You are on page 1of 11

Neuroscience and Biobehavioral Reviews 57 (2015) 209–219

Contents lists available at ScienceDirect

Neuroscience and Biobehavioral Reviews
journal homepage: www.elsevier.com/locate/neubiorev

Review

Mental number space in three dimensions
Bodo Winter a,∗ , Teenie Matlock a , Samuel Shaki b , Martin H. Fischer c
a
Cognitive and Information Sciences, University of California, Merced, 5200 North Lake Rd., 95340 Merced, CA, USA
b
Department of Behavioral Sciences, Ariel University, Ariel 40700, Israel
c
Division of Cognitive Sciences, University of Potsdam, Karl-Liebknecht-Str. 24/25, 14476 Potsdam OT Golm, Germany

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

Article history: A large number of experimental findings from neuroscience and experimental psychology demonstrated
Received 10 March 2015 interactions between spatial cognition and numerical cognition. In particular, many researchers posited
Received in revised form 1 September 2015 a horizontal mental number line, where small numbers are thought of as being to the left of larger
Accepted 8 September 2015
numbers. This review synthesizes work on the mental association between space and number, indicating
Available online 10 September 2015
the existence of multiple spatial mappings: recent research has found associations between number and
vertical space, as well as associations between number and near/far space. We discuss number space
Keywords:
in three dimensions with an eye on potential origins of the different number mappings, and how these
Embodiment
Intra-parietal sulcus
number mappings fit in with our current knowledge of brain organization and brain-culture interactions.
Mental number line We derive novel predictions and show how this research fits into a general view of cognition as embodied,
Metaphor grounded and situated.
Neglect © 2015 Elsevier Ltd. All rights reserved.
Spatial cognition
SNARC

Contents

1. Spatial representation of numbers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
2. Empirical evidence for 3-dimensional mental magnitudes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
2.1. Horizontal spatial–numerical associations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
2.2. Vertical spatial–numerical associations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
2.3. Distance-based or “sagittal” SNARC effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
3. Where do horizontal, vertical and radial SNARC effects come from? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
3.1. Hebbian learning, neuronal recycling and multi-causal origins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
3.2. Sources of horizontal mappings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
3.3. Sources of vertical mappings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
3.4. Sources of distance-based mappings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
3.5. An alternative account: polarity correspondence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214
4. Relationships between SNARC effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
4.1. Pitting horizontal and vertical SNARC effects against each other . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
4.2. Relating origins to predictions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216
5. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217

∗ Corresponding author.
E-mail addresses: bodo@bodowinter.com (B. Winter), tmatlock@ucmerced.edu (T. Matlock), samuel shaki@hotmail.com (S. Shaki), martinf@uni-potsdam.de
(M.H. Fischer).

http://dx.doi.org/10.1016/j.neubiorev.2015.09.005
0149-7634/© 2015 Elsevier Ltd. All rights reserved.

. Salillas et al. attention toward the right (Vuilleumier et al. When patients with right parietal damage and left magnitudes neglect are asked. 2005. Fischer and Campens.. 2014. participants are 2014) can induce corresponding arithmetic biases. with a focus on the embodied. Many studies et al. Across all these studies. 2014). when partic- to right visual space. the SNARC effect has large numbers (Shaki and Fischer. 2008. 1. 2014. When asked to quickly classify single digits by their parity (“is this number odd or even?”).1. with a right side while performing a spatial task. van Dijck ordered magnitudes. 2010). 1992. 1). 2006). 2004... with site direction of this associative link. observing eyes gazing in sions (see Fig. horizontal eye position predicts the As will be reviewed below. 2015). 2014) and even processing Random number generation (RNG) tasks have also been used left.. 2008. the spatial association of numbers magnitude of the next number that a participant generates in a extends from the horizontal into the vertical and radial dimen. This finding is taken to suggest that left neglect either abolishes The original SNARC study (Dehaene et al. an impairment that results in the loss of the cognitive rep- resentation of space contralateral to the side of their brain lesion 2. For example. A key behavioral finding that documents the close link between space and numbers is the Spatial–Numerical Association of Response Fig. 2008).vs. 2008. and more quickly to larger numbers. emphasizes the value of taking an embodied. when participants perform random number generation while line.. when solving addition problems and more leftward when solving and Fischer and Knops. therefore.b. 2014). 2006. Dodd et al. / Neuroscience and Biobehavioral Reviews 57 (2015) 209–219 1.. Horizontal spatial–numerical associations they erroneously respond “5” instead of “4” (Zorzi et al.. Simi- that people maintain a horizontally oriented “mental number larly. 2004. specifically. Chapman et al. 2007. Umiltà et al..” where small numbers are represented toward the left of walking. (Fischer. Marghetis et al.. “What number is half-way between 2 and 6?” they tend to exhibit a bias toward larger numbers. A similar bias to neglect small numbers observed across a wide range of measures and body parts. 2010). see Zanolie and Pecher. 2014. turn left more often than right when they say small compared to 1880a. free hand writing (Perrone et al. Masson and Pesenti. in which smaller Pinhas et al. A common interpretation of the SNARC effect is numbers when looking to the left (Loetscher et al.. numbers shift attention to left visual space. grounded and situated Converging evidence for a horizontal spatial association of num- approach to studying knowledge representations in the human bers comes from neurological patients who suffer from hemispatial mind.. Pinel et al. Winter et al. They have also been and eye movements (Schwarz and Keus. including can be induced in healthy adults with transcranial magnetic stim- foot movements (Schwarz and Müller. 2014. Kaufmann et al.. horizontal number-space associations have not just Song and Nakayama. The neural substrate for this close interaction is thought to be the bilateral intra-parietal sulcus which implements both spatial and numerical processing (Hubbard et al. Klein et al.. Galton..b. they systematically point more rightward et al. observed in mental arithmetic (Pinhas and Fischer. and smaller Shaki.. 1970). 2014.. When doing this while simulta- button (Dehaene et al. 2009. McCrink 2010. Our review. 2006). Werner and Raab. even when no lateralized effectors or move. Similar interactions have been 2009. Fischer and relatively larger numbers when looking to the right.. they produce larger numbers before taking an instructed larger numbers. 2014. 2014. Although Dehaene and colleagues were not the right turn instead of an instructed left turn. Winter et al. (Grade et al. or at least biases behavioral interaction between numerical magnitude and horizon.g. 2002). This finding has now been replicated neously performing horizontal head movements.. Galfano by pointing to a visually presented horizontal line that represents et al. 2012.. Even without explicit greatly popularized this idea.. Documenting the oppo- reliably associated with smaller magnitudes.. cultural and situated nature of such knowledge. 2014). 1993). ipants indicate the outcome of addition and subtraction problems ments are involved (Fischer et al. Seron et al. When partic- these multiple spatial–numerical mappings. 2008. sequence (Loetscher et al. 2008. Restle. tally arranged response buttons. 2014. All these mappings ipants observe others gazing toward the left. Code (SNARC) effect. Ruiz Fernández et al. 2003.210 B. 2012). right-branching syntactic structures (Scheepers and Sturt.. Finally... neglect. Wiemers et al. 2014).. 2004. healthy adults typically respond more quickly to smaller numbers with a left asked to call out a sequence of numbers as randomly as possible side button. Werner and Raab. Time and time again.. and larger numbers. Fischer et al.. directional physical activities larger magnitudes.. Knops et al.. have also reported an “attentional SNARC effect”.. Spatial representation of numbers Understanding the nature of knowledge representation in the brain is perhaps the most fundamental challenge for cognitive sci- entists and neuroscientists. to explore spatial–numerical associations. 2014. The domain of numerical knowledge lends itself to investigating this topic because of its universality and practical relevance.. Loetscher et al. 2004). The current review provides an update on recent insights into the cognitive and neural representation of num- ber knowledge. they generate smaller for thinking about numbers appear to stem from sensory-motor numbers than when they observe eyes looking toward the right interactions with the world around us. 2008a). 2011.. 2003b. left space is subtraction problems (Pinhas et al.. Empirical evidence for 3-dimensional mental (Karnath. been found in simple number processing. 2014). and right space.. and they spontaneously first to propose a spatial representation for numbers (e. We discuss the origin of and the relation among a particular direction also affects number generation. Here. . Finally. 1993) reported a the representation of small (=left-sided) numbers. they generate and extended hundreds of times (Wood et al. 2009a. (Wiemers et al. 2. Spatial–numerical associations along three dimensions. research on the mental representation of numbers has revealed a close connection between numerical and spatial cognition. Goffaux et al.. movements of the limbs. 2014. pointing movements ulation over right posterior parietal cortex (Göbel et al. 2006). 2013). Zorzi et al. for recent discussion.

