NeuroImage 46 (2009) 522529

Contents lists available at ScienceDirect

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

Quantity without numbers and numbers without quantity in the parietal cortex
Marinella Cappelletti , Neil Muggleton, Vincent Walsh
Institute of Cognitive Neuroscience and Dept of Psychology, University College London, 17 Queen Square, London WC1N 3AR, UK

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

Article history: A dominant view in numerical cognition is that processing the quantity indicated by numbers (e.g. deciding
Received 11 June 2008 the larger between two numbers such as 12.07 or 15.02) relies on the intraparietal regions (IPS) of the
Revised 1 February 2009 cerebral cortex. However, it remains unclear whether the IPS could play a more general role in numerical
Accepted 7 February 2009
cognition, for example in (1) quantity processing even with non-numerical stimuli (e.g. choosing the larger
Available online 21 February 2009
of bikini and coat); and/or (2) conceptual tasks involving numbers beyond those requiring quantity
processing (e.g. attributing a summer date to either 12.07 or 15.02).
In this study we applied fMRI-guided TMS to the left and right IPS, while independently manipulating
stimulus and task. Our results showed that IPS involvement in numerical cognition is neither stimulus-
specic nor specic for conceptual tasks. Thus, quantity judgments with numerical and non-numerical
stimuli were equally affected by IPS-TMS, as well as a number conceptual task not requiring quantity
comparisons. However, IPS-TMS showed no impairment for perceptual decisions on numbers without any
conceptual processing (i.e. colour judgment), nor for conceptual decisions that did not involve quantity or
number stimuli (e.g. summer object: bikini or coat?). These results are consistent with proposals that the
parietal areas are engaged in the conceptual representation of numbers but they challenge the most common
view that number processing is so automatic that the simple presentation of numbers activates the IPS and a
sense of magnitude. Rather, our results show that the IPS is only necessary when conceptual operations need
to be explicitly oriented to numerical concepts.
2009 Elsevier Inc. Open access under CC BY license.

Introduction performance in terms of increases in response times in number

Neuroimaging studies have shown that the parietal regions,
especially those around the intraparietal sulcus (IPS)1 are reliably
activated when processing number quantities, for instance when
deciding on the larger of two numbers (e.g. Dehaene et al., 2003;
Nieder, 2005). Some studies have further differentiated the ne-
grained structures within the IPS and provided evidence that the
bilateral horizontal segment of the IPS2 plays a role in quantity
processing (Dehaene et al., 2003). Further evidence of the involve-
ment of the IPS regions in number processing comes from studies
investigating how this process is affected by permanent neurological
damage in patients or temporary disruption following transcranial
magnetic stimulation (TMS). Patients with left parietal damage, for
example, can be impaired at processing number quantities (e.g.
Cipolotti et al., 1991; Dehaene and Cohen, 1991; Lemer et al., 2003;
Polk et al., 2001), and TMS studies have reported impaired
Corresponding author. Fax: +44 207 813 2835.
E-mail address: m.cappelletti@ucl.ac.uk (M. Cappelletti).
1
In the text we refer to the intraparietal sulcus in terms of the brain regions around
^
the sulcus, rather than the sulcus per se, although for brevity we indicate these as IPS or
IPS regions.
2
The horizontal segment of the IPS is the middle part of the sulcus in the dorsal
^
parietal lobe that intersects the postcentral sulcus (Ono et al., 1990).
comparison following IPS stimulation (e.g. Andres et al., 2005;
Cappelletti et al., 2007). Moreover, recent studies on developmental
dyscalculia showed that this is associated with right IPS dysfunctions
(Molko et al., 2003; Price et al., 2008; Rotzer et al., 2008).
Despite this converging evidence, it is still unknown whether the
IPS is critical for: (1) quantity processing irrespective of the stimuli
used, i.e. numerical or non-numerical; and (2) other conceptual
processing of numbers that do not require quantity manipulation (e.g.
attributing a summer date to either 12.07 or 15.02). A few studies
have examined the stimulus-specic nature of quantitative proces-
sing. These have proposed that quantitative judgments on different
types of stimuli may either rely on distinct subregions within the
parietal lobe, or that the parietal regions are involved in a generic
process of comparison of various types of stimuli. Behavioural support
for the hypothesis of a generic comparison mechanism can be found in
the evidence of a similar distance effect in many continua other than
numerical stimuli (e.g. Fulbright et al., 2003; Johnson, 1939; Moyer,
1973). That is, the time it takes to compare two stimuli increases as the
distance between the two stimuli decreases (i.e. distance effect,
Moyer and Landauer, 1967). However, the results of imaging studies
examining this issue are mixed: while some of these studies showed
common activation of the parietal areas in quantitative judgments
irrespective of the stimuli used (e.g. Fias et al., 2003), others
demonstrated that different stimuli are represented in distributed

1053-8119/ 2009 Elsevier Inc. Open access under CC BY license.

