Neuroscience and Biobehavioral Reviews 32 (2008) 13611372

Contents lists available at ScienceDirect

Neuroscience and Biobehavioral Reviews

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

Review

Visualizing numbers in the minds eye: The role of visuo-spatial processes

in numerical abilities
Maria Dolores de Hevia a,b,*, Giuseppe Vallar a,c, Luisa Girelli a
a

Universita` degli Studi di Milano-Bicocca, Italy

Harvard University, USA
c
Laboratorio di Neuropsicologia, IRCCS Istituto Auxologico Italiano, Milano, Italy
b

A R T I C L E I N F O

A B S T R A C T

Keywords:
Numerical cognition
Arithmetical abilities
Visuo-spatial representation of numbers

In the study of numerical and arithmetical abilities, there is compelling evidence demonstrating that
number and space representations are connected to one another. Historically the rst source of support
came more than a century ago, when Galtons investigations on mental imagery suggested that the
internal representation of numbers may evoke a stable, linear space. In the past few decades, empirical
evidence lent further support to the hypothesis that numerical representation is spatially coded into a
non-verbal mental number line, which in turn lead to considering this representation as the core of
number meaning. Visuo-spatial processing is intuitively involved in various aspects of number
processing and calculation: For instance, the meaning of a digit in a multi-digit number is coded following
spatial information, given its association to its relative position within the number; similarly, to solve a
complex written multiplication one has to know the correct location of the intermediate results. In this
review behavioral, neuropsychological, and neuroimaging data concerning the close relationship
between numerical abilities and visuo-spatial processes are considered.
2008 Elsevier Ltd. All rights reserved.

Contents
1.

2.

3.

Numbers and visuo-spatial processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1.1.
The mental number line hypothesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.2.
Experimental paradigms and effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.3.
Neuropsychological evidence. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.4.
Neuro-anatomical basis: lesion and neuroimaging data. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Calculation and visuo-spatial processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.1.
The hypothesis of a visuo-spatial medium for arithmetic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.2.
Experimental paradigms and effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.3.
Neuropsychological evidence. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.4.
Neural correlates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Acknowledgements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

In the study of numerical and arithmetical abilities, there is

compelling evidence demonstrating that number and space

* Corresponding author at: Harvard University, Department of Psychology, 33

Kirkland Street, Cambridge, MA 02138, United States.
E-mail address: dehevia@wjh.harvard.edu (M.D. de Hevia).
0149-7634/$ see front matter 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.neubiorev.2008.05.015

1362
1362
1362
1365
1366
1367
1367
1368
1369
1369
1370
1370
1370

representations are connected to one another (for recent reviews,

see de Hevia et al., 2006b; Fias and Fischer, 2005; Hubbard et al.,
2005). Historically, the rst source of support came more than a
century ago, when Galtons investigations on mental imagery
suggested that the internal representation of numbers may evoke a
stable, linear space (Galton, 1880). The so-called number forms
reect a series of visuo-spatial properties associated with

1362

M.D. de Hevia et al. / Neuroscience and Biobehavioral Reviews 32 (2008) 13611372

Fig. 1. A number form. Illustration of the mental image evoked by a subject when
thinking about numbers (from Galton, 1880).

numerical information, such as spatial orientation, color, and

brightness, which give rise to particular congurations occupied by
numbers (Galton, 1880) (see Fig. 1). Although originally anecdotal,
the intuition that the mental representation of numbers contains a
series of visuo-spatial properties has later found systematic
support (Seron et al., 1992). It has, at present, a major impact
on cognitive models of numerical processing (Dehaene, 1992;
Dehaene et al., 2003). In fact, some individuals report the
deployment of a visuo-spatial internal space when they represent
numerical information, and perform mental arithmetic. Interestingly, they claim that these mental images are not the product of
mathematical instruction at school, reecting instead the spontaneous process of visualizing numbers (Sagiv et al., 2006; Seron
et al., 1992).
In the past few decades, empirical evidence lent further support
to the hypothesis that numerical representations are spatially
coded into a non-verbal mental number line. This view, in turn,
leads to considering such a representation the core of number
meaning (Dehaene, 1992). Visuo-spatial processing is intuitively
involved in various aspects of number processing and calculation:
For instance, the meaning of a digit in a multi-digit number is
coded following spatial information, given its association to its
relative position within the number. Similarly, in order to solve a
complex written multiplication, one has to know the correct
location of the intermediate results.
In this review the behavioral, neuropsychological, and neuroimaging evidence concerning the close relationship between
numerical abilities and visuo-spatial processes is considered.
1. Numbers and visuo-spatial processing
1.1. The mental number line hypothesis
There is wide evidence supporting the view that the representation of numerical magnitude is mapped onto an analogue
mental number line. Early reports suggested that the representation of numbers is analogue, with subjects adding portions of the
number line when performing a mental addition (Moyer and
Landauer, 1967; Restle, 1970). Nowadays, the hypothesis of a
spatial numerical representation is central to models of numerical
cognition (in particular, the Triple Code model, Dehaene, 1992;
Dehaene et al., 2003), which postulate that among the verbal
auditory (e.g., twenty-seven), and the Arabic-visual (e.g., 27)

codes, in which numbers can be mentally represented, subjects

also activate an analogue-quantity code, which contains the
semantic information of a number. At this level, the numerical
representation is non-verbal, and only approximate, consisting of
variable distributions of activation along the spatial mental
number line (Dehaene et al., 2003).
The mental number line, in the form of a mental continuum
logarithmically compressed (see Gallistel and Gelman, 2000, 1992,
for the view of an analogue linear representation of magnitude),
provides the information of numerical magnitude (Dehaene et al.,
2003; Restle, 1970). This proposal was initially supported by two
psychophysical effects that characterize the comparison judgment
between two numerical values. First, the distance effect (i.e.,
subjects are faster and more accurate in comparing two numbers
the farther apart they are), and second, the size effect (i.e., by
keeping the numerical distance between two numbers constant,
the performance is faster, and more accurate, with smaller than
with larger numbers) (Moyer and Landauer, 1967). These effects
characterize judgments of inequality within other continuous
dimensions, such as brightness, or the length of a line, suggesting
that the internal representation of numerical magnitude is not
digital, but analogue (Moyer and Landauer, 1967), and that, as a
consequence, the subjective difference between two numerical
magnitudes obeys the WeberFechner law (Dehaene, 2003; Nieder
and Miller, 2003).
The distance and size effects have been described also with nonsymbolic numbers, such as dot patterns, both in adults and
children (Buckley and Gillman, 1974; Temple and Posner, 1998),
suggesting that the number line contains an abstract representation of magnitude. Moreover, there is evidence that the numerical
discrimination abilities present in non-verbal animals (Gallistel
and Gelman, 2000, for a review), and in pre-verbal human infants,
as young as 6-month-old (Feigenson et al., 2004, for a review), are
governed by the same properties (i.e., the success of discrimination
depends on the ratio between the two quantities). These ndings
support the hypothesis that the number line is a natural endowed
capacity of representing and discriminating numbers (Spelke and
Dehaene, 1999).
Accordingly, the mental number line representation provides
the capacity of understanding a numerical quantity in an
approximate way. Furthermore, a growing body of empirical
ndings suggests that this internal code consists of an oriented
spatial medium, in such a way that increasing numerical
magnitudes are represented in ascending order, and, as a
consequence, each number is associated with a specic spatial
location (Dehaene et al., 1993).
1.2. Experimental paradigms and effects
The spatial orientation of the representational continuum
within the hypothesis of the mental number line representation
is qualied as left-to-right oriented by the observation of a
stimulusresponse compatibility effect that emerges in numerical
classication tasks: The SNARC effect (Spatial Numerical Association of Response Codes) refers to the fact that subjects respond faster
to smaller numbers with the left than with the right hand, and to
larger numbers with the right hand (Dehaene et al., 1993; Fias
et al., 1996) (Fig. 2). This phenomenon suggests that the numerical
representation accessed in these tasks contains spatial information. The SNARC effect may be regarded as a sort of Simon effect
(Gevers et al., 2005; Keus and Schwarz, 2005; but see Mapelli et al.,
2003), in the form of a dimensional overlap between the (implicit)
relative position of the magnitude on the mental number line (in
representational space), and the location of the manual response to
the left and right sides (in physical egocentric space).

M.D. de Hevia et al. / Neuroscience and Biobehavioral Reviews 32 (2008) 13611372

Fig. 2. (A) The classical experimental setting in which the SNARC effect is observed.
In a numerical classication task, smaller magnitude numbers elicit left-sided
responses, larger magnitude numbers right-sided responses. (B) Expected
differences in RT (dRT) between right, and left hand responses, as a function of
the digit to be classied.

The implicit association of numbers to spatial locations can be

demonstrated in subjects who do not consciously experience the
phenomenon. Importantly, the SNARC effect may be triggered
automatically, being present whether or not the numerical
magnitude is intentionally processed: It has been obtained when
participants simply need to categorize numbers by their identity
(Dehaene and Akhavein, 1995; Zhou et al., 2008); it also emerges in
parity judgment tasks, where subjects are required to classify
numbers as even or odd (Dehaene et al., 1993; Fias et al., 1996; Ito
and Hatta, 2004); in phoneme detection tasks, where numbers are
classied as containing a specic phoneme in the corresponding
number word (Fias et al., 1996); or in orientation judgments,
where subjects need to process the spatial orientation of a gure
superimposed on a single Arabic digit (Fias et al., 2001).
The SNARC effect has been systematically replicated in a variety
of conditions: It can be obtained for different notations (Arabic vs.
verbal; Fias, 2001), and with non-symbolic number (dice dot
patterns; Nuerk et al., 2005). The effect has also been described for
different modalities of presentation (visual vs. auditory, see Nuerk
et al., 2005), and it extends to two-digit numbers (Brysbaert, 1995;
Dehaene et al., 1990; Zhou et al., 2008). Overall, these observations
support the view that the mental number line is to be conceived
as an amodal, and abstract visuo-spatial representation of the
numerical magnitude.
The mapping of numbers into specic spatial locations has been
observed to be a exible and context-dependent process. In
particular, it has been shown that the association of a number with
a spatial location is not absolute, depending instead on the relative
magnitude; thus, the same number considered in different
numerical intervals would be associated with different spatial
positions (e.g., the number 4 in intervals [04] vs. [49], Dehaene
et al., 1993; Fias et al., 1996). Furthermore, the advantage of
responding to small numbers with the left hand and to large
numbers with the right hand is related neither to handedness, nor
to hemispheric lateralization (Dehaene et al., 1993).
The direction along which magnitudes are arranged into the
horizontal dimension, however, seems to be shaped by reading