(2014) used a setup with two buttons on top of each other. 2013). Hence. books a day (which implies a comparatively larger quantity) facili. to express magnitudes is currently unknown. mid-sagittal axis investigated real or apparent movement along 1 However. 2012). or Italian un by a body-lifting device.. labeled the “low” key. 2011). Similarly. The eyes moved left when tures when talking about high numbers. Similar use of the terms left and right when being moved downward (Hartmann et al. extending along the front/back tion problems faster when they are moving upward in an elevator.. Ito and Hatta. The spatial imagery involved in Keus. When participants generate numbers while being moved upward German die Preise sind gestiegen ‘prices have risen’. vertical SNARC effects. 2008. spontaneous ges- and right when the interval was named in increasing order (e. 1987.2. Shaki ing upward movements interfered with subtraction (Wiemers et al. productivity is argued to show that metaphor is more than a 2010). For instance. and downward oriented the interval was named in decreasing numerical order (e. where ticipants verbally responded to spoken questions such as “which people were observed to perform upward directed manual ges- number is halfway between 4 and 8?”. 2014). Katz et al. Kövecses. 1980.e.. including the original SNARC study (Dehaene et al. 1993). Consistent with this. These studies often find that people respond of a visual target positioned relatively low on a computer screen.e. par. such as “radial” (Hartmann et al. 2014). they tend to generate higher numbers when looking process readily extended to other expressions such as plummet- up (Winter and Matlock. However. for example. gas research. 1980. Close reading of the method sections of published component when there was a mismatch between response and work reveals that several studies interpret results in terms of truly implied quantity (i.. 2013). in contrast. and they describe interest rates. A primary source of evidence Lakoff and Johnson. it should be noted that the arithmetical outcomes in Wiemers et al. and to large numbers. another study found a horizontal effect. Thus. ward from the body along the horizontal plane (cf. such as foot. ones when talking about low numbers.).. perceiving or performing downward response buttons along the mid-sagittal axis on a table top (Gevers movements interfered with addition while perceiving or perform.. 1998. when participants generate numbers while turning their head up Using vertical language to describe quantity is a productive and down. ically from the mid-sagittal axis. ERP measurements revealed a stronger N400 Tversky. when the response setup used a response There are also “attentional SNARC” effects along the vertical axis. Specifically. 1987. for vertical spatial–numerical mappings comes from the RNG task: talk about quantities in terms of height or vertical position (e. turing during speaking provides yet another source of evidence 4–8). 2008). numbers with the “up” button. Such predict numerical magnitude in the RNG task (Loetscher et al. Gibbs. and to smaller numbers with the “down” button. axis from the body into the distance. 2007.. The fashion. an analysis of T. whereas the results of subtraction problems were on operations (Tyler Marghetis. 8–4).. or words associated with the concepts “more” and “less” (Guan et al. labeled the “high” key. 2012) showed that The few studies that investigated vertical space/magnitude reading sentences such as More runs were being scored this game associations can be subdivided according to whether they mean leads to quicker upward directed responses. words associated with low space.. 2012). there are the above- and subtraction problems faster when moving downward (Lugli mentioned studies on “vertical” SNARC effects that positioned the et al. Winter et al. Other languages. process that structures how we think and reason (Lakoff. 2004... a table (Ito and Hatta. in the sense of extending out- bird (Lachmair et al.. same spontaneous association between upward gestures and larger More closely mirroring the classic SNARC paradigm. Distance-based or “sagittal” SNARC effects the lower key or the upper key. perhaps because early ber or This is a low number. processing metaphoric language about quantities is revealed by but not a vertical one (Loetscher et al. research on linguistic metaphors also provides evidence for vertical space/magnitude associations. 2006). with the farther button. Upward eye movements are followed by relatively larger rhetorical or poetic device. the results of addition ciation between space and magnitudes rather than between space and arithmetical problems were on average 46. 2004). more quickly to small numbers with the nearer response button On the other hand. Santens and Just as with the horizontal dimension. and Fischer. 2011).. In that case. whereas processing These distance-based effects have sometimes been called large numbers facilitates words associated with high space. reading sentences such as The old man had 2 books in his i.. 2012).g. It is instead viewed as a dynamic numbers. et al... average 22. participants respond faster to larger Gibbs. 2009).3. We commonly talk Fewer runs were being scored this game leads to quicker downward about far as “up” and near as “down” when mentioning positions directed responses. the result by Wiemers et al. Hartmann magnitudes can occur when adults are asked to point at the et al. A similar result was obtained with Chinese on a horizontal plane. Müller and Schwarz.g. they tend to generate larger numbers than numero alto ‘a high number’).9. First.V. news broadcasts (Winter et al. the major share of evidence comes specif- reported in relation to mental arithmetic. Other studies supporting a critical role of the 2014)1 . Vertical spatial–numerical associations Finally. i.e. 2006. . However. Müller and Schwarz. Shaki book case (which implies a small quantity). 2007. Participants solve addi... pad or keyboard that was in fact positioned on a horizontal plane.g. rather than hand movements. 2008a). In this case. Moreover. Kövecses. downward eye movements by relatively smaller ones. emphasized horizontal associations. vertical eye movements ing incomes. Schwarz and Matlock. skyrocketing prices or ever ascending tax rates. vertical effects have been Gevers. more/down and less/up). 2002. locations where they imagine small and large numbers in space Participants made parity judgments faster in response to larger (Fischer and Campens. 2011). when asked to respond with eye movements Lakoff and Johnson. facilitates the detection and Fischer. reading sentences such as The old man read 2 (close to the body). p.. the “top” half of a page. vertical SNARC. Such effects can be tates detection of a high target (Pecher and Boot. In a related study. A related study with a similar ver- tical setup of response keys (Sell and Kaschak. but the eyes did not move up or down in a systematic for vertical associations between space and number above. (2014) could potentially be an asso- (2014) were not balanced for addition and subtraction. when saying the “N” key is below the “U” key on a keyboard (cf.c. 2014). 2004. 2002). / Neuroscience and Biobehavioral Reviews 57 (2015) 209–219 211 2. regardless of whether the left hand was assigned 2. too. found on a numeric keypad with the “2” and “8” keys as near and processing small numbers facilitates the subsequent processing of far responses (Holmes and Lourenco.e. Moreover. prices and other abstract quantities as rising or falling (Lakoff. 2013). English Relatively little attention has been given to associations speakers frequently make statements such as This is a high num- between numbers and vertical space. B.2.. 1994) and that is driven by mental simulation (Gibbs and numbers when looking upward (i. reading the label “vertical” literally or metaphorically.