doi:10.1016/j.neuroimage.2009.02.016
 M. Cappelletti et al. / NeuroImage 46 (2009) 522529
and overlapping neural populations within the parietal areas (Fulb-
right et al., 2003; Cohen Kadosh et al., 2005; Pinel et al., 2004). These
discordant results may depend on the tasks used: for instance in
tasks such as number comparison, quantity information, has to be
extracted from the stimuli, whereas in the comparison of variables
such as luminance and physical size (e.g. Pinel et al., 2004) or angles
and lines (e.g. Fias et al., 2003), quantity information can be obtained
by visual inspection of the stimuli. Thus, there is an important
distinction between quantity processing based on extracting prior
knowledge associated with the stimuli, and quantity processing that
can be based on the physical properties of the stimulus. At present
there is still a need to know whether the IPS regions are involved in
quantity processing of non-numerical stimuli with similar task
demands.
Although the most distinctive meaning of numbers is to express
numerical quantity or cardinality (e.g. 5 items) this is not the only
information that numbers convey (Wiese, 2003). They carry other
types of information that do not imply quantity manipulations, for
instance hours (e.g. 7.15 a.m.), dates (e.g. 2006) and mathematical
constants (e.g. 3.14). Although these meanings are clearly different
from quantity information, this distinction has not yet been fully
explored. Suggestive evidence of this dissociation comes from lesion
studies showing dissociations between quantity manipulation and
processing of other non-quantity meanings of numbers (e.g. Cappel-
letti et al., 2008). Moreover, a recent fMRI study reported a distinction
between processing numbers in arithmetical and encyclopaedic
contexts, which are thought to rely on parietal and left temporal
regions respectively (Lochy and Van Turennout, 2005). Some recent
studies have focused on another type of non-quantity information
conveyed by numbers that is order. This refers to the use of numbers to
indicate the position of an item or event in an ordered sequence (e.g.
the 5th items, Butterworth, 1999; Dehaene, 1997; Jacob and Nieder,
2007). A few recent studies have shown that ordinal and quantity
information dissociate both behaviourally (e.g. Gevers et al., 2003;
Turconi et al., 2006; Zamarian et al., 2007) and temporally: namely
they take place in overlapping brain regions of the parietal cortex on
different time scales (e.g. Turconi et al., 2004). All this evidence
suggests a distinction between quantitative and non-quantitative
meanings of numbers including ordinal information. However, it is
still unknown to what extent the IPS regions are critical for these non-
quantity conceptual processes of numbers, and whether the IPS is
equally critical for number processes that do and do not involve
quantity information.
The present study
In this study, we applied fMRI-guided TMS and factorially
manipulated stimulus (numerical and non-numerical) and task
(quantity and non-quantity) independently. We aimed to test the
role of left and right IPS in: (1) quantity processing with numerical or
non-numerical stimuli in terms of object names; and in (2) conceptual
processing of numbers requiring or not requiring the manipulation of
quantity information.
To address these issues, subjects were engaged in three different
categorization tasks. Two focused on the conceptual attributes of
either quantity (e.g. choosing the larger of 12.07 and 15.02, or the
larger of bikini and coat) or category (e.g. attributing a summer date
to either 12.07 and 15.02 or choosing a summer object between
bikini and coat). The third was a perceptual task that focused the
subjects' attention on the colour of the stimulus (e.g. choosing a
number or object name drawn in red) and therefore involved both
attention to the stimulus and response selection as in the other tasks
but without requiring a conceptual decision. We chose a non-quantity
conceptual task (e.g. attributing a summer date to either 12.07 or
15.02) that required categorical processing of numerical information
but it was unlikely to rely on quantitative manipulations.
523
Material and methods
Stimuli and tasks
The experiment was controlled using the Cogent Graphics toolbox
(http://www.vislab.ucl.ac.uk/Cogent/) and Matlab7 software. The
viewing distance was approximately 0.5 m. A total of 144 Arabic
numbers and 144 object names were generated. Arabic numbers were
presented as pairs of 3 and 4-digits separated by a dot, e.g. 12.07. They
referred to a linear dimension of quantity, to dates (e.g. 12th July) or
times (e.g. seven minutes past twelve in the morning). Numerals
indicating quantities ranged from 1 to 31 for the rst half of the
numerical expression and from 01 to 59 for the second part (i.e. from
1.01 to 23.59). Numerals indicating dates were chosen to represent
either summer or winter days in the Northern hemisphere; therefore
summer dates included the months of June, July, and August, winter
dates included the months of December, January and February. Dates
were expressed in terms of day and month separated by a dot (e.g.
12th July = 12.07). Numerals referring to a date ranged from 01 to 31
for the rst half of each numerical expression and from 01 to 12 for the
second part (i.e. from 01.01 to 31.12). Each date indicating a summer
(or winter) month was presented with either a winter (or summer)
month, e.g. 12.07 and 15.02 or with a neutral month, i.e. 12.07 and
24.03. Numerals indicating times were chosen to refer to either a
sleeping or a working time approximately in terms of 8am to 6pm
working day. Therefore, working times were chosen between 8am and
6pm, and sleeping times between 10pm and 7am. Times were
expressed in terms of 24-hour clock with the rst pair of digits
referring to the hour and the other two digits, separated by a dot,
referring to the minutes past the hour (e.g. 16.30 is half past four in the
afternoon). Numbers referring to a time ranged from 00 to 23 for the
rst half of each numerical expression and from 01 to 59 for the
second part (i.e. from 00.01 to 23.59). Object names referred to
concrete, countable objects whose size could be unambiguously
identied and that could be used in both the quantity (e.g. larger
object: bikini or coat?) and the non-quantity tasks (e.g. summer
object: bikini or coat?). In the latter task, each object name in a pair
consisted of a stimulus unambiguously related to a season (e.g. bikini
for summer) and presented with another object name either
belonging to another season (e.g. coat for winter) or with a neutral
object, i.e. not related to any season (e.g. car). All the experimental
stimuli had been previously piloted in a study with different
participants.
Procedure
The 3 different tasks (quantity, non-quantity categorical and
perceptual-decision) with 2 stimuli (numbers and object names)
were blocked (6 stimuli per block) and presented in pseudo-random
order. Participants viewed pairs of stimuli presented one above the
other with a xation cross in the middle of the computer screen.
For each pair, they were instructed to indicate with a button press
which stimulus was the correct response to a question consisting of
two key words presented above the upper stimulus before and during
a block of 6 trials (see Fig. 1). On every trial, participants were
instructed to press the upper or the lower-arrow keys on the keyboard
for the upper or the lower stimulus respectively.3 Trials where the
correct answer was the upper or the lower stimulus were presented in
equal proportion.
In each pair, each stimulus was presented in one of four possible
colours (red, yellow, blue, green). In the quantity and in the non-
3
Participants' ngers rested on the two designated keys of the keyboard for the
whole duration of the experiment following a procedure common in two-choice
response experiments. This is unlikely to result in response priming as correct
responses were randomized.
524 M. Cappelletti et al. / NeuroImage 46 (2009) 522529

Fig. 1. Experimental design. In each trial, participants viewed pairs of stimuli presented one above the other with a xation cross in the middle of the computer screen. Stimuli could
be either two Arabic numbers (left column) or two object names (right column) presented in one of four possible colours (red, yellow, blue, green). Participants were instructed to
indicate with a button press which of these stimuli was the correct response to a question consisting of two key words presented above the upper stimulus before and during the
stimulus display. There were three types of tasks: for quantity tasks (A), there were four possible types of questions: larger?, smaller?, more?, less?. For non-quantity categorical
tasks (B), there were four different types of questions: summer month/object?, winter month/object?, sleeping time/object?, working time/object? (either month or object
was displayed depending on the stimulus condition). For the stimulus-colour decision (C) the questions were: blue/red/yellow/green number or object? The 3 different tasks
(quantity, non-quantity categorical and colour-decision) with both stimuli (numbers and object names) were blocked (6 stimuli per block) and presented in pseudo-random. Each
condition (e.g. quantity) was rst presented with numerical stimuli, e.g. larger number? (or object names, e.g. larger object?) and followed by another block with object names (or
numerical stimuli) in a counterbalanced order. Presentation of blocks of the same task with both stimuli was followed by about 16-second rest period where subjects were asked to
maintain xation on a cross in the middle of the computer screen. Trials where the correct answer was the upper or the lower stimulus were presented in equal proportion. (For
interpretation of the references to colour in this gure legend, the reader is referred to the web version of this article.)