1363

habits: left-to-right vs. right-to-left (Dehaene et al., 1993; Zebian,

2005). In fact, in the work by Dehaene et al. (1993), Iranian
participants show a weaker SNARC effect than French participants,
the former approaching a signicant reversed SNARC effect.
Similarly, in Lebanese participants, comparing two visually
presented Arabic digits, an advantage when numbers are
presented oriented from right-to-left (e.g., 9 1), in contrast to a
left-to-right orientation (e.g., 1 9) has been found (Zebian, 2005).
However, contrasting evidence is also on record: Japanese
participants, whose reading habits predict an association between
a top location with a small number and a low location with a large
number, exhibit the reverse pattern of association during a parity
judgment task (Ito and Hatta, 2004). Moreover, the preferred
spatial orientation might also depend on the specic numerical
notation in which numbers are presented. For instance, native
speakers of Chinese showed a horizontal left-to-right advantage
for Arabic digits, whereas for numbers written with Chinese
characters, which appear predominantly with a vertical top-tobottom directionality, a top-small and bottom-large association
was found to facilitate processing (Hung et al., 2008).
The evidence that the specic orientation of this numerical
continuum is shaped by cultural constraints suggests that the
spatial properties are acquired during school years, with the
acquisition of reading and writing habits as the main determinants.
In fact, the SNARC effect, assessed by the parity judgment task, has
been found to emerge in children after 3rd grade (910 year-old)
(Berch et al., 1999). Notwithstanding the apparent late appearance
of spatial properties in numerical processing, as indexed by the
SNARC effect, it should be noted that the automatic access to
magnitude information from Arabic symbols, as assessed by a
numerical-Stroop paradigm, emerges from age 10 (Girelli et al.,
2000). It is thus likely that the spatial organization of numerical
magnitude may be present even earlier in development. Importantly, once the spatial mapping of numbers has been established,
the spatial information is implicit in the processing of a number
(Dehaene et al., 1993; Fias, 2001; Fias et al., 1996). One may
suggest that the initial mapping of numbers onto space might be
partially triggered by the exposure to measurement devices, such
as rulers, or that it critically depends on extensive experience with
number in the visual modality (Cooper, 1984; Simon, 1997). This
view has been recently challenged in a study with congenitally or
early blind participants, who show a SNARC effect as strong as a
sighted control group, in a series of bimanual classication tasks of
verbal numbers presented auditorily (Castronovo and Seron,
2007).
As discussed earlier, the SNARC effect has typically been found
in numerical classication tasks requiring bimanual responses,
leading to an association between smaller numbers with the left
side and larger numbers with the right side of space (e.g., Dehaene
et al., 1993; Dehaene et al., 1990; Fias, 2001; Fias et al., 1996).
However, although the spatial arrangement might have a left-toright orientation by default, mainly determined by scanning habits
(Chokron and Imbert, 1993), the assignment of a spatial code to a
number has proved to be a exible phenomenon. For instance,
when numbers are visualized on a clock face, a reversed SNARC
effect can be obtained: If subjects are simply asked to classify
numbers as hours of the day coming before or after six oclock,
faster responses to small numbers with the right hand, and to large
numbers with the left hand, can be obtained (Bachtold et al., 1998).
Although it is worth noticing that in this study participants were
explicitly required to evoke the corresponding mental image of an
analogical clock, this nding suggests that the spontaneous
mapping of numbers into an internal visuo-spatial representation
may be subject to the specic context into which numbers are
manipulated. Additional evidence supports the view that the left-

1364

M.D. de Hevia et al. / Neuroscience and Biobehavioral Reviews 32 (2008) 13611372

to-right oriented mental number line is not the only representation which leads to visuo-spatial associations. In fact, when
subjects use all their ngers in a task that requires to identify
Arabic digits (each digit is assigned to a unique nger), they show
an advantage for small numbers responded to with the right hand,
provided that the specic digit-nger mapping imposed in the task
mirrors the personal nger counting strategies (Di Luca et al.,
2006). These studies provide evidence of further spatial frames
onto which numbers might be mapped, such as a nger counting
map, which may even overcome the canonical left-to-right
orientation of the mental number line.
In addition, the spatial representation of the mental number
line is not conned to the horizontal dimension. In fact, the
specic mapping of the responses has proved to be related to the
emergence of spatialnumerical compatibility effects. When
participants are required to perform the numerical classication
task with key responses arranged in a top-down fashion, a vertical
SNARC effect is observed, where smaller (larger) numbers are
preferentially associated to low (high) locations (Ito and Hatta,
2004; Schwarz and Keus, 2004). Moreover, the response advantage
driven by the numerical information is not specic to bimanual
responses, being also present with responses involving eye
movements (Fischer et al., 2003): When subjects are required to
detect a stimulus in the right or left sides of the computer screen,
there is an advantage in detecting left-sided stimuli when a smallmagnitude digit (i.e., 1, 2) is previously presented in the centre of
the screen, while right-sided stimuli are primed by a largermagnitude digit (i.e., 8, 9) (Fischer et al., 2003). These shifts of
attention towards the left or the right sides of space are also found
along the vertical dimension, with saccadic responses towards
bottom locations being faster for smaller numbers, saccadic
responses towards top locations faster for larger numbers
(Schwarz and Keus, 2004).
Similar lateralized effects, which emerge in the processing of
numerical information, have been observed in a variety of different
experimental paradigms, such as naming (Brysbaert, 1995; Zebian,
2005), unimanual pointing (Fischer, 2003; Ishihara et al., 2006),
and numerical classication tasks with pedal responses (Schwarz
and Muller, 2006). In these studies, participants show effects of
congruency between the numerical and the spatial information. In
naming tasks, subjects show faster responses when the numerical

display reects the preferred orientation [left-to-right 1 9

(Brysbaert, 1995), or right-to-left 9 1 (Zebian, 2005)]. Similarly,
in the performance of pointing tasks participants are faster in
pointing towards a left location when a smaller-magnitude
number has been presented, and towards a right location with
the preceding presentation of a larger-magnitude number (Fischer,
2003; Ishihara et al., 2006). Finally, the SNARC effect is present also
when participants provide bipedal responses in a numerical
classication task (Schwarz and Muller, 2006).
Recently, it has been shown that the spontaneous mapping of
information into representational space is not conned to
numerical information. In particular, a stimulusresponse compatibility effect has been described in the classication of ordered
non-numerical stimuli, whether or not the order information is the
attended dimension: When subjects classify a letter of the alphabet
(or a month of the year), as coming before or after a specic letter
(month), SNARC-like effects are observed: Letters coming before
the letter (month) of reference are classied faster with the left
hand than with the right hand, the reverse being true for letters
(months) coming after the letter (month) of reference. Critically,
the spatial-congruency effects are present even when these stimuli
are classied according to a property unrelated to the ordinal
meaning, such as in a phoneme monitoring task (Gevers et al.,
2003; Gevers et al., 2004; but see Dehaene et al., 1993, for a null
effect). This evidence undermines the hypothesis that the mapping
of the mental number line representation into a spatial dimension
derives from the meaning of quantity, rather than from the ordinal
meaning.
It has been also suggested, however, that numerical magnitudes
are mapped onto a continuous space, whereas for other ordered
sequences the spatial mapping corresponds to categorical (left/
right) positions (Ishihara et al., 2006; Zorzi et al., 2006). Even if this
were the case, it still remains unclear whether a continuous spatial
mapping is specic to the numerical domain, since it has been
observed that other continua, such as the pitch height, show spatial
compatibility effects along both the vertical and the horizontal
dimensions (Rusconi et al., 2006). In particular, when participants
are asked to compare the pitch of variable frequency tones with a
xed reference, by providing bimanual responses, they show faster
responses in responding to higher tones with the top, and the right
keys, and to lower tones with the bottom, and the left, keys;

Fig. 3. (A) Illustration of the bisection paradigm with numerical material. Participants are asked to set the subjective midpoint of horizontal lines, anked by digits. (B) Mean
spatial biases exhibited in the bisection task: rightward biases are observed when the larger digit is on the right-hand side, leftward biases when the larger digit is on the lefthand side of the line (reprinted from de Hevia et al., 2006b, with permission).

M.D. de Hevia et al. / Neuroscience and Biobehavioral Reviews 32 (2008) 13611372

moreover, the same effects are observed when participants classify

stimuli on the basis of the musical instrument producing the
sounds, suggesting an automatic spatial mapping of tones (Rusconi
et al., 2006).
A number of studies focus on the behavioral effects observed
when numerical and visuo-spatial processing are concurrently
engaged in a given task. Several numerical effects have been
described using a variety of visuo-spatial paradigms, in which the
numerical information is irrelevant, but still affects the performance in a visuo-spatial task. For instance, the task of manual
bisection of a segment has been used by means of anking each
side of horizontal lines by numbers, resulting in the observation of
systematic spatial biases towards the larger magnitude number
(de Hevia et al., 2006a; Fischer, 2001) (Fig. 3). The effect of numbers
on the perception of a line has been interpreted as if subjects
represent numerical ankers demarcating a specic portion of the
mental number line (Fischer, 2001), or as if subjects exhibit
spatial biases towards the larger magnitude, as a result of an
overestimation of the lateral extent closer to the larger digit (de
Hevia et al., 2006a).
Similarly, left- or rightward spatial biases have been reported
when the stimulus to bisect is a string of identical smaller or larger
numbers, respectively, in Arabic as well as in verbal notation,
interpreted on the basis of the SNARC effect (Calabria and Rossetti,
2005; Fischer, 2001). This automatic activation of left/right spatial
codes can actually guide our attention towards the left- or righthand sides of space, as it has been shown in detection tasks using
irrelevant digits as xation points (Fischer et al., 2003). Moreover,
motor performance is modulated by numerical magnitude
information: Motor planning and execution towards the left- or
the right-hand sides of space are inuenced by number magnitude
(Fischer, 2003; Ishihara et al., 2006).
Besides the automatic activation of left-oriented vs. rightoriented spatial codes, numerical magnitude can inuence spatial
processing by inducing a sort of cognitive illusions (de Hevia et al.,
2006a). In particular, underestimations and overestimations of a
spatial extension have been described by using a length
reproduction task, in which participants are required to reproduce
the space between two, previously presented, small or large Arabic
numbers, respectively (de Hevia et al., 2007). In this study,
participants underestimate the spatial gap between two smaller
magnitude numbers (e.g., 1 1). Conversely, the spatial extent is
overestimated when two large magnitude numbers (e.g., 9 9) are
used as ankers (de Hevia et al., 2007). This observation is
consistent with the idea that a given numerical magnitude is
internally represented by a spatial extension (Restle, 1970).
Recently, aperture and closure grasping movements have been
found to be inuenced by the magnitude information of numerical
stimuli (Andres et al., 2004). In this study, participants are required
to perform a closure or opening grip, depending on the parity of a
visually presented digit. Electromyographic recordings of hand
muscles indicate that smaller digits speed up grip closure, larger
digits grip opening (Andres et al., 2004). These interference effects
may be interpreted in the context of an integrative view:
Estimations of magnitude, time, and space are computed by a
general magnitude system, which represents the essential
information guiding the sensorimotor transformations required
for action (Walsh, 2003).
1.3. Neuropsychological evidence
The idea that number and space are intimately related to one
another found early support from neuropsychological studies,
showing that the parietal lobe, a long known basic neural
underpinning of spatial processing (Critchley, 1953; Jewesbury,