” there also support from multiple sources.. Western societies conventionally start counting with their left hand ittal?) of distance-based effects is not entirely clear at present. 1984. Shaki This.” In this setup. 2009. all studies in support of these effects have used the front/back Pitt and Casasanto.. people in many (Winter et al. They also Horizontal spatial associations are commonly taken as evidence generate smaller numbers when physically moved backward by for a spatially oriented “mental number line. 2014). Finally. see Santiago et al. gest a “radial” effect. such as the fact that when children count from a central position indicate a larger size. 2012.. e. visual primes affect temporal that it is fruitful to consider numerical cognition as a “greedy” sys.1. (Göbel et al. just as gesturing an array of objects.. 2. Sources of horizontal mappings that induce the visual perception of backward motion. while others orient them along the horizontal and tural reading/writing conventions (Zebian. Anderson. they habitually start on the left side (Briars away from the body’s midpoint is used to indicate large quantities and Siegler. extended the range for this horizontal mapping preference well tion. However. 2013.. 1993). Geary et al. in Western cafes and restaurants. where traditional left-to-right SNARC effects were observed when Implicitly. Hence. / Neuroscience and Biobehavioral Reviews 57 (2015) 209–219 the front/back axis. In the case of spatial–numerical mappings.. This num- bers when physically moved forward. 2014). However. 2009. 2011. cf. Zhang and You. 2007) that develops so as to be consistent with other et al. 2005. Bender and Beller. before most children are fluent readers and writers. Santens and Gevers (2008) in children only after 3 years of schooling provided initial support found that small numbers were associated with “close” responses for this argument (Berch et al. This includes the co-opting about earlier events and to the right when talking about later events of brain areas previously evolved for other tasks. there is convergent Using the term “radial” rather than “sagittal” implies circular cultural support for the orientation of the number line. 1991). 2003. whereas positive numbers are left-to-right axis (such as during reading) uses similar neural sub- associated with full-body movements going forward (Marghetis strates as doing mental arithmetic (Knops et al. 2006). 1999) but more recent work has and large numbers with “far” responses—regardless of orienta. for a related of number and time). they tend to generate smaller random numbers (Seno et al. Western adults are moreover constantly exposed to pictures... a confluence of cultural factors.. Alternatively. et al. this. Santens and Gevers (2008) gave participants a baseline The original SNARC study (Dehaene et al. with smaller 3. 2011). 2003a. respond quicker with their left hand to past events and with their tural practices (such as those involved in the visual representation right hand to future events (Weger and Pratt. including brain organization. judgments consistent with a left-to-right going mapping (Núñez tem (cf. and larger num. 2010. 2012). Evidence for the role of visual ciations... Walker cognition. such as spatial (Cooperrider and Núñez. 2010). in fact. 2012). Standard keyboard origins arrangements of numbers have the same orientation. For example..g. we point out that the space (Fischer. Moreover.. 2014). 2005. graphs and other representations that are structured in accordance 3. and Youngstrom. prices are often listed in columns. negative numbers are associated with full. In other words. 2009a. etc. Tversky. effect.) are increasing. in a rightward oriented setup. 1993) proposed that key “j” from which participants had to either move to a “close” key the horizontal SNARC reflects a spillover of spatial-directional scan- “h” or to a “far” key “g” to classify numbers by their magnitude.. orthography is likely not the only factor leading to a left- (2008) could also be interpreted in terms of a number-size asso. 2007). 2007. They discuss this effect in terms of a distance-based effect. besides this. expected based on recent insights into brain-culture 2012. reading and counting and calculating may share partly overlapping times associated with left space and positive numbers with right brain organization. 2008. but due to 2012). where movements away factors that play a role. when participants ber line has sometimes been called a “cultural mental number line” are asked to indicate their number associations freely in space. Spivey. is a consistency between horizontal associations of quantity and tural practices. For example.. We (Lindemann et al.. 2011) because evidence reveals a close link between they occasionally orient them along the sagittal axis (Fischer and the orientation of the horizontal mental number line and cul- Campens. Moreover. 2012). we will point out that each mapping may have convergent In line with the idea of “convergent cultural support. Winter et al. to-right bias. Elementary school children may already know these biases and reproduce them When researchers talk about the origins of space/number asso. into pre-school age (Hoffmann et al. calendars list days and interactions (Dehaene and Cohen. ning habits from reading into the domain of numerical cognition. 2014)—just like negative numbers are some. neuronal recycling and multi-causal prices on the left and larger prices on the right. 1992). they often emphasize one particular origin. together with the sagittal effects reported above. symmetry. Henik and Tzelgov. In contrast to smaller numbers are on the right side). cul.. 1982). in sorting tasks (Tversky et al. the associated For example. horizontal representations also comes from a study of Bächtold et al. 2012). and the natural world. Finger counting direction has also been will continue to use the term “sagittal” given the fact that almost shown to modulate or even reverse the SNARC effect (Fischer. suggesting that manual practices besides dimension. Casasanto and Jasmin. Pitt and Casasanto (2014) but a reverse SNARC effect (right-to-left) was observed when par- emphasize the role of counting direction (people also count from ticipants were primed with a picture of a clock face (where overall left to right). might sug. Shaki et al. many such discussions assume that one origin takes people were primed with a picture of a horizontally oriented ruler. When participants view optic flow patterns 3.b).g. Anderson (2010) points out evidence suggesting that numerals (day 1.. but de-emphasize the role of writing. the exact spatial nature (radial or sag. 2011). is.2. months from left to right and corresponding to this. Moreover. horizontal SNARC effect may not only be due to writing. the “close” key was The observation that the SNARC effect begins to become reliable “k” and the far key “l. (1998) SNARC effects stemming from writing (Dehaene et al. writing can influence the orientation of the mental number line. and —similar to the horizontal SNARC—people also neural structures (such as those involved in reading) and other cul. Hebbian learning.” according to which a body-lifting device (Hartmann et al. 2008. Hence. 2008. 2015).. Orienting spatial attention along the body movements going backward. there may be other ciation (e. vertical and radial SNARC effects with the left-to-right orientation of the mental number line (Maass come from? and Russo. For example. Opfer et al. Shaki and Petrusic. 2009). . vertical ones. 3. small numbers are placed to the left of larger numbers. Behaviorally. Where do horizontal.. When looking at early ontogeny. we will see and Cooperrider. priority over another. Such a multi-causal proposal horizontal associations of time (recently reviewed in Bonato et al.212 B. the results of Santens and Gevers Hence. it has been evolutionarily more recent practices (such as math) depend on found that speakers of English gesture toward the left when talking comparatively larger brain networks. Shaki and Fischer.