quantity categorical tasks, subjects were instructed to ignore the
colour of the stimuli, and to focus on the question presented above the
stimuli. In the perceptual task (colour-decision) they were asked to
choose the stimulus whose colour corresponded to the colour
indicated by the question above the stimuli. Subjects were instructed
to select the stimulus according to the colour of the ink and not
according to the colour of the object (e.g. they should not select red
just because the object name was for example strawberries or
tomatoes).
Participants were told that the number stimuli could indicate
either: i) quantities, ii) dates, or iii) times. The instructions for the
number stimuli were as follows: For the larger/smaller and more/less
questions, participants were told that numbers referred to an amount
and that they should choose the larger (or smaller) number in each
pair irrespective of the wording of the question (i.e. larger or
more). For summer/winter questions, participants were told that
each number indicated either a summer or a winter month in the
Northern hemisphere (all participants were British and raised in the
UK). They were told that summer months were June, July, and
August and winter moths were December, January and February
and that these months followed a day (131) separated with a dot
(12.07) rather than the more familiar slash (12/07). They were
instructed to select either the summer or the winter month in each
pair of stimuli depending on the question. For the working/sleeping
questions, participants were told that working or sleeping times were
in terms of a 24-hour clock; and that working times were between
 M. Cappelletti et al. / NeuroImage 46 (2009) 522529
8am and 6pm, and sleeping times were between 10pm and 7am.
Participants were discouraged from considering jobs that include
night shifts.
For object names, the instructions were the same as those for the
numbers except that the processing required for more/less ques-
tions was not the same as that required for larger/smaller questions.
Instead, during the more/less questions participants were instructed
to select the stimulus that was more (or less) numerous than the
other, for example socks vs thermos, stars vs moon, bed vs blanket,
deck chair vs swimming pool, snowakes vs snowman or cherries
vs melon. Prior to the experiment, participants underwent a practice
session in order to familiarize themselves with the task procedure.
TMS design
TMS was delivered over two brain sites and sham stimulation was
also given over the same regions. TMS was applied using a Magstim
Rapid Rate stimulator (Magstim Company Ltd., UK) and a focal 8-
shaped coil with each wing measuring 70 mm in diameter. TMS was
applied at 60% of the maximum output of the stimulator machine at
10 Hz frequency for 500 ms at the stimulus onset. A xed level of
stimulation was used as it has been shown that there is little
correlation of TMS stimulation thresholds between different brain
areas (Stewart et al., 2001). Sham stimulation was produced by
orienting the coil sideways, so that the sound and feeling of physical
contact with the head were similar to those produced by real
stimulation. For TMS, the stimulation coil rested tangentially on the
subject's scalp and the handle pointing posteriorly parallel to the
subject's midsagittal plane as calculated by frameless stereotaxy (Paus
and Wolforth, 1998). Target regions were identied by a previous fMRI
study run with the identical experimental design on different
subjects4 (Cappelletti et al., in press) and consisted of two regions
on the left and right IPS (MNI coordinates: 42, 40, 42 and 38,
44, 40 respectively). These areas corresponded to the brain regions
commonly activated by numbers and object names and identied by
the main effect of conceptual tasks (over quantity and non-quantity
for numbers and object names) relative to xation. To ensure that
these effects were not driven by one condition only, we used the
inclusive masking option in SPM to identify the main effect of
conceptual tasks relative to xation in areas that were activated by
both (i) conceptual tasks on numbers and (ii) conceptual tasks on
object names at p b 0.01. To control for any correlation between
conditions, a correction was made for non-sphericity using standard
SPM5 procedures.
In order to locate the site of stimulation accurately, we used a
frameless stereotactic system (Brainsight software, Rogue Research,
Montreal, Canada). This system allows the precise localization of
anatomical areas using the MRI images of the participants (e.g.
Dorward et al., 1999), and has been successfully used in previous
fMRI-guided TMS studies (e.g. Devlin et al., 2003). Prior to the
experiment, a high-resolution T1-weighted MRI scan of the brain of
each TMS participant was obtained in order to locate the two target
regions relative to external landmarks on the head. These landmarks
consisted of the bridge and tip of the nose and the tragus of the ears
that are visible on both the subject's MRI scan and on his/her head.
The 3D location of these landmarks was registered using an optical
tracking system. This system uses an infrared camera that can detect
reectors attached to the objects of interest (i.e. the coil and the
4
Functional image analysis was performed using Statistical Parametric Mapping
software (SPM5 software, Wellcome Trust Centre for Neuroimaging, London; http//
www.l.ion.ucl.ac.uk/spm). Scans were realigned, unwarped and spatially normalized
(Friston et al., 1995) to the Montreal Neurological Institute (MNI) standard space.
Functional images were then smoothed in the spatial domain with a Gaussian kernel of
6 mm FWHM to improve the signal to noise ratio. A high pass lter was used with a
cutoff period of 128s.
525
subject's head, Paus, 1999). The individual Brainsight localizations can
be seen in Supplementary Material.
Stimulation locations were calculated for each subject individually
using their structural scan and the coordinates obtained from the fMRI
analysis, similar to other TMS studies (e.g. Kalla et al., 2008; Sack et al.,
2009). Briey, this involved normalisation of the structural scan
against a standard template (using the FSL software package, FMRIB,
Oxford, UK) which produced both a normalised structural scan and a
mathematical description of the applied transformation. This trans-
formation was then reverse applied to the coordinates to be targeted,
giving their location in untransformed space. These were marked on
the scan in Brainsight, and the scalp locations overlying each served as
the targets for TMS. The precise localization of the target regions of a
representative participant as well as the results of the fMRI group
analysis is shown in Fig. 2.
For each area stimulated there were nine blocks, three for each
experimental condition (TMS, sham and no-stimulation respectively).
Each block consisted of 18 trials for each task (quantity, non-quantity
and colour decision) for each stimulus (numbers or object names);
therefore each subject performed a total of 54 trials per task per
stimulus in each experimental condition for a total of 324 trials for
each experimental condition. The order of the experimental condi-
tions and of the block pairs (i.e. each task with both stimuli) was
presented in pseudo-random order across subjects to avoid learning
and practice effects.
Participants
Six neurologically normal and native-English participants gave
their informed consent to participate in the TMS study (3 males, mean
age 22.2, range 2123). The study had been approved by the Ethics
Committee of the Institute of Neurology in London and was performed
in adherence with TMS safety guidelines (Wasserman, 1998).
Analysis of TMS data
The proportion of errors and the mean reaction times (RTs) were
calculated for each subject in each condition. Response times below
200 ms (i.e. anticipatory responses) and above 2 standard deviations
of the overall mean of each individual (i.e. delayed responses) were
excluded from the data set following a procedure which is common
practice in data analysis and in previous TMS studies (e.g. Martin et al.,
2004; Naeser et al., 2005). Furthermore, using this approach, we
decreased the likelihood that our results were driven by outliers as we
used tests that assume normal data distribution. By using the above
2SD criterion, 3.8% of the total number of responses were outliers and
were therefore excluded from the analysis. Bartlett's test was used to
check for homogeneity of variance and no data transformation was
necessary.
A t-test compared the sham and the non-stimulation conditions
across tasks to test whether there was any difference between them.
As no difference emerged (see analysis below), we compared the TMS
condition to sham rather than no-stimulation as sham is the most
similar to TMS in terms of the sound produced and physical contact
with the head.
A 4 2 3 analysis of variance (ANOVA) with condition [left and
right TMS-IPS and left and right sham-IPS], stimulus-type [numbers
and object names] and task [quantity, non-quantity categorical and
colour-decision] as within-subject factors was conducted on RTs of
correct answers. The analysis aimed at testing for stimulation effects
and at verifying whether these effects interacted with task and/or
stimulus. In addition, two-way ANOVAs were performed on RTs of
correct answers for each task with condition [left and right TMS-IPS
and sham] and stimulus-type [numbers and object names] as factors.
We also tested whether TMS differentially affected numbers that were
close vs far apart, i.e. distance effect. In the quantity tasks with
526 M. Cappelletti et al. / NeuroImage 46 (2009) 522529