1365

1969), is also involved in numerical processing. Parietal damage is

frequently associated to acalculia, a decit of the numerical domain
of cognition, which may be independent of other cognitive
disorders (Hecaen et al., 1961; Henschen, 1919).
A striking instance of the putative relationship between
processing of space and number may be provided by the
Gerstmanns syndrome, a four-symptom disorder, typically associated with damage to the inferior parietal lobule (angular gyrus)
of the left hemisphere. The syndrome comprises acalculia, agraphia,
nger agnosia, and left-right disorientation (Gerstmann, 1940). The
syndrome has a denite value for anatomical localization,
suggesting a lesion of the inferior-posterior parietal lobe of the
left hemisphere (Morris et al., 1984; Roeltgen et al., 1983; Strub
and Geschwind, 1983). The functional unity of the syndrome,
however, and the hypothesis of a single pathological mechanism
underlying the four components of it have been long criticized
(reviews and discussion in Benton, 1992; Critchley, 1966; Morris
et al., 1984). The main argument, mainly developed in the 1960s by
the north-American neuropsychologist Arthur Benton on the basis
of the investigation of large series of brain-damaged patients (e.g.,
Benton, 1961), is that the four symptoms of the syndrome may
occur in isolation. The association, accordingly, has a relevant
anatomical value, suggesting a focal left posterior parietal lesion,
but is not functionally relevant: the syndrome is anatomical, not
functional (Vallar, 2000).
Critically for the present review, the hypothesis that the tetrad
of symptoms characterizing the Gerstmann syndrome share a
common underlying decit of visuo-spatial nature (reviews in
Critchley, 1966; Vallar, 2007) has been revived in recent years. On
the basis of the investigation of individual left-brain damaged
patients with Gerstmann syndrome, suggestions have been made
that the core impairment may concern difculties in deriving the
relative position of an object along the horizontal axis (Gold et al.,
1995), or the mental transformation of visual images (e.g., rotation,
translation) (Mayer et al., 1999; see also Carota et al., 2004).
Moreover, the performance of 5- and 7-year-old children in tests
assessing nger representation, leftright orientation, constructional abilities, and handwriting, is an excellent predictor of
arithmetical skills, but not of verbal performance (Noel, 2005).
More recent evidence for a fundamental association between
number and space comes from studies assessing number processing in brain-damaged patients suffering from the syndrome of
unilateral spatial neglect, that is more frequent and severe after a
right hemispheric lesion, involving the left, contralateral, part of
space (Vallar, 1998). The decit consists of the inability to explore
the side of space contralateral to the side of the lesion, and to report
stimuli presented in that portion of space (Bisiach and Vallar,
2000). Critical to the present review is the fact that unilateral
spatial neglect extends to the scanning of mental images generated
by the patient (Bisiach and Luzzatti, 1978). With regard to
numerical processing, when right-brain-damaged patients with
neglect are asked to compare numbers with a reference one, shown
centrally at xation, their latencies are increased for those located
to the immediate left of the reference, along the mental number
line, such as 4 vs. a reference number 5. This imagery decit
extends to the left-sided positions of the hours onto an imagined
clockface, with slower latencies to hours such as 9, and is not
found in right-brain-damaged patients without neglect (Vuilleumier et al., 2004). These ndings support the hypothesis that the
number comparison task invokes an internal space of magnitude,
oriented from left-to-right. In patients with left spatial neglect, the
left side of such a representation may be not available for
numerical processing. In sum, a unilateral impairment of spatial
representation and attention (in the absence of primary numerical
decits or acalculia) may affect performance in some numerical

1366

M.D. de Hevia et al. / Neuroscience and Biobehavioral Reviews 32 (2008) 13611372

tasks. This suggests that spatial representations may contribute to

aspects of numerical processing.
The task recently exploited for the assessment of the mental
number line representation is the numerical bisection task, which
consists in detecting the midpoint of a numerical interval (e.g.,
which is the central number between 2 and 8?). This task is
supposed to rely explicitly on the representation of the mental
number line (Dehaene, 1992; but see Nuerk et al., 2002).
Accordingly, performance should primarily reect visuo-spatial
internal representations. Some studies (Zorzi et al., 2006; Zorzi
et al., 2002) have argued that, when assessed by this task, the
performance of brain-damaged patients with left neglect parallels
the rightward error found in the bisection of physical lines
(Heilman and Valenstein, 1979; Schenkenberg et al., 1980), by
showing a displacement of the numerical centre towards the right
side, namely the larger number. Double dissociations between the
bisection of physical lines vs. numerical imagined lines are also on
record, however, with patients being selectively impaired only in
one of these tasks (Doricchi et al., 2005; Rossetti et al., 2004). These
ndings extend to the spatial representations of numbers the
distinction between left neglect for objects in extra-personal space
vs. in internally generated visuo-spatial mental images (Anderson,
1993; Bisiach and Luzzatti, 1978; Guariglia et al., 1993).
Notwithstanding a defective processing of the left side of the
representation of the mental number line, right-brain-damaged
patients with left neglect may show a SNARC effect, as assessed by
a task in which numbers are classied according to their parity
with left-right unimanual responses (Priftis et al., 2006). This
nding suggests that a left-to-right visuo-spatial numerical
representation is still preserved in brain-damaged patients with
left neglect. These patients, however, show impairment when the
task requires a more explicit processing (Priftis et al., 2006). More
generally, these ndings are in line with the view that the core
decit of the neglect syndrome concerns perceptual awareness of
left-sided, contralesional, representations (e.g., Berti, 2002).
Furthermore, manipulations that transiently ameliorate a number
of manifestations of the syndrome of spatial neglect (i.e., the prism
adaptation procedure, see review in Rossetti and Rode, 2002) also
improve the performance of patients with left neglect in the
number bisection task (Rossetti et al., 2004). This nding suggests
that sensorimotor transformations operating in the visuo-spatial
domain are able to exert an inuence on higher cognitive abilities,
such as the internal representations of numerical magnitude, and
further support their spatial nature.
1.4. Neuro-anatomical basis: lesion and neuroimaging data
The view of a functional interaction between numerical and
spatial processes, suggested by the previously reviewed evidence,
is further supported by the partial overlapping of the posterior

parietal areas devoted to the processing of either types of

information. Neuropsychological studies suggest that decits in
numerical processing may involve verbal vs. quantity numerical
representations. Damage to the parietal lobe is usually associated
to a decit in quantitative tasks that tackle the internal
manipulation of quantity, such as magnitude comparison, estimations of numerosity, and bisection of numerical intervals, while
other language-related abilities, such as counting, and retrieval of
arithmetic facts, are well preserved (Dehaene and Cohen, 1997;
Delazer and Benke, 1997; Delazer et al., 2006a,b; Lemer et al.,
2003).
As evidence for such a dissociation, an aphasic patient affected
by a left fronto-temporal atrophy was more impaired in multiplication fact retrieval (highly dependent on rote verbal memory),
than in subtraction (primarily solved via non-verbal quantity
meaning). The patient compared and calculated with large nonsymbolic quantities (arrays of dots) better than with tasks
requiring verbal counting (Lemer et al., 2003). On the other hand,
a patient who presented with a Gerstmanns syndrome following a
left parietal ischemic lesion, was more deteriorated in the
manipulation of non-symbolic quantities and in subtraction, than
in multiplication and counting procedures (Lemer et al., 2003).
These dissociations between quantity and verbal numerical
processes support the view of a basic distinction, within numerical
processing, between a verbal system involving the left frontal and
temporal (language) regions, and a non-verbal quantity system,
based on the left and right intraparietal regions (Dehaene et al.,
2003). The participation of specic brain regions to numerical
processing may be modulated by practice. In fact, one study
(Delazer et al., 2003) has shown, during a training session of
multiplication facts, a shift of the focus of activation from the
intraparietal sulcus to the angular gyrus (see however Rickard
et al., 2000, for contrasting evidence on the activation of the
angular gyrus in an arithmetic task).
In the last decade, a large number of brain imaging studies have
attempted to localize the brain regions involved in the processing
of numerical information. Early functional-anatomical models
suggested a role of the bilateral posterior parietal regions for the
representation of numerical information (Dehaene and Cohen,
1995). More recent studies have rened, and further specied, the
regions involved in the processing of number magnitude. PET
studies show posterior parietal activations in simple numerical
tasks, such as single digits multiplication, comparison, naming, and
subtraction (Dehaene et al., 1996; Pesenti et al., 2000; Zago et al.,
2001). fMRI studies reveal that manipulation of numerical
information leads to an activation of the posterior-inferior parietal
lobule (Chochon et al., 1999; Pinel et al., 2001). The specialized role
of the posterior parietal cortex for numerical processing is
supported by studies showing that the intraparietal sulcus (IPS)
is activated when participants are engaged in subtraction

Fig. 4. Bilateral parietal network involved in numerical processing. A bilateral-intraparietal system may be devoted to the more abstract representation of quantity (grey), the
left angular gyrus to verbal numerical representations (black), and a bilateral posterior-superior parietal system to spatial and non-spatial attention (cross-hatched).
(Adapted from Dehaene et al., 2003).