graphing conventions. when it comes to sources of vertical number-space asso.” where “more” numerosity corresponds to “more” Tversky. From the 3. linguistic expressions such as high number or low number. 2013). Seno et al. and their gestures can cal mappings might also stem from another set of conventions. as we pointed out above. The cultural representations and Youngstrom (2014) discuss connections between a sagittal number artifacts where the mapping is consistent with an upward oriented line and a sagittal time line. The distance-based effects would then stem from the fact thermometers. for example.g. On calendars and league tables.. This type of research suggests that 2 Some linguistic metaphors may actually create novel conceptual mappings. suggests that these cultural reflections from the glass. 2012).. and the level goes down. distance-based effects might be an indication of see Fischer et al. Kövecses. Sources of vertical mappings perspective of child ontogeny and Hebbian learning.g. which has been argued to stem from expressions such as right-wing about high numbers and low numbers)2 . First. cf. these apparent inconsistencies can be resolved by point. 2015... 2011.2). the major share of evidence ples run counter to this orientation. away. 2012. 2005) observe the principle of “more is up” (cf. tions between verticality and quantity. Hutchins. natural world make a set of stimuli available to the growing child ciations.” “second place” and so on. cognitive linguistic research has shown that the processing of linguistic metaphors involves active con- ceptualization of the metaphorical source domain such as space (Katz et al. gestural (Winter et al. nearby space is a small spatial magnitude ber words. which maps onto small numbers. Winter et al. including all the linguistic and graphic reflections. for example. such as distance. Whereas writing orientation along the horizontal axis corresponds to the horizontal SNARC effect (left-right). Lakoff (1987: 276). a large Instead. In other cases. 2008). left to right) and artifacts where space and number are correlated. the level rises. Thus. 2008). Bueti and Walsh. 1949). it is clear that writing where vertical space and quantity (or its symbolic expression) are direction does not explain the orientation of vertical SNARC effects: associated in a highly predictable fashion. with one telephone number tural support from space/time mappings (e.” Repeatedly observing may have a causal role in shaping space/number mappings. Fischer. but sim. Hubbard et al.e.. that both magnitudes access the same neural substrate (often pos- and floor numbers in elevators. such hearing about metaphorical vertical language in the context of number is likely as the mapping of horizontal left/right space onto political positions (Oppenheimer going to reinforce those mappings. Winter et al. 2001. 2005: pp. Grade et al.. Sources of distance-based mappings the vertical axis (top-down) would predict that smaller numbers should be associated with higher vertical space and larger numbers Distance-based effects are sometimes discussed in relation to a with lower vertical space (see Hubbard et al.. reveal a sagittal orientation (Núñez et al. to a pile. the graphic. a the- been found. If chil. compatible with sagittal mappings. “more is up” is thought to mentally engrain the corresponding map- dren are constantly exposed to practices (such as counting from ping (Lakoff. number words written in Chinese script. When using cellphones... “more is more. observed (e.. They also numbers and quantities in terms of vertical space supports vertical mental associations. not being “more” or “less” than another telephone number. 2011). Marghetis and ply identifying a specific identity. but only with Chinese participants when responding to ory of magnitude or ATOM: Walsh. B. verti.4. and may in some cases. the mental association between vertical space and cardinal num- ber. On the other hand. i. they more quickly as discussed above (Section 2. it is conceivable that the natural cor- this necessarily leads to neural and mental associations between relation between verticality and quantity is the ultimate source of co-occurring sensory-motor and conceptual activations. soon-to-occur events onto near space. 2010. effects that are radial. / Neuroscience and Biobehavioral Reviews 57 (2015) 209–219 213 From this perspective. Gibbs and Matlock. (2005) point to the potential role of spatial magnitude (i. generalized magnitude system in the brain that codes visual and cf. and future events onto far measuring cups) where “more” and “less” are clearly defined. and by consis. the use of numbers (with ticular axis. for exam- systems influence our mental representations (Bender et al. this orientation has spatial as well as temporal and motoric magnitudes (i. 2015). Speakers of English. override the small/low-large/high association. 2011. 2009. 2013: 128). Holmes and Lourenco. body height measurement devices. see Walsh. space. which maps the presence on the “ego. 1998. However. at least not in the same way we talk and Trail. Indeed. In fact. 2012). discussion in Ito and Hatta. For example. talk about looking forward In line with the idea of convergent cultural support.3. or the front/back dimension.. if anything. observes that “whenever we add more of a substance—say. Marghetis and Youngstrom. Holmes and Lourenco. calculators and the number keys on cellphones. distance-based effects. Hence. bar plots: in line with ATOM. In contrast to magnitude-based proposals. The prevalence of talking about process future related words (Hartmann and Mast. writing direction is just one of several However. bly assign magnitudes to any spatial dimension that a task provides writing orientation stands in opposition to the commonly observed (see also. which is frequently writ.. 1987.g. is a purely ordinal one. 1994. 2004). but not horizontal ones (we do not talk about right numbers and left numbers. . That we are constantly water to a glass—the level goes up.. 2002. writing orientation along 3.g. the picture is more complex. 2011). 2010). league for distance-based effects stems from experiments that test specif- tables. 2010). Remove objects from the pile or water tent space-time mappings. however. 2003....e. on the other hand. Most often. ATOM would seem to predict distance-based “1” being at the top. “first place.. For vertical SNARC effect. it does not concretely explain ing out that numbers are not used in a cardinal sense in such why distance-based mappings should be oriented along this par- situations. 1995. even though vertical SNARC effects might stem from components of cultural support for horizontal spatial mappings cultural or linguistic conventions. Lakoff and Johnson. 2003). When participants namely. 2006. many cultural exam. While ATOM is however. such as Chinese num. The argument is that participants flexi- ten top-down in columns (Hung et al. scientific graphs (e. calendars. Hence. And so do other cultural devices. ically the sagittal axis. Cognitive anthropologists have gin. Zhang and Norman. tend to emphasize quantities (e. 2014). 437–438. 2014) and linguistic associa- gent cultural support for vertical mappings of numbers onto space. to the future and thinking back to the past. others allude to cul- we use numbers in a nominal sense. However. When we add more objects surrounded by horizontal space-number mappings. Together. far space) maps onto large numbers. there is a lot of conver. ple.e. are moved forward using a body-lifting device. Gibbs. measuring cups. Instead. they probably have a deeper ori- of quantities and their symbols. the natural correlation between verticality and quantity stands next to Similar to the horizontal mappings. It has often been pointed out that verticality and quantity noted that external representations such as graphs and notation are correlated in the natural world. rather than the bottom).. 2006). This is and interacting with natural quantities that obey the principle of indeed expected because of Hebbian learning (Hebb. as has been frequently tulated to lie within the intraparietal sulcus. 1980. culture and the However. conservative and liberal left (Casasanto.” number line.