Fig. 2. Location of the TMS target areas based on fMRI coordinates. The TMS target areas in the (A) left and (B) right IPS on the sagittal, coronal and axial view of the brain of a
representative participant (anti-clock wise from top right). Target regions were identied by a previous fMRI study run with the identical experimental design on different subjects
(Cappelletti et al., under review) and consisted of two regions on the left and right IPS (bottom right of A and B). These areas corresponded to the brain regions commonly activated
by numbers and object names and identied by the main effect of conceptual tasks (over quantity and non-quantity categorical for numbers and object names) relative to xation.
Precise location of TMS target regions was obtained using a frameless stereotactic system (Brainsight software, Rogue Research, Montreal, Canada).

numbers, the distance between 2 numbers was calculated by
subtracting one number from the other. In contrast, in the non-
quantity tasks with numbers, rather than the internal distance
between the 2 numbers, we calculated the distance between each
numerical stimulus in the pair and the next standard and we then
averaged these values for the numerical stimuli in each pair. Reported
results were corrected for non-sphericity using GreenhouseGeisser
correction.
When appropriate, we performed post-hoc paired comparisons
using student t-test and used Fisher LSD correction for multiple
comparisons. Statistical signicance refers to a two-tailed p value b0.05.
names), [F(2,10) = 9.31, p b 0.006]. Specically, performance in quan-
tity tasks with numbers (e.g. which is larger: 12.07 or 15.02?) was
signicantly impaired (slower performance) by stimulation over the
IPS relative to sham stimulation of the same areas, with longer RTs
following rTMS to either left [871.63 ms, 34.5 ms difference with
sham, t(5) = 9.56, p b 0.001] or right IPS [855.56 ms, 17.8 ms difference
with sham, t(5) = 4.91, p = 0.004]. Left and right IPS stimulation
differed signicantly, left IPS-TMS inducing a larger impairment than
right IPS-TMS [t(5) = 3.25, p b 0.02] (See Fig. 3). Moreover, left and
right IPS-TMS impairment was greater for comparisons of close
numbers relative to numbers far apart, i.e. distance effect [left IPS: t