M.D. de Hevia et al. / Neuroscience and Biobehavioral Reviews 32 (2008) 13611372

problems with Arabic numbers, among a variety of tasks including

pointing, visual saccades, phoneme detection, and attention
(Simon et al., 2002).
In fact, a domain-specic localization hypothesis for numerical
processing has been proposed, according to which the horizontal
segment of the intraparietal sulcus (IPS), frequently related to
visuo-spatial processing, is involved in the coding of numerical
magnitude (Dehaene et al., 2003) (Fig. 4). In line with this view, a
bilateral region in the horizontal IPS is activated in a greater extent
when participants are involved in a detection task with Arabic
numbers, in comparison to the presentation of letters or colors
(Eger et al., 2003; but see contrasting evidence in Fias et al., 2007,
and in Shuman and Kanwisher, 2004). Moreover, the horizontal IPS
appears to be sensitive to the numerical distance effect, with the
numerical comparison of closer digits leading to a higher degree of
parietal activation than that of digits farther apart (Pinel et al.,
2001). Similar activations have been obtained with passive
viewing tasks (Piazza et al., 2004, 2007). The activation of the
IPS is stronger in approximate than in exact calculation problems,
supporting its role in the representation of numerical quantity
(Dehaene et al., 1999). In line with these ndings, more IPS
activation is observed during subtraction than during multiplication (Chochon et al., 1999).
The left IPS is activated by comparison judgments between
symbolic (Arabic numbers), and non-symbolic (lines and angles)
magnitudes (Fias et al., 2003). A recent study has shown a bilateral
activation of the IPS for the estimation of discrete non-symbolic
numerosities, both in sequential (temporal) and in simultaneous
(visuo-spatial) presentations, compared to the estimation of
analogue quantity (i.e., extension) (Castelli et al., 2006). These
ndings suggest the existence of a number-specic region in the
parietal lobe. Additional converging evidence for the role of the
posterior parietal regions, bilaterally, in basic number processing
comes from recent studies using the rTMS technique. In particular,
the participants performance in the standard numerical comparison task is disrupted after bilateral stimulation of the angular
gyrus (Gobel et al., 2001), and of the left inferior parietal lobule
(Sandrini et al., 2004). Finally, stimulation of the right parietal
cortex affects performance in a numerical bisection task, shifting
rightwards the numerical midpoint (Gobel et al., 2006; Oliveri
et al., 2004).
In monkeys trained to perform a numerical matching task, cells
both in the lateral prefrontal cortex and in the IPS selectively
respond to numerosity (Nieder and Miller, 2004). Importantly, IPS
neurons require less time than those in the prefrontal cortex to
become selective to numerosity, and their response adjustment is
increasingly imprecise as numerosity increases. This response
change follows the Weber-fraction, and thus mirrors the internal
representation of magnitude (Nieder and Miller, 2004; Nieder and
Miller, 2003). Finally, the level of abstraction of these numerosityselective neurons is demonstrated by the nding that, although in
the monkeys cortex the codication of number is segregated for
spatial (simultaneous) and temporal (sequential) visual presentations, a common neural population ultimately encodes a more
abstract quantity representation, irrespective of their numerical
format (Nieder et al., 2006).
To further investigate the possibility of numerosity tuning in
humans, the fMRI adaptation paradigm has been employed (Piazza
et al., 2004; Shuman and Kanwisher, 2004). In these studies, the
fMRI response, in brain areas supposed to process the quantity
dimension, is expected to be lower for repeated stimuli, due to
neural adaptation, than for stimuli changing in quantity. Consistent with the hypothesis that the horizontal segment of the IPS
processes number, activation is attenuated with repeated presentation of the same numerical quantity conveyed by an Arabic

1367

symbol, or by a number word (Naccache and Dehaene, 2001).

Similar adaptation effects occur during the presentation of dots
arrays with the same quantity, leading to an increased activity in
the IPS bilaterally, when new numerosities are presented (Piazza
et al., 2004). Moreover, a cross-notational adaptation effect
between symbolic (digits) and non-symbolic (arrays of dots)
number has been found (Piazza et al., 2007). Recent evidence from
4-year-old children (Cantlon et al., 2006) further supports the
hypothesis that the IPS is implicated in a notation-independent
representation of numerical magnitude (Dehaene et al., 1998).
However, while the evidence is accumulating in support for the
activation of the IPS in number processing, the hypothesis that its
horizontal segment is specic for numbers has been recently
challenged. In particular, when presenting non-symbolic numerical information, a signicant involvement of the IPS is found,
although not higher than the activation observed in the same area
for color judgments. These results suggest that the IPS participates
in processes concerned not only with number processing, but also
with the analysis of other types of information. These results
challenge the hypothesis of a single, domain-specic, region in the
IPS that underlies both symbolic and non-symbolic number
processing (Shuman and Kanwisher, 2004). In fact, parallel to
the behavioral evidence of SNARC-like effects for non-numerical
ordinal information (Gevers et al., 2003, 2004), a recent fMRI study
using a comparison task has shown that the anterior region of the
horizontal IPS is equally involved in the processing of numbers and
letters (Fias et al., 2007). This evidence suggests that the IPS region
may represent numerical ordinality, and not only cardinality (Fias
et al., 2007; Nieder, 2005).
2. Calculation and visuo-spatial processes
2.1. The hypothesis of a visuo-spatial medium for arithmetic
The importance of visuo-spatial metaphors in grasping
mathematical concepts has been stressed by many authors (e.g.,
ez, 2000). In fact, a number
Bryant and Squire, 2001; Lakoff and Nun
of studies have emphasized the fact that participants spontaneously rely on visuo-spatial processes when they engage in
arithmetic and numerical processing (e.g., Dehaene, 1992; Restle,
1970; Seron et al., 1992). For instance, the suggestion has been
made that visuo-spatial imagery consists of one of several
strategies among which subjects may choose to solve arithmetic
problems (Siegler and Lemaire, 1997).
Visuo-spatial representations may be one resource, among
other numerical codes, that subjects may preferentially use to
maintain numerical information in an active state (Noel and Seron,
1993; Seron et al., 1992). The distinction between auditory and
visual calculators, neatly introduced by Smith (1983), in describing
strategies and performances of famous mathematics prodigies, has
been conrmed by a recent case-study of a prodigy turning to
visual strategies when calculating (Pesenti et al., 2001). At the
same time, other authors have attached a more fundamental role
to visuo-spatial imagery in calculation (Hayes, 1973; Heathcote,
1994; Restle, 1970). Furthermore, visualizing mathematical
concepts helps their acquisition: For instance, in formal school
teaching, the number line diagram has been successfully employed
to improve and train subjects in less than and more than kind of
relations (Grifn et al., 1994).
In the study of calculation processing, models of numerical
cognition assume the existence of a long-term memory store of
arithmetical facts, containing known sums, products, and the like
(e.g., 2  4), which are retrieved by participants without following
the corresponding calculation algorithm (e.g., Ashcraft, 1992). This
semantic network is thought to rely on language-based mechan-

1368

M.D. de Hevia et al. / Neuroscience and Biobehavioral Reviews 32 (2008) 13611372

isms, since simple arithmetic facts are supposed to be acquired in

school by verbal rote learning (Dehaene, 1992). However, multidigit operations (e.g., 34  25) are supposed to require, besides
retrieval of arithmetical facts, the mental manipulation of a spatial
image of the operation in Arabic format (Dehaene, 1992). Its
maintenance in a short-term representational medium, while the
dedicated solution algorithm is applied, is also required. Furthermore, whether or not the calculation problem needs to be solved
by writing it down, mastering the spatial information contained in
numbers is crucial, for instance, to know the value of digits which,
syntactically organized, form a complex number (e.g., units,
decades, hundreds), and to know which number must operate
with which other number in a calculation problem (e.g., in
multiplication, addition, and subtraction). Thus, cognitive models
of numerical processing consider the contribution of visuo-spatial
and visuo-constructive abilities in arithmetical operations
(Dehaene, 1992; McCloskey et al., 1985), and propose the
involvement of both verbal and visuo-spatial working memory
during complex calculation (Dehaene and Cohen, 1995).
The involvement of mental imagery in arithmetic has been
considered crucial in order to hold the problem operands, and the
interim results, active in a visuo-spatial working memory system
during calculation (Hayes, 1973; Hitch, 1978). For instance, it has
been suggested that visual imagery is used as a strategy to solve
arithmetic problems, since it allows the maintenance of notations
that would be performed on paper in written arithmetic (Hayes,
1973). The view that calculation is partially mediated by mental
visual images raised interest in the contribution of working
memory in calculation and counting processes. In particular, a
signicant quantity of errors in complex addition may be the
consequence of a failure in maintaining crucial information in
working memory, such as the carries, and the interim results
(Hitch, 1978). However, the latter observations do not focus on the
specic system(s) of working memory (WM) crucial for arithmetic
processing (review in Baddeley, 2007). The question as to whether
the visuo-spatial component of working memory, the visuo-spatial
sketch pad (VSSP), plays a role in arithmetic has been investigated
in behavioral studies, by means of interfering with its WM capacity
during calculation.
2.2. Experimental paradigms and effects
The relationship between the visuo-spatial component of
memory and calculation skills has been the object of a few
experimental studies, and has mainly been investigated by means
of the dual task-paradigm. This methodology has been used to
identify which WM components are engaged in performing
cognitive tasks (e.g., Logie et al., 1989). The logic of this method
consists of engaging participants in a primary cognitive task, while
a secondary task, known to place load on the WM components, is
concurrently performed. The observed pattern of interference (or
lack thereof) provides information about the differential engagement of specic WM subsystems in a given cognitive task.
Adopting this rationale, Logie et al. (1994) required subjects to
perform a cumulative addition task, consisting of adding a series of
two-digit numbers presented in the visual or auditory modality,
while the verbal and visuo-spatial WM components were loaded by
a secondary task. Spatial suppression, obtained by asking participants to perform a spatial tapping task (i.e., hand movement),
produces an interference effect. However, the interference effects
are minor in size, and limited to the condition in which numbers are
presented in the visual modality. These results suggest a minor
involvement of the VSSP in mental arithmetic (Logie et al., 1994).
Further support for the importance of visual imagery in
calculation comes from the observation that multi-digit addition