2014). which are [−] polar. processing is the same cognitive processes. slowed down. this unmarked member of a binary pair is generally Hutchinson and Louwerse. 2006. a question about a person’s height using How tall is she? but not The polarity correspondence principle furthermore makes How short is she? (cf. and the sagittal number-space mapping remains to be shown. above. 2010. 2005). However. 2012) considered in this paper. 7. Loetscher ATOM associates far space with “more”. who found spatial numerical associations in a vertical and distal space are cognitively and neuronally associated. discussion in Roettger and Domahs. and Fischer. Near and low space tal (e. 2007. we often do so dimensions. is correct about SNARC effects.. This proposal has considerable considered in this paper so far. it is not asymmetrical dimensions. whether cognitive con. and mental arithmetic (is addition [+] polar?). but it does not account for all patterns observed and makes Santens and Gevers. we would gen- from a study (Matlock et al. . we tend to ask responses to plurals). This account is based on the observation that many for late response times. . 2008b. 2008). such as tall versus short. in Roettger and Domahs. .g. we should notice that there is consistency between these with the random number generation experiments discussed above. tions for all three of the axes considered in this paper because right. Initial support for a link between time lines and number lines comes up and far would all be considered as [+] polar (e. Dehaene et al. For early response times. . many time lines and ges. The above-discussed embodied . inconsistent predictions with respect to word frequency: Linguists The pole of a dimension that can stand in for the entire dimension generally assume that the unmarked member of a pair (e. the polarity correspondence principle may be a powerful that the right side and large quantities are [+] polar. However. again. the table-top response example. seem to be tapping into some of If the stimulus and response dimension mismatch. Winter et al. if polarity correspondence support from neuroscience.) leads to reliable shifts in more frequent than their opposites left. likewise if the [−] polar left. and both and Shaki (2015). different accounts for the origins of spatial–numerical mappings. Müller and Schwarz. This is the case for many of the horizon- low and near space (Mennemeier et al. 2014). there are several strands of evidence to suggest that (Levine and McAnany. and toward their body when reading than left. This can explain the SNARC effect if one assumes So. for instance for the SNARC effects considered in this paper. we are faced with Second. are also associated in our ecology and in linguistic metaphors. First. Hence.g. and the words are (5.. These consistent with the polarity correspondence principle (faster left oppositions are often asymmetrical.g. down and near). house versus houses) with singulars 3. 1992). these effects would not be because tional association between the lower visual field and proximal of spatial mental representations per se. given their directional polarity of the stimulus dimension match. 6. 1990). large numerals. consistency with SNARC effects. Winter and Matlock. and the left explanation of SNARC effects in some of the binary tasks considered side and small quantities are [−] polar (Proctor and Cho.g. correspondence-based accounts may coexist. 15. 2015). An alternative account: polarity correspondence being faster with left responses and plurals being faster with right responses. Moreover. As such. . That is. These studies reveal a The polarity correspondence account makes the right predic- mental sagittal axis for the conceptualization of event sequences.. For example. SNARC effects would arise because of structural in the lower visual field. lapping asymmetries of the corresponding numerical and spatial When performing actions in peri-personal space. tall) is called the “unmarked” case. / Neuroscience and Biobehavioral Reviews 57 (2015) 209–219 respond more quickly with a response away from their body when and given that the corresponding adjective right is more frequent reading about future events. and reports have been made about the polarity correspondence principle cannot be the whole story neural associations between vertical and distal space. for 2011) and distance-based mappings (e. 2011)... 16. 2011: 506). This result is consistent with a spatial representation A specific proposal that can potentially account for of quantity but inconsistent with polarity correspondence. Finally. Processing temporal reasoning..g. we often do so in the upper visual field.. go/no-go task involving a single response. 1993). 2006). which. 2013). novel exper- if people (and children) automatically build up consistent neural imental evidence indicates that both space-based and polarity and cognitive associations based on sets of correlated stimuli. processing is facilitated.g. 1990).. 2006). down and near responses match with One alternative proposal for the origin of distance-based SNARC the [−] polar small numbers. it is not clear whether polarity correspondence multiple potential sources that are not mutually exclusive. Such a extends to the above-discussed associations between space and situation is to be expected if culture reflects brain organization. 2015). et al. Third.. we commonly understand a page oriented on a table in options in Ito and Hatta.5. its primary in a stroke patient with neglect of both up and far space (Shelton explanatory domain contains tasks that involve binary stimulus et al. In everyday also more frequent in spoken language and texts (for discussion see language. ultimately. but rather due to over- space and the upper visual field and distal space (Previc. was observed only Cho. and in another stroke patient with neglect of both and response dimensions. The right side is typically assumed to some inconsistent predictions. however. Holmes and Lourenco. are more fre- 2015.) or “downward” (17. vertical (e. results were dimensions have binary opposites. as shown by the case of neglect.. the polarity correspondence effects is that they might be derived from vertical effects (see principle is a unifying explanatory account for many of the effects also Holmes and Lourenco. 2015). for which there is no binary response dimension (e.g. People more quickly perceptual similarity. 2011).214 B.. what exactly counts as be [+] polar because right-handedness is the more frequent case [+] polar and [−] polar is not always straightforward (see discussion in the population (see discussion in Roettger and Domahs. The SNARC the so-called “polarity correspondence principle” (Proctor and effect in Roettger and Domahs (2015). Shaki terms of “up” (far) and “down” (near) (Tversky. Given these quent than small numerals. detect stimuli in the upper visual field when they appear distal However. 2011) showing that counting “upward” erally ask How far is it? rather than How near is it?. Arguments have been made for a func. about past events (Sell and Kaschak. but it is not the case Again. is predicted to be faster if these [+] polar poles of the respective gruency between spatial mappings of time and number underlies dimensions match with [+] polar quantities (large numbers). 2004. which more frequent than the marked member (Roettger and Domahs. tall) is (in this case. .. see also Hutchinson and Louwerse. which spatial–numerical associations along all three dimensions is regards singulars as [+] polar and plurals as [−] polar. but at different time scales: Roettger and Domahs (2015) find a SNARC effect for singu- lar versus plural nouns (e. are assumed to be [+] polar (Proctor and Cho. Critically. the polarity correspondence principle clear that the polarity correspondence principle applies to effects states that if the polarity of the response dimension and the that relate to space/time mappings. And when looking toward extra-personal similarity between response and stimulus rather than because of space. Another experiment tures of time lines have the same association (with the future and that has no binary response dimension was conducted by Fischer associated larger numerical values being in the distance).