Results

TMS did not affect the participants' accuracy in any of the tasks
[overall 4.6% of errors, no signicant difference between TMS and sham
across tasks and stimuli, p N 0.1]. This is consistent with similar
paradigms or tasks used in posterior parietal cortex TMS showing that
tasks performed at high levels of accuracy are likely to result in RT decit
rather than increase in error rate following TMS (Alexander et al., 2005;
Ashbridge et al., 1997; Cappelletti et al., 2007).
There was no signicant difference between sham and no-
stimulation across any of the tasks or stimuli used [t(5) = 0.45,
p = 0.67, n.s.].
The three-way ANOVA revealed a signicant main effect of task
[F(2,10) = 41.74, p b 0.001]. Moreover, the interaction of stimulus-
type and task [F(2,10) = 9.36, p b 0.005], and the three-way interac-
tion of condition, stimulus-type and task [F(6,30) = 2.36, p b 0.05]
were all signicant. Fig. 3. Performance impairments following TMS. Bars indicate the mean difference in
response time in ms in the left (black bars) and right IPS (grey bars) after TMS relative
Quantity task to sham stimulation to the same areas, for quantity, non-quantity categorical and colour
decisions with numbers and object names respectively. TMS conditions in which
response times signicantly differed from sham stimulation or between conditions are
A two-way ANOVA revealed a signicant interaction of condition indicated by an asterisk above the bars. Numbers below bars indicate the mean
(left and right TMS and sham) and stimulus-type (numbers and object differences in error rates between TMS and Sham in each experimental condition.
 M. Cappelletti et al. / NeuroImage 46 (2009) 522529
(5) = 2.593, p b 0.04 right IPS t(5) = 11.14, p b 0.001]. A distance effect
was also observed in both the sham [t(5) = 2.76, p b 0.03] and the non-
TMS conditions [t(5) = 10.99, p b 0.001].
Similarly, performance in quantity tasks with object names (e.g.
larger object: bikini or coat?) was signicantly delayed after
stimulation over the left [1083.2 ms, 27.03 ms difference with
sham, t(5) = 8.38, p b 0.001] or right IPS [1069.86 ms, 13.3 ms
difference with sham, t(5) = 9.56, p = 0.001] compared to sham
stimulation over the same areas. Left and right IPS stimulation
differed signicantly, left IPS-TMS again inducing a larger impairment
than the right IPS-TMS [t(5) = 3.49, p b 0.017].
Non-quantity categorical task
A two-way ANOVA revealed a signicant interaction of condition and
stimulus-type [F(2,10)= 7.28, p b 0.01]. In particular, performance in the
non-quantity categorical task with numerical stimuli (e.g. summer date:
23.07 or 15.02?) was signicantly delayed following stimulation to the
left [1118.57 ms, 24.9 ms difference with sham, t(5) = 6.42, p b 0.001] or
right IPS [1113.44 ms, 19.7 ms difference with sham, t(5) = 4.5,
p = 0.006] compared to sham stimulation of the same areas. Unlike
the quantity task, there was no signicant difference between left and
right IPS stimulation [t(5) = 0.76, p = 0.5].
Left and right IPS-TMS impairment was greater for comparisons of
close numbers when they referred to times, e.g. working time: 23.07 or
15.02? [left IPS: t(5) = 16.7, p b 0.001; right IPS t(5) = 6.5, p b 0.001],
but not when they referred to dates, e.g. summer month: 23.07 or
15.02? [left IPS: t(5) = 0.61, p = 0.56, n.s.; right IPS t(5) = 2.12,
p = 0.87, n.s.]. Similarly, a distance effect was also observed in both the
sham [t(5) = 5.7, p b 0.002] and the non-TMS conditions [t(5) = 11.94,
p b 0.001] for numbers referring to time but not to dates [t(5) = 0.8,
p = 0.4 and t(5) = 0.38, p = 0.72 for sham and non-TMS respectively].
Relative to sham, performance following IPS-TMS in the non-
quantity categorical task with object names (e.g. summer object:
bikini or coat?) remained unchanged following stimulation to the
left [926.5 ms, 7.37 ms difference with sham, t(5) = 2.53, p = 0.06]
and right IPS [929.4 ms, 4.4 ms difference with sham, t(5) = 8.3,
p = 0.45]. There was no signicant difference between left and right
IPS-TMS stimulation [t(5) = 0.07, p b 0.95].
Perceptual (colour-decision) task
A two-way ANOVA showed a non-signicant interaction of
condition and stimulus-type [F(2,10) = 3.53, p b 0.07].
Discussion
This study investigated the role of the left and right IPS regions in
processing quantity with numerical and non-numerical stimuli, and in
conceptual tasks with numbers requiring or not quantity manipula-
tion. We showed that quantity judgments with numerical and non-
numerical stimuli (e.g. choosing the larger of 12.07 and 15.02 or the
larger in size of bikini and coat) were equally affected following IPS-
TMS. Moreover, number conceptual processing either involving or not
involving quantity manipulation (e.g. choosing the larger of 12.07
and 15.02, or attributing a summer date to either 12.07 or 15.02)
was also equally impaired following IPS-TMS. Our results also showed
a distance effect in both quantity and non-quantity categorical
judgments. This is consistent with previous results indicating that
distance effect can be observed not just with any continuous quantity
dimension such as size of symbols (e.g. Cohen Kadosh et al., 2005;
Kaufmann et al., 2005; Pinel et al., 2004; Tang et al., 2006) and
luminance (Cohen Kadosh et al., 2005; Pinel et al., 2004), but also with
non-quantity dimensions such as letters of the alphabet (e.g. Fulbright
et al., 2003) or social status (Chiao et al., 2004). It may be suggested
that the distance effect induced by IPS-TMS is simply due to TMS
527
affecting the most difcult conditions, i.e. close vs distant numbers,
rather than interfering with numerical processing per se. We note,
however, that if this were the case TMS-induced distance effects
should have been observed in all numerical tasks. In contrast, we
showed that in the non-quantity categorical tasks with numbers
(summer/winter month; working/sleeping time), the distance effect
was affected by TMS in one of these tasks (working/sleeping time) but
not the other (summer/winter month), although they were the same
in terms of difculty at baseline. Some previous studies have already
shown that IPS-TMS may induce different degrees of interference on
the distance effect in numerical task of the same difculty level at
baseline (e.g. Cappelletti et al., 2007).
These ndings challenge past results by showing that the role of
IPS in quantity processing is stimulus-independent and its role in
numerical cognition is not specic for the conceptual task performed
with numbers. Two further critical conditions revealed the true
specicity of the IPS contribution to numerical processing: IPS-TMS
showed no impairment for perceptual decisions on numerical stimuli
in the absence of any conceptual task (i.e. colour judgment), nor for
conceptual decisions that did not involve quantity or number stimuli
(e.g. choosing a summer object between bikini and coat). From this,
we suggest that the simple presence of numbers is not sufcient for