problems containing visually similar digits are more prone to

errors than addition problems with dissimilar digits. This nding
suggests that visual WM is implied in the encoding stage of
numerical information (Heathcote, 1994). In this study, visuospatial suppression disrupts calculation performance, the more so
in complex problems involving carrying. Visuo-spatial representation may be critical for maintaining the information of the carries
(Heathcote, 1994).
The role of the VSSP in complex arithmetic has been further
investigated by manipulating the visual similarity of the addends
in multi-digit addition (Noel et al., 2001). This study did not nd
any disrupting effect on the overall performance of participants, as
a group; however, emphasizing the role of individual differences in
arithmetic performance (LeFevre and Kulak, 1994), and in
manipulating number representation (Noel and Seron, 1993),
evidence has been provided, in the performance prole of each
participant, for a sensitivity, in the individual subject, to the
manipulation of the visual features of the stimuli. Accordingly,
some individuals may preferentially rely on the visualization of the
numerical information in computing calculations (Noel et al.,
2001; see also Hatta et al., 1989), particularly those for whom the
visual form constitutes a preferred code of number representation
(Noel and Seron, 1993).
Recent empirical evidence suggests that the recruitment of
visuo-spatial codes in arithmetic depends upon contextual factors,
such as the problems format. Previous ndings indicate that the
presentation format affects level of performance for arithmetic facts
(Campbell, 1994; Noel and Seron, 1997), and this seems to apply also
to complex written calculation. Trbovich and LeFevre (2003)
manipulated the vertical and horizontal alignment of addition
problems to be solved by participants, while retaining a memory
load. Indeed, the vertical presentation of the operands, which
mirrors the arrangement of operands in written arithmetic, was
expected to elicit the same unit-to-decade addition algorithm used
to solve written calculation, thus requiring more visual resources
than horizontally presented operands. As expected, the results
indicate a cost of the visual load limited to the problems presented in
vertical format, suggesting a selective recruitment of the VSSP
component of WM in this condition (Trbovich and LeFevre, 2003).
Besides written calculation, the role of visuo-spatial processing
in simple arithmetic has been investigated using a dual-task
paradigm, for either phonological or visuo-spatial suppression (Lee
and Kang, 2002). In particular, specic predictions derived from the
hypothesis of different forms of number representation involved in
different operations, as postulated by the Triple Code model
(Dehaene and Cohen, 1997; Dehaene et al., 1996). According to this
model, an auditory-verbal code is preferentially used for multiplications learned by verbal rote during school-years, whereas
subtraction depends on the manipulation of quantity meaning,
thus implying the activation of semantic representations, conceived as a mental number line (Dehaene, 1992). This hypothesis
predicts an operation-specic relationship between arithmetic
performance and working memory subsystems, as indeed reported
by Lee and Kang (2002): Phonological suppression affects multiplication but not subtraction, whereas visuo-spatial suppression
interferes with subtraction but not multiplication (Lee and Kang,
2002). These ndings lend support to the hypothesis of discrete,
operation-specic, numerical representations in simple arithmetic
(Dehaene et al., 1999).
It is worth noticing that brain-damaged patients with spatial
neglect may present with no signs of acalculia and, specically, be
unimpaired in arithmetic tasks, such as simple arithmetic fact
retrieval with one- and two-digit numbers (Vuilleumier et al.,
2004), and numerical comparison and subtraction tasks (Zorzi
et al., 2002). Furthermore, patients may be impaired exploring

M.D. de Hevia et al. / Neuroscience and Biobehavioral Reviews 32 (2008) 13611372

explicitly a spatially organized numerical representation, showing

left-sided neglect in a mental bisection task, but no effects of
neglect on the SNARC effect (Priftis et al., 2006). To account for
these ndings, one might imply that numerical visuo-spatial
representations are invoked only implicitly when supporting
subtraction and/or addition problems, or the SNARC effect.
2.3. Neuropsychological evidence
Intuitively, one may consider visuo-spatial processing as
intrinsically related to number processing and calculation, all
the more so when treating multi-digit numbers. Difculties of
visuo-spatial nature would thus manifest themselves in the
assembly and organization of numbers, or in the spatial layout
of written calculation. In the neuropsychological literature, the role
of visual and spatial processes in the occurrence of numerical
decits was in fact recognized very early on, although mainly
grounded on anecdotal and scattered clinical observations (e.g.,
Kleist, 1934; Peritz, 1918; Singer and Low, 1933).
The link between numerical abilities and space was formalized
by Hecaen who introduced the term spatial acalculia, with
reference to difculties in dealing with written digits in a specic
order and position, including calculation decits associated with
spatial neglect (Hecaen et al., 1961). Typical errors associated with
spatial acalculia include incorrect alignment of digits in columns,
confusions with reversals (32 for 23), or with visually similar digits
(6 for 9), and difculties in maintaining the decimal place (for a
review, see Hartje, 1987) (Fig. 5). Since the seminal paper by Hecaen
et al. (1961), the spatial category of acalculia refers to calculation
decits considered to be secondary to visuo-spatial disorders. On the
basis of this assumption, acalculia has been investigated in group
studies of right-brain-damaged patients (Ardila and Roselli, 1994;
Basso et al., 2000; Dahmen et al., 1982; Rosselli and Ardila, 1989).
Most of these studies have conrmed the presence of spatially
related difculties in multi-digit written calculation in righthemisphere-damaged patients, namely: Problems with carrying
and borrowing, organization of interim products, and failure in the
correct alignment of digits. These reports, however, do not provide
any interpretation concerning the functional origin of these errors.
Only recently, efforts have been made to identify which
components of multi-digit calculation rely on spatial cognition
and, in turn, to verify whether spatial errors may be related to
calculation-specic spatial disturbances, rather than to a generic
spatial decit (Grana` et al., 2006). In particular, Grana` et al. (2006)
report a right-hemisphere-damaged patient who systematically
fails on multiplication procedures. Specically, while knowing
what, when, and how to carry out the various steps required to

1369

solve a multi-digit multiplication, the patient does not know

where. The patients difculties are traced back to the impairment of a visuo-spatial store, containing a layout representation
specic to multiplication. This system may support neurologically
unimpaired calculators in overcoming the WM demands of
complex calculation, by representing the information of where
each sub-step should be placed (Grana` et al., 2006). The role for
such a memory-based visuo-spatial store resource, or representation, in calculation had been previously suggested (Dehaene and
Cohen, 1995), but was awaiting for experimental support.
Other neuropsychological syndromes, like posterior cortical
atrophy, which includes other complex visual disorders and leftto-right disorientation (Benson et al., 1988), provide further
support for the close relationship between visuo-spatial decits
and impairment of calculation abilities. A recent case study
presents a detailed neuropsychological examination of numerical
and arithmetical abilities in a patient with a severe atrophy of the
bilateral posterior parietal regions, more pronounced on the right
side (Delazer et al., 2006a,b). The patients severe impairment in
visuo-spatial processing was accompanied by a series of numerical
decits in tasks assessing counting large series of dots arrays,
numerical bisection, approximation, and estimation. Moreover,
while the recitation of the counting sequence was well preserved
(as predicted by the preserved language abilities), the patient was
severely impaired in the subtraction and division arithmetical
abilities (both mental and written) (Delazer et al., 2006a,b).
Aside from the neuropsychological evidence in adult patients
with acquired brain damage, hints for the differential contribution of
verbal and non-verbal competence to numerical cognition comes
from the investigation of number abilities in children with genetic
disorders, and specically those whose cognitive impairment is
characterized by visuo-spatial decits (i.e., Williams syndrome and
Turner Syndrome). In particular, since Williams syndrome (WS) is
marked by a relative strength in language, coupled with a severe
decit in visuo-spatial abilities, attention has recently been directed
to the decit in numerical cognition, typically associated with these
disorders (Ansari et al., 2003; Paterson et al., 2006). Critically, the
impairment in visuo-spatial cognition prevents the normal development of exact number representation, regardless of adequate
language abilities (Ansari et al., 2003). Indeed, in the context of an
overall delay in cardinality understanding, language predicts the
success of WS children in numerical tasks better than visuo-spatial
abilities, which is opposite to what occurs for normally developing
unimpaired children, whose visuo-spatial abilities predict their
performance in numerical tasks. Overall, these studies point to the
role that visuo-spatial cognition plays in the normal development of
numerical understanding, strengthening the view that basic number
competence is driven by a non-verbal system of representations
(Dehaene et al., 1999; Gelman and Butterworth, 2005).
2.4. Neural correlates

Fig. 5. Examples of spatial errors in written calculation: misalignment of partial

product in complex multiplication (adapted from Grana` et al., 2006).

The observation that number processing and calculation rely on

brain areas devoted to visuo-spatial processes has been highlighted in numerous neuroimaging studies (e.g., Chochon et al.,
1999; Dehaene et al., 1999; Dehaene et al., 1996; Gruber et al.,
2001; Pesenti et al., 2001; Simon et al., 2002; Zago et al., 2001; Zago
and Tzourio-Mazoyer, 2002). Complex calculation includes a wide
range of cognitive components, from number-specic processes,
such as accessing the quantity meaning of numbers, to processes
shared by arithmetical and non-arithmetical tasks, such as
attention, and the active maintenance of the numerical information in spatial WM (Simon et al., 2002). The brain areas that are
engaged in complex calculation comprise a network of prefrontal,
premotor, and parietal cortices (e.g., Gruber et al., 2001).

