Wendling F, Miranda PC, et al. Transcranial direct-current stimulation 2012;6:32–6.
modulates synaptic mechanisms involved in associative learning in behaving [90] Wilson AJ, Dehaene S, Dubois O, Fayol M. Effects of an adaptive game
rabbits. Proceedings of the National Academy of Sciences 2012;109 interventjion on accessing number sense in low-socioeconomic-status kin-
(17):6710–5. dergarten children. Mind, Brain, and Education 2009;3(4):224–34.
[64] Moeller K, Fischer U, Link T, Wasner M, Huber S, Cress U, et al. Learning and [91] Wilson AJ, Dehaene S, Pinel P, Revkin SK, Cohen L, Cohen D. Principles
development of embodied numerosity. Cognitive Processing 2012;13(0): underlying the design of “The Number Race”, an adaptive computer game
271–4. for remediation of dyscalculia. Behavioral and brain functions 2006;2(1):19.
[65] Mussolin C, De Volder A, Grandin C, Schlogel X, Nassogne MC, Noel MP. Neural [92] Wilson AJ, Revkin SK, Cohen D, Cohen L, Dehaene S. An open trial assessment
correlates of symbolic number comparison in developmental dyscalculia. of “The Number Race”, an adaptive computer game for remediation of
Journal of cognitive neuroscience 2010;22(5):860–74. dyscalculia. Behavioral and Brain Functions 2006;2(1):20.
[66] Neves G, Cooke SF, Bliss TVP. Synaptic plasticity, memory and the hippocam- [93] Wißmann J, Heine A, Handl P, Jacobs AM. Förderung von Kindern mit
pus: a neural network approach to causality. Nature Reviews Neuroscience isolierter Rechenschwäche und kombinierter Rechen- und Leseschwäche:
2008;9(1):65–75. Evaluation eines numerischen Förderprogramms für Grundschüler [Remedia-
[67] Nitsche MA, Cohen LG, Wassermann EM, Priori A, Lang N, Antal A, et al. tion for children with mathematical difficulties: evaluation of a numerical
Transcranial direct current stimulation: state of the art 2008. Brain Stimula- intervention program for primary school children.]. Lernen und Lernstörun-
tion 2008;1:206–23. gen 2013;2(2):91–109.
[68] Nitsche MA, Paulus W. Excitability changes induced in the human motor [94] Wißmann J, Heine A, Jacobs AM. Evaluation eines numerisch-konzeptuellen
cortex by weak transcranial direct current stimulation. Journal of Physiology Förderprogramms für rechenschwache Grundschulkinder (Evaluation of a numer-
2000;527:633–9. ical intervention program for primary school children). Paper presented at the
[69] Parsons S, Bynner J. Does numeracy matter more? London: NRDC; 2005. the 1st international symposium of the German federal research initiative
[70] Paulsen DJ, Woldorff MG, Brannon EM. Individual differences in nonverbal Entwicklungsstörungen schulischer Fertigkeiten; 2012.
number discrimination correlate with event-related potentials and measures [95] Wright RJ, Martland J, Stafford A. Early numeracy: assessment for teaching and
of probabilistic reasoning. Neuropsychologia 2010;48(13):3687–95. intervention. 2nd ed. London: Paul Chapman; 2006.
[71] Price GR, Holloway I, Räsänen P, Vesterinen M, Ansari D. Impaired parietal [96] Zamarian L, Ischebeck A, Delazer M. Neuroscience of learning arithmetic—
magnitude processing in developmental dyscalculia. Current Biolology Evidence from brain imaging studies. Neuroscience and Biobehavioral
2007;17(24):R1042–1043. Reviews 2009;33(6):909–25.