ined environments. There are also horizontal and the vertical mapping that results from an association asymmetries of horizontal and vertical space in our environment. 2012 and Gevers have not taken into account the differences between sagittal and et al. ries may be confounded with differences in the strength of the spatial–numerical associations. For example.e. adults tend to perform smaller vertical head movements than delivered along the “right-diagonal. of small numbers with the lower and the left side. 1980). tion. Higashiyama. verti. ments. cognition and culture (Section 3). Ikeda and Takeuchi.. SNARC effects. The evidence from perceptual organization and zontal and vertical SNARC effects are of similar strengths. i. 2012. such asymmetries could matter. The logic along the horizontal than the vertical axis (Haith. 1947. On the other hand. legitimacy of generalizing from the presence of purely horizontal Table 1 suggests equivocal evidence with respect to the strength and vertical effects (as in Schwarz and Keus. . Experiment 1B) Grade et al.. Lechelt et al. 1993). aligned either horizontally or vertically (e. numbers are generally not presented diagonally or grid-like in our culture. tal. 2001) and ver- response location for small numbers to an upper right response tical lines are perceived to be longer than horizontal lines of equal location for large numbers. there is a congruency between the length (Finger and Spelt. So. we would like to point out that visual discrimination is best along the vertical and horizontal axis.g. left-top/right-bottom).) (cf. other types of perceptual and spatial asymmet- and Kaschak. 2013. Grade of numbers are unnatural or unrelated to the other documented et al. participants position numbers in three-dimensional space (Fischer inant dimension in the organization of number space” (Wiemers and Campens. 1962. Children also scan more widely and frequently (Holmes and Lourenco. bar plots. and most people more frequently move horizontally than precisely. Gevers et al. and have the advantage of fitting firmly into the body of existing Franklin and Tversky (1990) argue that when people search imag- linguistic and anthropological research. Wiemers et al. Children. we can question the that have tested at least two axes for the same participants. Are the three spatial–numerical associations (horizontal. numerical cognition is consistent with other domains of brain orga- c Results from Experiment 1. 2006). radial) equally well entrenched? Or is one axis more cognitively lists...g. one. / Neuroscience and Biobehavioral Reviews 57 (2015) 209–219 215 Table 1 smaller number on the left side is compatible with the horizontal Comparison of strength of horizontal. sagittal). vertical. 2013) and an eye tracking study (Schwarz and Keus. For example. (2010) RNG H* < V* In a similar setup.g. Pelz et al. when blindfolded some researchers asserting that vertical space may be the “predom. 1976. there is disagreement about learn numbers on the number line before they learn the Cartesian whether the horizontal or the vertical SNARC effect is stronger. Winter and Matlock. (2012) RNG H* > V* observed a sagittal but no horizontal effect. Gevers et al. More- behind these studies is that in a congruent condition. but no ver- Sagittal tical effect in the left-diagonal incongruent condition. 1982). sentation is not the preferable one” (Gertner et al.. some studies indicate that vertical effects are More generally. and others saying that “the vertical mode of repre. menu price cal. Pitting horizontal and vertical SNARC effects against each asymmetrical in perception and action—regardless of any associa- other tions to numbers. 2004). dominant than the others? Currently. Table 1 provides a list of those studies horizontal mappings. from an upper left response location to a lower right loca.g. a setup Truly vertical that is not truly vertical but relates to the sagittal effects discussed Schwarz and Keus (2004) Eye H* > V* above. rooms are generally more horizontally than vertically ent condition. 2014. grid-like mappings. verti- et al. they preferentially access objects along specific axes.” see Appelle. (2008) Eye H* > V in an incongruent diagonal condition (e. responses are over. as with cultural representations suggests that diagonal representations random number generation studies (Loetscher et al. Howard. In line with this evidence. the star indicates whether space are primarily oriented along distinct cardinal axes (horizon- the corresponding effect was found to be significant. and many diagonal orientations (as in Holmes and Lourenco.1. discussion in Tversky.. Sell against each other. Previc and Blume. 2006) to other experiments that report purely vertical and truly vertical SNARC effects. from a lower left horizontal ones (Glenn and Vilis.. Here. 1972.. it is important to ask Ito and Hatta (2004)b SNARC H* < S* whether SNARC effects are truly defined along the diagonal axis. B. a Results from Experiment 1 (Experiment 2 is diagonal).. Müller and Schwarz (2007)c SNARC H < S* Several findings suggest that associations between numbers and “>” and “<” indicates which effect is numerically larger. 2009). A similar task with Winter and Matlock (2013) RNG H* < V* a diagonal response setup on a vertically mounted touch screen Wiemers et al. cf. 2004). etc. A response to a the horizontal and the vertical axis. it has been suggested that the visual search field is of greater horizontal than vertical extent (Chaiken A few studies have directly pitted horizontal and vertical SNARC et al. rather than being map-like (as suggested by b Results not strictly speaking comparable due to dependence on hand assign.” precisely. 2013: 1354). but instead. 2012). 1992). vertical and sagittal effects for those studies number line but simultaneously incompatible with the vertical that had multiple response dimensions in the same task. Thus. responses are delivered along the “left-diagonal. 2011). 1992. vertically. In an incongru. not a diagonal axis (the so-called “oblique accounts of three-dimensional SNARC effects are equally plausible effect. there was a horizontal effect.. They observed neither a horizontal nor a vertical effect Loetscher et al. 2004) to diagonal or of vertical and horizontal SNARC.. Some studies show that hori. for example. cal or radial dimensions—but critically no diagonal orientations. Along the same lines. (2013) RNG H* > V* Hartmann et al. 2. Schwarz and Keus. there is a conflict between For the experiments reporting explicit comparisons between the horizontal and vertical mapping preferences. when pitting the horizontal and vertical axis stronger (e. 1975. nization. as effects against each other by using diagonal response mappings depicted in Fig. and likewise for lower right responses to large numbers. We can therefore question the legitimacy of generalizing from very few studies explicitly compared these two effects... in this incongruent condition. Relationships between SNARC effects the body are equally available. Horizontal and vertical space are 4. Winter et al. 2014: 12). they spontaneously exploit horizontal. with coordinate system. Holmes and Lourenco (2011. 2012). Task Results Gevers and colleagues (2006) used keypad responses.. (2014) Arithmetic H < V* found that the horizontal effect “trumps” the vertical one (Holmes Sell and Kaschak (2012) Linguistic H < V* and Lourenco. 2010. Moreover. 2011. Yet. e. Remember that broadly construed. Loetscher et al. (2006)a SNARC H* > S To interpret the relevance of these findings..” extended. rejecting the hypothesis that all directions extending from 4.