1370

M.D. de Hevia et al. / Neuroscience and Biobehavioral Reviews 32 (2008) 13611372

Few neuroimaging studies, aiming at characterizing the WM

systems engaged during the resolution of a complex calculation,
have shown that visuo-spatial working memory is massively
exploited in arithmetic performance (Pesenti et al., 2000; Zago
et al., 2001; Zago and Tzourio-Mazoyer, 2002). For instance, when
comparing the brain activations during a complex calculation task
and a non-numerical visuo-spatial task, a bilateral involvement of
the posterior parietal regions is found (Zago and Tzourio-Mazoyer,
2002). In particular, the task of multiplying a pair of two-digit
numbers relies, among other areas, on a bilateral fronto-parietal
network (supposed to be engaged on holding multi-digit numbers
in visuo-spatial WM), as well as on the inferior temporal gyrus
bilaterally (possibly involved in mental imagery) (Zago et al.,
2001). Moreover, it has been shown that, during the development
of arithmetical abilities, signicant changes in neural responses are
observed (Rivera et al., 2005). In this study, 8-to-20-year-old
participants performed a verication task with a series of two-digit
addition and subtraction problems: Older participants exhibit
more activation in the left supramarginal gyrus and the IPS,
whereas in younger subjects there is more activity in the prefrontal
cortex. These ndings suggest a developmental process of
functional specialization of the posterior parietal cortex for mental
arithmetic (Rivera et al., 2005).
With regard to the role of visuo-spatial WM, the upper part of
the right supramarginal gyrus may be a relevant region (Becker
et al., 1999). The role of the inferior parietal lobule (angular and
supramarginal gyri) in calculation has been repeatedly suggested
(Dehaene et al., 1999; Gruber et al., 2001; Menon et al., 2000; Zago
and Tzourio-Mazoyer, 2002). Particularly, a bilateral increase of
activation in the angular gyrus of the inferior parietal cortex, in
response to increasing arithmetic complexity, has been reported
(Menon et al., 2000). Activation of the right supramarginal gyrus is
associated with both complex calculation on visually presented
numbers, and non-numerical tasks that tap visuo-spatial WM
(Zago and Tzourio-Mazoyer, 2002). These observations concur to
suggest that the storage of numbers, such as the carries, needed for
mental calculation, is accomplished by holding information in a
visuo-spatial short-term representational medium (Pesenti et al.,
2000; Zago and Tzourio-Mazoyer, 2002).
3. Conclusions
Following very early suggestions, starting from the end of the
XIX century (Galton, 1880; Moyer and Landauer, 1967; Restle,
1970), there is converging evidence to the effect that numerical
and arithmetical abilities involve visuo-spatial resources, for the
purpose of processing and temporary storage. Overall, neuropsychological reports, neuroimaging and behavioral ndings support
the view that visuo-spatial processes and numerical representation are intimately related. This relationship may constitute an
early and fundamental link which, although partially shaped by
our cultural constraints, remains an essential component of our
cognitive architecture.
Acknowledgements
Supported in part by a MUR Prin 2005 Grant to G.V. and L.G.
M.D.dH. was supported by a Spanish MEC-Fulbright Postdoctoral
Fellowship.
References
Anderson, B., 1993. Spared awareness for the left side of internal visual images in
patients with left-sided extrapersonal neglect. Neurology 43, 213216.
Andres, M., Davare, M., Pesenti, M., Olivier, E., Seron, X., 2004. Number magnitude
and grip aperture interaction. Neuroreport 15, 27732777.

Ansari, D., Donlan, C., Thomas, M.S., Ewing, S.A., Peen, T., Karmiloff-Smith, A., 2003.
What makes counting count? Verbal and visuo-spatial contributions to typical
and atypical number development. Journal of Experimental Child Psychology
85, 5062.
Ardila, A., Roselli, M., 1994. Spatial acalculia. International Journal of Neuroscience
78, 177184.
Ashcraft, M.H., 1992. Cognitive arithmetic: a review of data and theory. Cognition
44, 75106.
Bachtold, D., Baumuller, M., Brugger, P., 1998. Stimulusresponse compatibility in
representational space. Neuropsychologia 36, 731735.
Baddeley, A.D., 2007. Working Memory, Thought, and Action. Oxford University
Press, Oxford.
Basso, A., Burgio, F., Caporali, A., 2000. Acalculia, aphasia and spatial disorders in left
and right brain-damaged patients. Cortex 36, 265280.
Becker, J.T., MacAndrew, D.K., Fiez, J.A., 1999. A comment on the functional
localization of the phonological storage subsystem of working memory. Brain
and Cognition 41, 2738.
Benson, D.F., Davis, R.J., Snyder, B.D., 1988. Posterior cortical atrophy. Archives of
Neurology 45, 789793.
Benton, A.L., 1961. The ction of the Gerstmann syndrome. Journal of Neurology,
Neurosurgery and Psychiatry 24, 176181.
Benton, A.L., 1992. Gerstmanns syndrome. Archives of Neurology 49, 445447.
Berch, D.B., Foley, E.J., Hill, R.J., Ryan, P.M., 1999. Extracting parity and magnitude
from Arabic numerals: developmental changes in number processing and
mental representation. Journal of Experimental Child Psychology 74, 286308.
Berti, A., 2002. Unconscious processing in neglect. In: Karnath, H.-O., Milner,
A.D., Vallar, G. (Eds.), The Cognitive and Neural Bases of Spatial Neglect.
Oxford University Press, Oxford, pp. 313326.
Bisiach, E., Luzzatti, C., 1978. Unilateral neglect of representational space. Cortex 14,
129133.
Bisiach, E., Vallar, G., 2000. Unilateral neglect in humans. In: Boller, F., Grafman,
J., Rizzolatti, G. (Eds.), Handbook of Neuropsychology. second ed. Elsevier
Science, Amsterdam, pp. 459502.
Bryant, P., Squire, S., 2001. Childrens mathematics: lost and found in space. In:
Gattis, M. (Ed.), Spatial Schemas and Abstract Thought. MIT Press, Cambridge,
MA, pp. 175200.
Brysbaert, M., 1995. Arabic number reading: on the nature of the numerical scale
and the origin of phonological recoding. Journal of Experimental Psychology:
General 124, 434452.
Buckley, P., Gillman, C., 1974. Comparisons of digits and dot patterns. Journal of
Experimental Psychology 103, 11311136.
Calabria, M., Rossetti, Y., 2005. Interference between number processing and line
bisection: a methodology. Neuropsychologia 43, 779783.
Campbell, J.I.D., 1994. Architectures for numerical cognition. Cognition 53, 144.
Cantlon, J., Brannon, E.M., Carter, E.J., Pelphrey, K., 2006. Notation-independent
number processing in the intraparietal sulcus in adults and young children.
PLoS Biology 4, 111.
Carota, A., Di Pietro, M., Ptak, R., Poglia, D., Schnider, A., 2004. Defective spatial
imagery with pure Gerstmanns syndrome. European Neurology 52, 16.
Castelli, F., Glaser, D.E., Butterworth, B., 2006. Discrete and analogue quantity
processing in the parietal lobe: a functional MRI study. Proceedings of the
National Academy of Sciences 103, 46934698.
Castronovo, J., Seron, X., 2007. Semantic numerical representation in blind subjects:
the role of vision in the spatial format of the mental number line. Quarterly
Journal of Experimental Psychology 60, 101119.
Chochon, F., Cohen, L., van de Moortele, P.F., Dehaene, S., 1999. Differential contributions of the left and right inferior parietal lobules to number processing.
Journal of Cognitive Neuroscience 11, 617630.
Chokron, S., Imbert, M., 1993. Inuence of reading habits on line bisection. Cognitive
Brain Research 1, 219222.
Cooper, R.G., 1984. Early number development: discovering number space with
addition and subtraction. In: Sophian, C. (Ed.), Origins of Cognitive Skills.
Erlbaum, Hillsdale, NJ, pp. 157192.
Critchley, M., 1953. The Parietal Lobes. Hafner, New York.
Critchley, M., 1966. The enigma of Gerstmanns syndrome. Brain 89, 183198.
Dahmen, W., Hartje, W., Bussing, A., Sturm, W., 1982. Disorders of calculation in
aphasic patientsspatial and verbal components. Neuropsychologia 20, 145
153.
de Hevia, M.D., Girelli, L., Bricolo, E., Vallar, G., 2008. The representational space of
numerical magnitude: illusions of length. Quarterly Journal of Experimental
Psychology 61, 14961514.
de Hevia, M.D., Girelli, L., Vallar, G., 2006a. Numbers and space: a cognitive illusion?
Experimental Brain Research 168, 254264.
de Hevia, M.D., Vallar, G., Girelli, L., 2006b. Visuo-spatial components of numerical
representation. In: Vecchi, T., Bottini, G. (Eds.), Imagery and Spatial Cognition.
John Benjamins, Amsterdam, pp. 155171.
Dehaene, S., 1992. Varieties of numerical abilities. Cognition 44, 142.
Dehaene, S., 2003. The neural basis of the WeberFechner law: a logarithmic mental
number line. Trends in Cognitive Sciences 7, 145147.
Dehaene, S., Akhavein, R., 1995. Attention, automaticity, and levels of representation in number processing. Journal of Experimental Psychology: Learning,
Memory, and Cognition 21, 314326.
Dehaene, S., Bossini, S., Giraux, P., 1993. The mental representation of parity and
number magnitude. Journal of Experimental Psychology: General 122, 371
396.