the IPS to be critically involved, but conceptual-level operations are
also necessary. However, conceptual operations on their own are also
not sufcient to engage the IPS, but must be specically oriented to
numerical concepts.
One or more quantity mechanisms?
The issue of whether the quantity expressed by different stimuli is
processed in the same way or via distinct mechanisms is a long
standing issue. The seminal work of Moyer and Landauer (1967)
suggesting that the magnitude of all stimuli is processed in the same
way has received support at the theoretical level (e.g. Campbell, 1994;
Dehaene, 1997; McCloskey, 1992; Noel and Seron, 1993), and more
recently has motivated research into the neuronal correlates of
magnitude processing. These have been identied mainly in the
parietal regions (Cappelletti et al., 2007, in press; Cohen Kadosh et al.,
2005, 2007a,b; Dehaene et al., 2003; Pinel et al., 2001, 2004; Piazza et
al., 2004, 2007), and a recent theoretical proposal has suggested these
areas as the locus of stimulus-independent magnitude processing
according to a common metric (Walsh, 2003). However, dissociations
between quantity processing of Arabic numbers and of other
magnitude dimensions such as the size of physical stimuli have
been reported in neuropsychological patients (e.g. Cipolotti et al.,
1991; Lemer et al., 2003) and not all functional imaging studies have
shown consistent results as to whether the bilateral IPS is involved in
stimulus-independent magnitude processing (e.g. Cohen Kadosh et al.
2007a,b; Pinel et al., 2004) or whether the left or the right IPS is
differentially involved (e.g. Castelli et al., 2006; Cohen Kadosh et al.,
2007a,b; Pinel et al., 2004 for right IPS activation; Fias et al., 2003;
Cohen Kadosh et al., 2005 for left IPS activation). One factor that may
account for these discrepancies is the way in which magnitudes
change luminance is continuous, for example, while number
changes are discrete (e.g. Cohen Kadosh et al., 2007a,b; Pinel et al.,
2004) irrespective of the underlying representation. Another factor, as
formalised by Stevens (1975), is the way in which magnitude changes
are experienced weight differences are experienced as more than
or less than another weight, while frequency and saturation changes
modify the categorical perception of pitch and colour respectively. A
third reason for the discrepancy of previous results is the possible
confound between numerical processing and response selection, both
engaging the parietal lobes (e.g. Cappelletti et al., in press; Goebel et
al., 2005). However, it was beyond the purpose of this study to address
this issue. A fourth difference is that numerical magnitude judgments
(i.e. is 65 bigger than 55?) may necessitate the retrieval of learnt
528 M. Cappelletti et al. / NeuroImage 46 (2009) 522529
information while physical size and luminance decisions may be
based on visual stimulus properties rather than memory-related
strategies. Our study avoided this confound by using magnitude
dimensions with similar processing requirements: choosing the larger
of two Arabic numbers (e.g. 12.07 vs 15.02) or of two objects (e.g.
bikini vs coat) relies in both cases on the retrieval of associated
information which is not contained in the stimuli. By using this novel
experimental manipulation we have provided new evidence suggest-
ing that the bilateral IPS is necessary for processing magnitude
expressed by different stimuli when these are based on similar
cognitive resources.
Quantity and non-quantity conceptual number processing in the IPS
We have also shown that the IPS regions are equally involved in
quantity processing and in other conceptual tasks with numbers that
do not require quantity manipulation, i.e. categorical tasks. The idea
that number semantics can be extended to include non-quantity
conceptual operations on numbers has been proposed by some
previous theoretical accounts (Butterworth, 1999; Dehaene, 1997).
However, this proposal has not yet been systematically investigated as
most of the previous studies focused on quantity processing. By
showing that the parietal regions are involved not only in processing
the quantitative features of numbers but also in other non-
quantitative conceptual manipulations our results support the idea
that number semantics in the IPS is not limited to the processing of
magnitude but can be extended to include other numerical conceptual
processing (Butterworth, 1999; Dehaene, 1997).
Our ndings show that the involvement of the IPS regions in
number processing is modulated by the nature of the number task
performed: responses were delayed only when subjects were
performing quantitative and non-quantitative conceptual judgments
but not perceptual judgments with numerical stimuli. These results are
consistent with proposals that the parietal areas are engaged in the
conceptual representation of numbers (Dehaene, 1997; Dehaene and
Cohen, 1995), and with a recent fMRI study using the same
experimental design showing the involvement of the IPS only in
conceptual (quantity and non-quantity) but not in perceptual number
tasks (Cappelletti et al., in press). How do these results combine with
the most common view that number processing is so automatic that
the simple presentation of numbers activates a sense of magnitude
(e.g. Ansari et al., 2006; Dehaene and Akhavein, 1995; Eger et al., 2003;
Thioux et al., 2005)? One possibility is that a magnitude representation
might have been automatically activated in the colour task but TMS
had no effect on colour processing, therefore no disruption by TMS was
seen in that task. Another possibility is that different types of
numerical representations may be triggered by different types of
task requirements (Fischer and Rottmann, 2005; Shaki and Petrusic,
2005). In our study these numerical representations may differ in the
conceptual (quantity and non-quantity) relative to perceptual tasks.
Moreover, these representations may rely on different brain regions or
may pose different task requirements on the same regions, therefore
resulting in different TMS effects. As such, our results suggest a
renement of the idea of task-independent automatic magnitude
activation and are in keeping with proposals suggesting that
magnitude representations may interact with task requirements.
It still remains to be explained why some previous imaging studies
showed similar parietal activation in numerical tasks whether they
required quantity processing or not (e.g. Gbel et al., 2005; Eger et al.,
2003; Shuman and Kanwisher, 2004; Tang et al., 2006; Thioux et al.,
2005). For instance, Thioux et al. (2005) found no difference in IPS
activation when contrasting number comparison to a task where
subjects have to decide if numbers were written in plain characters or
not. Similarly, Eger et al. (2003) reported IPS activation in a simple
number detection task (i.e. indicate when a stimulus is a number) not