4).e.. Relating origins to predictions spontaneously mapped numbers onto different spatial axes with idiosyncratic. More studies implementing concrete to find associations between numerical magnitude and space. inclined to think about numbers horizontally or radially? That the tions. see Fig. space/number association is thought to be grounded in natural cor. p. we expect to see task-dependent differences on a priori should be most strongly correlated with vertical SNARC effects grounds. As described above. head movements along the horizontal and head movements along tal and vertical dimension is.52 and r = 0. Finally. 2013). at present many studies exploring both horizon- multifarious cultural and non-cultural phenomena that may lead to tal and vertical/sagittal space for the same subjects fail to report space/number mappings. 1993. the task dimensions? In general. 2011. SNARC effects with truly vertically aligned response buttons were found in response to lin- guistic stimuli such as More/less runs were being scored in this game. 2012. 1980). p. The visual search field is less extended along the et al. and cultural world. at least compared to horizontal effects that have no such linguistic support. more so than with horizontal SNARC effects. numerical magnitude interacts with the Fig.216 B.. Göbel et al. the screen on which the numbers were displayed in us to expect the different axes to be at least partially dissoci- Holmes and Lourenco (2012. and for other axes in other across different individuals and different tasks. low number.2. 1046) had more horizon. vertical and sagittal effects are correlated stronger for some axes in some tasks. which are known to be culturally relative than horizontal effects. hence limiting our conclusions one effect being stronger than the other in all contexts. Cohen Kadosh et al. cf. 2008) found asymmetries for movements—including eye movements (Collewijn a somewhat weaker correlation for horizontal and vertical num- et al. At present. etc. 2010. (Dehaene et al. 2. 2703] with fixation at X and center vertical location implied by words such as foot and bird (Lachmair of the ellipsoid at the black dot. 2011. at this point. and they cal comparisons if the horizontal and vertical dimensions are not interact differently with the size congruency effect and the numer- equally scaled psychophysically. / Neuroscience and Biobehavioral Reviews 57 (2015) 209–219 horizontal and vertical scenarios to contextualize number mean- ings are needed. besides its cultural reflections.. Lakoff. we might expect that sagittal SNARC effects tion 3).. Given this. reasoning about quantities of SNARC effects should be found in any culture. 2007). suggesting neural dissociation. others found that response orientation with equal physical extent in the horizon. Another prediction that can be derived from the above discus- sion is that the vertical SNARC effect—because of its connection to language (e.). as discussed above. we may not want to make claims about whether these effects are correlated. but no horizontal effects emerged in this situation. 2007. our discussion in Section 3 leads us to ask why should Gertner et al. The view that vertical SNARC effects stem from the physical relations (Fischer. with permission from Elsevier. is it the number generation (Winter and Matlock. whereas horizontal SNARC effects largely stem 1987. Fischer and Brugger. given the fact Given that spatial–numerical associations may arise from a that near/far space frequently correspond to each other in percep- wealth of different cultural and embodied phenomena (see Sec. perceived as equal. participant-specific choices.)—should be stronger in linguistic contexts. by virtue of asking participants to verbalize. Lakoff and Johnson. Winter and Matlock. it is difficult to draw hard and fast Horizontal and vertical orientations of number lines can further- conclusions about relative strength from horizontal versus verti. which. tical effects in both linguistic and non-linguistic contexts is still Source: Reprinted from Previc and Blume (1993). 1988)—we cannot take for granted that a diagonally oriented ber line bisection (r = 0. Winter et al. 2005. Gevers and colleagues (Gevers et al. the vertical SNARC effect re-emerged only when participants were primed to think about building floors More and more studies on numerical cognition are beginning (first floor. etc. because of horizontal–vertical SNARC effects (r = 0. However. Zebian. Future research needs to explore more systemati- experimental effects of spatial–numerical associations might be cally how horizontal. 1. the vertical across individuals. Visual search field sketched by [108. second floor. Similarly. Finally. Moreover. with differ- diction. 2012) suggests that this may indeed be the case. The diverse set of potential cognitive origins furthermore leads For example. Initial evidence already supports this pre. not . 2012). more be dissociated in neglect (Cappelletti et al. 2009). Likewise. attentional SNARC effects with sentences that imply concrete quan- tities.47). Finally.. compared to hor- objects or amounts of liquids). if the vertical axes where largely unrelated when it came to random we find one effect to be stronger in a particular task. Loetscher et al. Initial evidence (Sell and Kaschak. 2013).. 2011). we might expect that from cultural conventions also makes the prediction that vertical in concrete physical situations (i. participants in this result of representational space or the differential perception of study predominantly either had a horizontal or a vertical mapping. one effect be stronger than the other across the board? Given the Unfortunately.. Grade et al. we relate the multi-causal origins (Section more inclined to think about numbers vertically. This saliency of the horizontal perceptual this is not straightforward. vertical effects would be stronger izontal SNARC effects. 2012. 2013. In the next section. tion (see Section 3. where different participants 4. ical distance effect (Gertner et al.. Are some people tasks. many studies that did find vertical effects were random number gener- ation tasks (Hartmann et al.. So.. and others more 3) of space/number mappings to specific task-dependent predic. As discussed. high number. a direct comparison of horizontal and ver- vertical axis than along the horizontal axis. such as The man had two books in his bookcase—but not with numerals outside of any physical contexts.. Instead. the evidence for tal than vertical extent...g. 2013.75) while others (Bogdanova et al. Conclusions (Holmes and Lourenco. Pecher and Boot (2011) found vertical ent orientations for different cultures. In another experiment 5. ated across spatial–numerical tasks. 2009). in fact. dimension could have biased participants in favor of horizontal 2006) found a strong correlation between horizontal and sagittal number-space mappings. necessarily have a linguistic component.. outstanding. answer may turn out to be “yes” is suggested by a free placement task (Fischer and Campens. 2014)..