M.D. de Hevia et al. / Neuroscience and Biobehavioral Reviews 32 (2008) 13611372

Dehaene, S., Cohen, L., 1995. Towards an anatomical and functional model of
number processing. Mathematical Cognition 1, 83120.
Dehaene, S., Cohen, L., 1997. Cerebral pathways for calculation: double dissociation
between rote verbal and quantitative knowledge of arithmetic. Cortex 33, 219
250.
Dehaene, S., Dehaene-Lambertz, G., Cohen, L., 1998. Abstract representations of
numbers in the animal and human brain. Trends in Neurosciences 21, 355361.
Dehaene, S., Dupoux, E., Mehler, J., 1990. Is numerical comparison digital? Analogical and symbolic effects in two-digit number comparison. Journal of Experimental Psychology: Human Perception and Performance 16, 626641.
Dehaene, S., Piazza, M., Pinel, P., Cohen, L., 2003. Three parietal circuits of number
processing. Cognitive Neuropsychology 20, 487506.
Dehaene, S., Spelke, E., Pinel, P., Stanescu, R., Tsivkin, S., 1999. Sources of mathematical thinking: behavioral and brain-imaging evidence. Science 284, 970
974.
Dehaene, S., Tzourio, N., Frak, V., Raynaud, L., Cohen, L., Mehler, J., et al., 1996.
Cerebral activations during number multiplication and comparison: a PET
study. Neuropsychologia 34, 10971106.
Delazer, M., Benke, T., 1997. Arithmetic facts without meaning. Cortex 33, 697710.
Delazer, M., Benke, T., Trieb, T., Schocke, M., Ischebeck, A., 2006a. Isolated numerical
skills in posterior cortical atrophyan fMRI study. Neuropsychologia 44, 1909
1913.
Delazer, M., Domahs, F., Bartha, L., Brenneis, C., Lochy, A., Trieb, T., et al., 2003.
Learning complex arithmetican fMRI study. Cognitive Brain Research 18, 76
88.
Delazer, M., Karner, E., Zamarian, L., Donnemiller, E., Benke, T., 2006b. Number
processing in posterior cortical atrophya neuropsycholgical case study. Neuropsychologia 44, 3651.
Di Luca, S., Grana`, A., Semenza, C., Seron, X., Pesenti, M., 2006. Finger-digit compatibility in Arabic numeral processing. Quarterly Journal of Experimental Psychology 59, 16481663.
Doricchi, F., Guariglia, P., Gasparini, M., Tomaiuolo, F., 2005. Dissociation between
physical and mental number line bisection in right hemisphere brain damage.
Nature Neuroscience 8, 16631665.
Eger, E., Sterzer, P., Russ, M.O., Giraud, A.L., Kleinschmidt, A., 2003. A supramodal
number representation in human intraparietal cortex. Neuron 37, 719725.
Feigenson, L., Dehaene, S., Spelke, E.S., 2004. Core systems of number. Trends in
Cognitive Sciences 8, 307314.
Fias, W., 2001. Two routes for the processing of verbal numbers: evidence from the
SNARC effect. Psychological Research 65, 250259.
Fias, W., Brysbaert, M., Geypens, F., dYdewalle, G., 1996. The importance of
magnitude information in numerical processing: evidence from the SNARC
effect. Mathematical Cognition 2, 95110.
Fias, W., Fischer, M.H., 2005. Spatial representation of number. In: Campbell, J.I.D.
(Ed.), Handbook of Mathematical Cognition. Psychology Press, New York, pp.
4354.
Fias, W., Lammertyn, J., Caessens, B., Orban, G.A., 2007. Processing of abstract
ordinal knowledge in the horizontal segment of the intraparietal sulcus. Journal
of Neuroscience 27, 89528956.
Fias, W., Lammertyn, J., Reynvoet, B., Dupont, P., Orban, G.A., 2003. Parietal representation of symbolic and non-symbolic magnitude. Journal of Cognitive Neuroscience 15, 4756.
Fias, W., Lauwereyns, J., Lammertyn, J., 2001. Irrelevant digits affect feature-based
attention depending on the overlap of neural circuits. Cognitive Brain Research
12, 415423.
Fischer, M.H., 2001. Number processing induces spatial performance biases. Neurology 57, 822826.
Fischer, M.H., 2003. Spatial representations in number processing: evidence from a
pointing task. Visual Cognition 10, 493508.
Fischer, M.H., Castel, A.D., Dodd, M.D., Pratt, J., 2003. Perceiving numbers causes
spatial shifts of attention. Nature Neuroscience 6, 555556.
Gallistel, C.R., Gelman, R., 1992. Preverbal and verbal counting and computation.
Cognition 12, 4374.
Gallistel, C.R., Gelman, R., 2000. Non-verbal numerical cognition: from reals to
integers. Trends in Cognitive Sciences 4, 5965.
Galton, F., 1880. Visualised numerals. Nature 21, 252256.
Gelman, R., Butterworth, B., 2005. Number and language: how are they related?
Trends in Cognitive Sciences 9, 610.
Gerstmann, J., 1940. Syndrome of nger agnosia, disorientation for right and
left, agraphia, and acalculia. Archives of Neurology and Psychiatry 44, 398
408.
Gevers, W., Caessens, B., Fias, W., 2005. Towards a common processing architecture
underlying Simon and SNARC effects. The European Journal of Cognitive Psychology 17, 659673.
Gevers, W., Reynvoet, B., Fias, W., 2003. The mental representation of ordinal
sequences is spatially organized. Cognition 87, B87B95.
Gevers, W., Reynvoet, B., Fias, W., 2004. The mental representation of ordinal
sequences is spatially organized: evidence from days of the week. Cortex 40,
171172.
Girelli, L., Lucangeli, D., Butterworth, B., 2000. The development of automaticity in
accessing number magnitude. Journal of Experimental Child Psychology 76,
104122.
Gobel, S., Calabria, M., Farne, A., Rossetti, Y., 2006. Parietal rTMS distorts the mental
number line: simulating spatial neglect in healthy subjects. Neuropsychologia
44, 860868.

1371

Gobel, S., Walsh, V., Rushworth, M.F., 2001. The mental number line and the human
angular gyrus. Neuroimage 14, 12781289.
Gold, M., Adair, J.C., Jacobs, D.H., Heilman, K.M., 1995. Right-left confusion in
Gerstmanns syndrome: a model of body centered spatial orientation. Cortex
31, 267283.
Grana`, A., Hofer, R., Semenza, C., 2006. Acalculia from a right hemisphere lesion
dealing with where in multiplication procedures. Neuropsychologia 44,
29722986.
Grifn, S.A., Case, R., Siegler, R.S., 1994. Rightstart: providing the central conceptual
prerequisites for rst formal learning of arithmetic to students at risk for school
failure. In: McGilly, K. (Ed.), Classroom Lessons: Integrating Cognitive Theory
and Classroom Practice. M.I.T. Press, Cambridge, pp. 2550.
Gruber, O., Indefrey, P., Steinmetz, H., Kleinschmidt, A., 2001. Dissociating neural
correlates of cognitive components in mental calculation. Cerebral Cortex 11,
350359.
Guariglia, C., Padovani, A., Pantano, P., Pizzamiglio, L., 1993. Unilateral neglect
restricted to visual imagery. Nature 364, 235237.
Hartje, W., 1987. The effect of spatial disorders on arithmetical skills. In: Deloche,
G., Seron, X. (Eds.), Mathematical Disabilities. Erlbaum, Hillsdale, NJ, pp.
121125.
Hatta, T., Hirose, T., Ikeda, K., Fukuhara, H., 1989. Digit memory of soroban experts:
evidence of utilization of mental imagery. Applied Cognitive Psychology 7, 265
281.
Hayes, J.R., 1973. On the function of visual imagery in elementary mathematics. In:
Chase, W.G. (Ed.), Visual Information Processing. Academic Press, New York,
pp. 177214.
Heathcote, D., 1994. The role of visual-spatial working memory in the mental
addition of multi-digit addends. Cahiers de Psychologie Cognitive 13, 207245.
Hecaen, H., Angelergues, R., Houillier, S., 1961. Les varietes cliniques des acalculies
au cours des lesions retro-rolandiques: approche statistique du proble`me.
Revue Neurologique 105, 85103.
Heilman, K.M., Valenstein, E., 1979. Mechanisms underlying hemispatial neglect.
Annals of Neurology 5, 166170.
Henschen, S.E., 1919. Ueber sprach- musik- und rechen-. Mechanismen und ihre
lokalisationen im grosshirn. Zeitschrift fur die gesamte Neurologie und Psychiatrie 52, 273298.
Hitch, G.J., 1978. The role of short-term working memory in mental arithmetic.
Cognitive Psychology 10, 302323.
Hubbard, E.M., Piazza, M., Pinel, P., 2005. Interactions between number and space in
parietal cortex. Nature Reviews Neuroscience 6, 435448.
Hung, Y.-h., Hung, D.L., Tzeng, O.J.-L., Wu, D.H., 2008. Flexible spatial mapping of
different notations of numbers in Chinese readers. Cognition 106, 14411450.
Ishihara, M., Jacquin-Courtois, S., Flory, V., Salemme, R., Imanaka, K., Rossetti, Y.,
2006. Interaction between space and number representations during motor
preparation in manual aiming. Neuropsychologia 44, 10091016.
Ito, Y., Hatta, T., 2004. Spatial structure of quantitative representation of numbers:
evidence from the SNARC effect. Memory and Cognition 32, 662673.
Jewesbury, E.C.O., 1969. Parietal lobe syndromes. In: Vinken, P.J., Bruyn, G.W.
(Eds.), Handbook of Clinical Neurology. North Holland, Amsterdam, pp. 680
699.
Keus, I.M., Schwarz, W., 2005. Searching for the functional locus of the SNARC effect:
evidence for a response related origin. Memory & Cognition 33, 681695.
Kleist, K., 1934. Gehirnpathologie. Johann Ambrosius Barth, Leipzig.
ez, R.E., 2000. Where Mathematics Comes From: How the Embodied
Lakoff, G., Nun
Mind Brings Mathematics into Being. Basic Books, New York.
Lee, K.M., Kang, S.Y., 2002. Arithmetic operation and working memory: differential
suppression in dual tasks. Cognition 83, B63B68.
LeFevre, J.A., Kulak, A.G., 1994. Individual differences in the obligatory activation of
addition facts. Memory & Cognition 22, 188200.
Lemer, C., Dehaenene, S., Spelke, E., Cohen, L., 2003. Approximate quantities and
exact number words: dissociable systems. Neuropsychologia 41, 19421958.
Logie, R., Baddeley, A., Mane, A.M., Donchin, E., Sheptak, R., 1989. Working memory
in the acquisition of complex cognitive skills. Acta Psychologica 71, 5387.
Logie, R.H., Gilhooly, K.J., Wynn, V., 1994. Counting on working memory in arithmetic problem solving. Memory & Cognition 22, 395410.
Mapelli, D., Rusconi, E., Umilta`, C., 2003. The SNARC effect: an instance of the Simon
effect? Cognition 88, B1B10.
Mayer, E., Martory, M.D., Pegna, A.J., Landis, T., Delavelle, J., Annoni, J.M., 1999. A
pure case of Gerstmann syndrome with a subangular lesion. Brain 122, 1107
1120.
McCloskey, M., Caramazza, A., Basili, A., 1985. Cognitive mechanisms in number
processing and calculation: evidence from dyscalculia. Brain and Cognition 4,
171196.
Menon, V., Rivera, S.M., White, C.D., Glover, G., Reiss, A.L., 2000. Dissociating
prefrontal and parietal cortex activation during arithmetic processing. Neuroimage 12, 357365.
Morris, H.H., Luders, H., Lesser, R.P., Dinner, D.S., Hahn, J., 1984. Transient neuropsychological abnormalities (including Gerstmanns syndrome) during cortical
stimulation. Neurology 34, 877883.
Moyer, R.S., Landauer, T.K., 1967. The time required for judgments of numerical
inequality. Nature 215, 15191520.
Naccache, L., Dehaene, S., 2001. Unconscious semantic priming extends to novel
unseen stimuli. Cognition 80, 215229.
Nieder, A., 2005. Counting on neurons: the neurobiology of numerical competence.
Nature Reviews Neuroscience 6, 177190.