requiring quantity manipulation. The simplest possibility is that not
all activations reported in fMRI studies are indicative of an area being
necessary for normal performance of that task. There are two reasons
for this: rst there is some 'redundancy' in perceptual and cognitive
systems (e.g. Price and Friston, 2002); second, fMRI activations are the
result of a particular subtraction and thus indicate activity elicited by
one kind of stimulus/task relative to another stimulus/task. Necessity
of an area to those stimuli/tasks can only be ascertained by
interference. Another reason may lie in the type of tasks used or the
response selection demands of the task. For instance, tasks such as
stimulus detection (Eger et al., 2003) activate the parietal lobes
because they require conceptual processing in the form of identity
recognition (i.e. when distinguishing a number from another
stimulus). Moreover, the perceptual tasks used in some of these
studies (e.g. Gbel et al., 2005; Tang et al., 2006; Thioux et al., 2005)
involved visual search processes that have been shown to activate the
superior parietal lobes (e.g. Coull et al., 2003; Pollmann et al., 2003)
whereas a colour detection task placed minimal demands on visuo-
spatial mechanisms because it could be based on any unit of the string.
Our ndings that the left and right IPS are critically involved in
conceptual but not perceptual processing of numbers is also in
keeping with some previous studies based on fMRI adaptation
paradigms (fMRIA, e.g. Cohen Kadosh et al., 2007a,b; Cantlon et al.,
2006; Piazza et al., 2007). These studies showed that IPS adaptation
(i.e. in the absence of any explicit task) is modulated by numerical
magnitude but not by changes in the colour of numerical stimuli. Our
results further expand this evidence by showing that the IPS plays a
critical role in number processing and that this is specic for
conceptual (quantity and non-quantity) but not perceptual number
processing.
In conclusion, we have shown that the role of the IPS in numerical
cognition is dependent on a combination of stimuli and task
operations. Specically, the left and right IPS are critical for: (1)
performing quantity and non-quantity conceptual operations with
numbers; and for (2) quantity processing irrespective of the stimuli
used, i.e. numbers or object names. However, this study shows that
the IPS is not critically involved in perceptual decisions based on
numbers or in conceptual tasks that do not involve quantity or
numerical stimuli.
Acknowledgments
This work was supported by the Wellcome Trust and the Royal
Society (MC), and by the MRC (VW and NM).
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.neuroimage.2009.02.016.
References
Alexander, I., Cowey, A., Walsh, V., 2005. The right parietal cortex and time perception:
back to Critchley and the Zeitraffer phenomenon. Cogn. Neuropsychol. 22,
1030610315.
Andres, M., Seron, X., Olivier, E., 2005. Hemispheric lateralization of number
comparison. Cogn. Brain Res. 25, 283290.
Ansari, D., Dhital, B., Siong, S.C., 2006. Parametric effects of numerical distance on the
intraparietal sulcus during passive viewing of rapid numerosity changes. Brain Res.
1067, 181188.
Ashbridge, E., Walsh, V., Cowey, A., 1997. Temporal aspects of visual search studies by
transcranial magnetic stimulation. Neuropsychologia 35, 11211131.
Butterworth, B., 1999. The Mathematical Brain, McMillan.
Campbell, J.I.D., 1994. Architectures for numerical cognition. Cognition 53, 144.
Cantlon, J.F., Brannon, E.M., Carter, E.J., Pelphrey, K.A., 2006. Functional imaging of
numerical processing in adults and 4-y-old children. PLoS Biol. 4, e125.
Cappelletti, M., Barth, H., Fregni, F., Pascual Leone, A., Spelke, E., 2007. rTMS over the left
and the right intraparietal sulcus disrupts discrete and continuous quantity
processing. Exp. Brain Res. 179, 631642.
Cappelletti, M., Jansari, A., Butterworth, B., Kopelman, M.D., 2008. A case of selective
impairment of encyclopedic numerical knowledge or when December 25th is no
longer Christmas day, but 20 + 5 is still 25. Cortex 44, 325336.
 M. Cappelletti et al. / NeuroImage 46 (2009) 522529
Cappelletti, M., Lee, H.L., Freeman, E.D., Price, C.J., in press. The role of the right and left
parietal lobes in the conceptual processing of numbers.
Castelli, F., Glaser, D.E., Butterworth, B., 2006. Discrete and analogue quantity processing
in the parietal lobe: a functional MRI study. Proc. Natl. Acad. Sci. 103, 46934698.
Cipolotti, L., Butterworth, B., Denes, G., 1991. A specic decit for numbers in a case of
dense acalculia. Brain 114, 26192637.
Chiao, J.Y., Bordeaux, A.R., Ambady, N., 2004. Mental representations of social status.
Cognition 93, B49B57.
Cohen Kadosh, R., Henik, A., Rubinsten, O., Mohr, H., Dori, H., van de Ven, V., et al., 2005.
Are numbers special? The comparison systems of the human brain investigated by
fMRI. Neuropsychologia 43, 12381248.
Cohen Kadosh, R., Cohen Kadosh, K., Schuhmann, T., Kaas, A., Goebel, R., Henik, A., Sack,
A.T., 2007a. Virtual dyscalculia induced by parietal-lobe TMS impairs automatic
magnitude processing. Curr. Biol. 17, 15.
Cohen Kadosh, R., Cohen Kadosh, K., Kaas, A., Henik, A., Goebel, R., 2007b. Notation-
dependent and independent representations of numbers in the parietal lobes.
Neuron 53, 307314.
Coull, J.T., Walsh, V., Frith, C.D., Nobre, A.C., 2003. Distinct neural substrates for visual
search amongst spatial versus temporal distracters. Cogn. Brain Res. 17, 368379.
Dehaene, S., 1997. The Number Sense. Oxford University Press, Penguin press, New York,
Cambridge (UK).
Dehaene, S., Akhavein, R., 1995. Attention, automaticity, and levels of representation in
number processing. J. Exper. Psychol., Learn., Mem., Cogn. 21, 314326.
Dehaene, S., Cohen, L., 1991. Two mental calculation systems: a case study of severe
acalculia with preserved approximation. Neuropsychologia 29, 10451074.
Dehaene, S., Cohen, L., 1995. Towards an anatomical and functional model of number
processing. Math. Cogn. 1, 83120.
Dehaene, S., Piazza, M., Pinel, P., Cohen, L., 2003. Three parietal circuits for number
processing. Cogn. Neuropsychol. 20, 487506.
Devlin, J.T., Matthews, P.M., Rushworth, M.S.F., 2003. Semantic processing in the left
inferior prefrontal cortex: a combined functional magnetic resonance imaging and
transcranial magnetic stimulation study. J. Cogn. Neurosci. 15, 7184.
Dorward, N.L., Alberti, O., Palmer, J.D., Kitchen, N.D., Thomas, D.G.T., 1999. Accuracy of
true frameless stereotaxy: in vivo measurement and laboratory phantom studies.
J. Neurosurg. 90, 160168.
Eger, E., Sterzer, P., Russ, M.O., Giraud, A.L., Kleinschmidt, A., 2003. A supramodal
number representation in human intraparietal cortex. Neuron 37, 719725.
Fias, W., Lammertyn, J., Reynvoet, B., Dupont, P., Orban, G.A., 2003. Parietal
representation of symbolic and non-symbolic magnitude. J. Cogn. Neurosci. 15,
4756.
Fischer, M.H., Rottmann, J., 2005. Do negative numbers have a place on the mental
number line? Psychol. Sci. 47, 2232.
Friston, K.J., 1995. Neuronal transients. Proc. Biol. Sci. 22, 401405.
Fulbright, R.K., Manson, S.C., Skudlarski, P., Lacadie, C.M., Gore, J.C., 2003. Quantity
determination and the distance effect with letters, numbers, and shapes: a
functional MR imaging study of number processing. Am. J. Neuroradiol. 24, 193200.
Gevers, W., Reynvoet, B., Fias, W., 2003. The mental representation of ordinal sequences
is spatially organized. Cognition 87, B87B95.
Gbel, S.M., Johansen-Berg, H., Behrens, T.E., Rushworth, M.F.S., 2005. Response-
selection related parietal activation during number comparison. J. Cogn. Neurosci.
16, 15361551.
Jacob, S.N., Nieder, A., 2007. The ABC of cardinal and ordinal number representations.
Trends Cogn. Sci. 12, 4143.
Johnson, D.M., 1939. Condence and speed in the two-category judgment. Arch.
Psychol. 241, 150.
Kalla, R., Muggleton, N.G., Juan, C.H., Cowey, A., Walsh, V., 2008. The timing of the
involvement of the frontal eye elds and posterior parietal cortex in visual search.
NeuroReport 2 (19), 10671071.
Kaufmann, L., Koppelstaetter, F., Delazer, M., Siedentopf, C., Rhomberg, P., Golaszewski,
S., Felber, S., Ischebeck, A., 2004. Neural correlates of distance and congruity effects
in a numerical Stroop task: an event-related fMRI study. NeuroImage 25 (3),
888898.
Lemer, C., Dehaene, S., Spelke, E., Cohen, L., 2003. Approximate quantities and exact
number words: dissociable systems. Neuropsychologia 41, 19421958.
Lochy, A., Van Turennout, M., 2005. Semantics of numbers: a cortical shift? Poster
presented at OHBM, Toronto.
Martin, P., Naeser, M.A., Theoret, H., Tormos, J.M., Nicholas, M., Kurland, J., Fregni, F.,
Seekins, H., Doron, K., Pascual-Leone, A., 2004. Transcranial magnetic stimulation as
a complementary treatment for aphasia. Semin. Speech Language 25, 181191.
McCloskey, M., 1992. Cognitive mechanisms in numerical processing: evidence from
acquired dyscalculia. Cognition 44, 107157.
Moyer, R.S., 1973. Comparing objects in memory: evidence suggesting an internal
psychophysics. Percept. Psychophys. 7, 82138.
529
Moyer, R.S., Landauer, T.K., 1967. The time required for judgements of numerical
inequality. Nature 215, 15191520.
Molko, N., Cachia, A., Riviere, D., Mangin, J.F., Bruandet, M., Le Bihan, D., et al., 2003.
Functional and structural alterations of the intraparietal sulcus in a developmental
dyscalculia of genetic origin. Neuron 40 (4), 847858.
Naeser, M.A., Martin, P.I., Nicholas, M., Baker, E.H., Seekins, H., Helm-Estabrooks, N.,
Cayer-Meade, C., Kobayashi, M., Theoret, H., Fregni, F., Tormos, J.M., Kurland, J.,
Doron, K.W., Pascual-Leone, A., 2005. Improved picture naming in chronic aphasia
after TMS to part of right Broca's area: an open-protocol study. Brain Lang. 93,
95105.
Nieder, A., 2005. Counting on neurons: the neurobiology of numerical competence. Nat.
Neurosci. Rev. 6, 177190.
Noel, M.P., Seron, X., 1993. Arabic number reading decit: a single-case study or when
236 is read (2306) and judged superior to 1258. Cogn. Neuropsychol. 10, 317339.
Ono, M., Kubik, S., Abernathey, C.D., 1990. Atlas of the Cerebral Sulci. Thieme.
Paus, T., 1999. Imaging the brain before, during, and after transcranial magnetic
stimulation. Neuropsychologia 37, 219224.
Paus, T., Wolforth, M., 1998. Brainsight, Rogue Research, Montreal, Quebec, Canada.
Hum. Brain Mapp. 6, 399402.
Piazza, M., Izard, V., Pinel, P., Le Bihan, D., Dehaene, S., 2004. Tuning curves for
approximate numerosity in the human intraparietal sulcus. Neuron 44, 547555.
Piazza, M., Pinel, P., Le Bihan, D., Dehaene, S., 2007. A magnitude code common to
numerosities and number symbols in human intraparietal cortex. Neuron 53,
293305.
Pinel, P., Dehaene, S., Riviere, D., LeBihan, D., 2001. Modulation of parietal activation by
semantic distance in a number comparison task. NeuroImage 14, 10131026.
Pinel, P., Piazza, M., Le Bihan, D., Dehaene, S., 2004. Distributed and overlapping cerebral
representations of number, size, and luminance during comparative judgments.
Neuron 41, 983993.
Polk, T.A., Reed, C.L., Keenan, J.M., Hogarth, P., Anderson, C.A., 2001. A dissociation
between symbolic number knowledge and analogue magnitude information. Brain
Cogn. 47, 545563.
Pollmann, S., Weidner, R., Humphreys, G.W., Olivers, C.N., Muller, K., Lohmann, G.,
Wiggins, C.J., Watson, D.G., 2003. Separating distracter rejection and target
detection in posterior parietal cortexan event-related fMRI study of visual
marking. NeuroImage 18, 310323.
Price, G.R., Friston, K.J., 2002. Degeneracy and Cognitive anatomy. Trends in Cognition
Science 6, 416421.
Price, G.R., Holloway, I., Rsnen, P., Vesterinen, M., Ansari, D., 2008. Impaired parietal
magnitude processing in developmental dyscalculia. Curr. Biol. 17 (24),
R1042R1043.
Rotzer, S., Kucian, K., Martin, E., Aster, M.v., Klaver, P., Loenneker, T., 2008. Optimized
voxel-based morphometry in children with developmental dyscalculia. Neuro-
Image 39, 417422.
Sack, A.T., Cohen Kadosh, R., Schuhmann, T., Moerel, M., Walsh, V., Goebel, R., 2009.
Optimizing functional accuracy of TMS in cognitive studies: a comparison of
methods. J. Cogn. Neurosci. 21, 207221.
Shaki, S., Petrusic, W.M., 2005. On the mental representation of negative numbers:
context-dependent SNARC effects with comparative judgments. Psychon. Bull. Rev.
12, 931937.
Shuman, M., Kanwisher, N., 2004. Numerical magnitude in the human parietal lobe:
tests of representational generality and domain specicity. Neuron 44, 557569 28.
Stevens, S., 1975. Psychophysics: Introduction to its Perceptual, Neural, and Social
Prospects. John Wiley & Sons, New York.
Stewart, L.M., Walsh, V., Rothwell, J.C., 2001. Motor and phosphene thresholds: a
transcranial magnetic stimulation correlation study. Neuropsychologia 39,
415419.
Tang, J., Critchley, H.D., Glaser, D.E., Dolan, R.J., Butterworth, B., 2006. Imaging
informational conict: an fMRI study of numerical stroop. J. Cogn. Neurosci. 18,
20492062.
Thioux, M., Pesenti, M., Costes, N., De Volder, A., Seron, X., 2005. Task independent
semantic activation for numbers and animals. Cogn. Brain Res. 24, 284290.
Turconi, E., Jemel, B., Rossior, B., Seron, X., 2004. Electrophysiological evidence for
differential processing of numerical quantity and order in humans. Cogn. Brain Res.
21, 2238.
Turconi, E., Campbel, J.I.D., Seron, X., 2006. Numerical order and quantity processing in
number comparison. Cognition 98 (3), 273285.
Walsh, V., 2003. A theory of magnitude: common cortical metrics of time, space and
quantity. Trends Cogn. Sci. 7, 483488.
Wasserman, E.M., 1998. Electroencelophagr. Clin. Neurophysiol. 108, 116.
Wiese, H., 2003. Numbers, Language and the Human Mind. Cambridge University Press.
Zamarian, L., Egger, C., Delazer, M., 2007. The mental representation of ordered
sequences in visual neglect. Cortex 43, 542550.