1372

M.D. de Hevia et al. / Neuroscience and Biobehavioral Reviews 32 (2008) 13611372

Nieder, A., Diester, I., Tudusciuc, O., 2006. Temporal and spatial enumeration
processes in the primate parietal cortex. Science 313, 14311435.
Nieder, A., Miller, E., 2004. A parieto-frontal network for visual numerical information in the monkey. Proceedings of the National Academy of Sciences of United
States of America 101, 74577462.
Nieder, A., Miller, E.K., 2003. Coding of cognitive magnitude: compressed scaling of
numerical information in the primate prefrontal cortex. Neuron 37, 149157.
Noel, M.P., 2005. Finger gnosia: a predictor of numerical abilities in children? Child
Neuropsychology 11, 413430.
Noel, M.P., Desert, M., Aubrun, A., Seron, X., 2001. Involvement of short-term
memory in complex mental calculation. Memory & Cognition 29, 3442.
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. Cognitive Neuropsychology 10, 317339.
Noel, M.P., Seron, X., 1997. On the existence of an intermediate semantic representation. Journal of Experimental Psychology: Learning, Memory, and Cognition 23, 697720.
Nuerk, H.C., Geppert, B.E., van Herten, M., Willmes, K., 2002. On the impact of
different number representations in the number bisection task. Cortex 38, 691
715.
Nuerk, H.C., Wood, G., Willmes, K., 2005. The universal SNARC effect: the association
between number magnitude and space is amodal. Experimental Psychology 52,
187194.
Oliveri, M., Rausei, V., Koch, G., Torriero, S., Turriziani, P., Caltagirone, C., 2004.
Overestimation of numerical distances in the left side of space. Neurology 63,
21392141.
Paterson, S.J., Girelli, L., Butterworth, B., Karmiloff-Smith, A., 2006. Are numerical
impairments syndrome specic? Evidence from Williams syndrome and
Downs syndrome. Journal of Child Psychology and Psychiatry 47, 190204.
Peritz, G., 1918. Zur pathopychologie des rechnens. Deutsche Zeitshrift fur Nervenheilkunde 61, 234340.
Pesenti, M., Thioux, M., Seron, X., De Volder, A., 2000. Neuroanatomical substrates of
Arabic number processing, numerical comparison and simple addition: a PET
study. Journal of Cognitive Neuroscience 12, 461479.
Pesenti, M., Zago, L., Crivello, F., Mellet, E., Samson, D., Duroux, B., et al., 2001.
Mental calculation in a prodigy is sustained by right prefrontal and medial
temporal areas. Nature Neuroscience 4, 103107.
Piazza, M., Izard, V., Pinel, P., Le Bihan, D., Dehaene, S., 2004. Tuning curves for
approximate numerosity in the human intraparietal sulcus. Neuron 44, 547
555.
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., Rivie`re, D., LeBihan, D., 2001. Modulation of parietal activation
by semantic distance in a number comparison task. Neuroimage 14, 1013
1026.
Priftis, K., Zorzi, M., Meneghello, F., Marenzi, R., Umilta`, C., 2006. Explicit vs. implicit
processing of representational space in neglect: dissociations in accessing the
mental number line. Journal of Cognitive Neuroscience 18, 680688.
Restle, F., 1970. Speed of adding and comparing numbers. Journal of Experimental
Psychology 83, 274278.
Rickard, T.C., Romero, S.G., Basso, G., Wharton, C., Flitman, S., Grafman, J., 2000. The
calculating brain: an fMRI study. Neuropsychologia 38, 325335.
Rivera, S.M., Reiss, A.L., Eckert, M.A., Menon, V., 2005. Developmental changes in
mental arithmetic: evidence for increased functional specialization in the left
inferior parietal cortex. Cerebral Cortex 15, 17791790.
Roeltgen, D.P., Sevush, S., Heilman, K.M., 1983. Pure Gerstmanns syndrome from a
focal lesion. Archives of Neurology 40, 4647.
Rosselli, M., Ardila, A., 1989. Calculation decit in patients with right and left
hemisphere damage. Neuropsychologia 27, 607617.
Rossetti, Y., Jacquin-Courtois, S., Rode, G., Ota, H., Michel, C., Boisson, D., 2004. Does
action make the link between number and space representation? Psychological
Science 15, 426430.
Rossetti, Y., Rode, G., 2002. Reducing spatial neglect by visual and other sensory
manipulations: non-cognitive (physiological) routes to the rehabilitation of a
cognitive disorder. In: Karnath, H.-O., Milner, A.D., Vallar, G. (Eds.), The Cognitive and Neural Bases of Spatial Neglect. Oxford University Press, Oxford, pp.
375396.
Rusconi, E., Kwan, B., Giordano, B.L., Umilta`, C., Butterworth, B., 2006. Spatial
representation of pitch height: the SMARC effect. Cognition 99, 113129.

Sagiv, N., Simner, J., Collins, J., Butterworth, B., Ward, J., 2006. What is the relationship between synaesthesia and visuo-spatial number forms? Cognition 114
128.
Sandrini, M., Rossini, P.M., Miniussi, C., 2004. The differential involvement of
inferior parietal lobule in number comparison: a rTMS study. Neuropsychologia
42, 19021909.
Schenkenberg, T., Bradford, D.C., Ajax, E.T., 1980. Line bisection and unilateral visual
neglect in patients with neurologic impairment. Neurology 30, 509517.
Schwarz, W., Keus, I.M., 2004. Moving the eyes along the mental number line:
comparing SNARC effects with saccadic and manual responses. Perception &
Psychophysics 66, 651664.
Schwarz, W., Muller, D., 2006. Spatial associations in number-related tasks: a
comparison of manual and pedal responses. Experimental Psychology 53, 415.
Seron, X., Pesenti, M., Noel, M.P., Deloche, G., Cornet, J.A., 1992. Images of numbers,
or when 98 is upper left and 6 sky blue. Cognition 44, 159196.
Shuman, M., Kanwisher, N., 2004. Numerical magnitude in the human parietal lobe:
tests of representational generality and domain specicity. Neuron 44, 557
569.
Siegler, R.S., Lemaire, P., 1997. Older and younger adults strategy choices in
multiplication: testing predictions of ASCM using the choice/no-choice method.
Journal of Experimental Psychology: General 126, 7192.
Simon, T.J., 1997. Reconceptualizing the origins of number knowledge: a nonnumerical account. Cognitive Development 12, 349372.
Simon, O., Mangin, J.-F., Cohen, L., Bihan, D.L., Dehaene, S., 2002. Topographical
layout of hand, eye, calculation, and language-related areas in the human
parietal lobe. Neuron 33, 475487.
Singer, H.D., Low, A.A., 1933. Acalculia (henschen): a clinical study. Archives of
Neurology and Psychiatry 29, 476498.
Smith, S.B., 1983. The Great Mental Calculators. The Psychology, Methods, and Lives
of Calculating Prodigies, Past and Present. Columbia University Press, New York.
Spelke, E., Dehaene, S., 1999. Biological foundations of numerical thinking.
Response to T.J. Simon. Trends in Cognitive Sciences 3, 365366.
Strub, R.L., Geschwind, N., 1983. Localization in Gerstmann syndrome. In: Kertesz,
A. (Ed.), Localization in Neuropsychology. Academic Press, New York, pp. 295
321.
Temple, E., Posner, M.I., 1998. Brain mechanisms of quantity are similar in 5-yearold children and adults. Proceedings of the National Academy of Sciences of
United States of America 95, 78367841.
Trbovich, P.L., LeFevre, J.A., 2003. Phonological and visual working memory in
mental addition. Memory & Cognition 31, 738745.
Vallar, G., 1998. Spatial hemineglect in humans. Trends in Cognitive Sciences 2, 87
97.
Vallar, G., 2000. The methodological foundations of human neuropsychology:
studies in brain-damaged patients. In: Boller, F., Grafman, J., Rizzolatti, G.
(Eds.), Handbook of Neuropsychology. second ed. Elsevier, Amsterdam, pp.
305344.
Vallar, G., 2007. Spatial neglect, Balint-Homes and Gerstmanns syndrome, and
other spatial disorders. CNS Spectrum 12, 527536.
Vuilleumier, P., Ortigue, S., Brugger, P., 2004. The number space and neglect. Cortex
40, 399410.
Walsh, V., 2003. A theory of magnitude: common cortical metrics of time, space and
quantity. Trends in Cognitive Neurosciences 7, 483488.
Zago, L., Pesenti, M., Mellet, M., Crivello, F., Mazoyer, B., Tzourio, N., 2001. Neural
correlates of simple and complex mental calculation. Neuroimage 13, 314
327.
Zago, L., Tzourio-Mazoyer, N., 2002. Distinguishing visuospatial working memory
and complex mental calculation areas within the parietal lobes. Neuroscience
Letters 331, 4549.
Zebian, S., 2005. Linkages between number concepts, spatial thinking, and directionality of writing: the SNARC effect and the reverse SNARC effect in English
and Arabic monoliterates, biliterates, and illiterate Arabic speakers. Journal of
Cognition and Culture 5, 165190.
Zhou, X., Chen, C., Chen, L., Dong, Q., 2008. Holistic or compositional representation
of two-digit numbers? Evidence from the distance, magnitude, and SNARC
effects in a number-matching task. Cognition 106, 15251536.
Zorzi, M., Priftis, K., Meneghello, F., Marenzi, R., Umilta`, C., 2006. The spatial
representation of numerical and non-numerical sequences: evidence from
neglect. Neuropsychologia 44, 10611067.
Zorzi, M., Priftis, K., Umilta`, C., 2002. Brain damage: neglect disrupts the mental
number line. Nature 417, 138139.