M. 607–618. Giraux. 149–153. A. 1988. DC.. cognitive addition: a comparison of normal and mathematically disabled Bogdanova. Psychol.. Medin. N. Umiltà. Bächtold. we might view this multitude as part and parcel of numerical associations. 36. Measuring spatial–numerical associations: evidence for a purely conceptual link. for helpful comments and suggestions. K... Stahnke. J. (Eds.. M. Designing bar graphs: orientation Acknowledgments matters. M. 2008. 2012. Process. V. Dodd. H.. 2011) and fingers (Di Luca et al.H. Ch. ing multiple ways of thinking about numbers might help us deal Fischer. S. Perform. Adil Leghari. 2005. Hutchins. 37. Physiol. abacus. Overall. line in physical space: vertical and horizontal visual number line orientation in Gertner. Psychol. 2011..C. Cohen Kadosh.. R. M. Cohen. 2006.. we have synthesized the evidence for these other. but Berch. Cipolotti. 2914–2923. 1461–1483. Anthropology in cognitive science. The evidence for vertical and sagittal SNARC Casasanto... Whereas the set of tasks that have explored horizontal effects is Goodale.L. Y. Bueti. Linguist.L. Appl. Psychol.H. Acta Psychol.. The Power of as discussed above. A. 2007.. Y. Wood. D. P.. abaci (Frank and Barner. 183–197.H. Brain Sci. Oculomotor bias induced by Appelle. Rev. R. and the body.. R. 1327–1328. 2012. R. M. S. 10.H. Exp. 2009. 181–206. Shaki. culture... V. non-horizontal mappings between space and number. Across time.. time line. 2003b. Neuropsychologia 45. Dehaene. 2010. Child Psychol. J. Psychol. Bonato. 1988. 3.M. P. D.D. Cronin-Golomb. reminds us that number knowledge is. F. Exp. which lead us to temporal gestures.D. Experiential origins of mental metaphors: language..C.. Language and Action in Cognitive Neurosciences. Cogn. Cogn. O. Bartolo.M. Di Luca. 731–735. Attentional cueing in numerical cognition. magnitudes. 67. 5. Pointing to numbers and grasping magnitudes. Only with more studies on non. B. Nature 21. Exp. D. B 12. Neuropsychologia 46. H. 643–674. M. J.. Gallivan.J. 40. 494–495. 134–149. ble ways of thinking about numbers that arise through experience. K. Nat. Mapping a field of short-time visual search. Rusconi. Psychol. B. D.D. 14 (30). make testable predictions for specific tasks environments in future Dehaene. Psychon. Psychol. Marghetis and Núnez. The mental representation of parity and experiments.. Finger – digit associations between numbers and size (Henik and Tzelgov.. Noël.. Brugger. 63–76. Frank. 2008. D. All of these can be argued to stem from expe- Finger. The middle house or the middle multiple spatial mappings (cf.W. N. highlighted that the evidence points toward the co-existence of Cappelletti. Neural reuse: a fundamental organizational principle of the Franklin. 54. Mapping spatial frames of reference onto time: a review Hartmann..H. 2012. Most importantly. Henik. Zorzi. Culham. 20. that knowledge representation is rich and flexible.... A. Cognition 132. 2013. X. Yao. Soc. 2014).. M. 2000. S... 6. Children’s Counting and Concepts of Number.. Hav. B. Exp. Exp. 23. Fischer. J. Searching imagined environments. L. Spatial associations in numerical cognition: from single digits to arithmetic.. Hill. Stimulus–response compatibility in Fuson. we have also pointed out that the relative strength Chaiken. number magnitude.... Fischer. M. Steinman. 810–818. Exp. is up”: hand and foot responses reverse the vertical association of number Bender. 51. 2009... Pratt. Y. Visualised numerals. R. D. The unique role of the visual word form area in ber lines on the three axes reviewed here. Cultural recycling of cortical maps. Vision Cogn. Psychol. Exp. 374–385.. D. Shaki. Fischer. R. orientation: the “oblique effect” in man and animals. Bender. across the body: transversal horizontal. Hum. Number magnitude orients attention.. Milne. Percept. F. R. Science 138...H...D. D. Psychol. Galton. Thus. M. Psychol. Hill. 252–256. Psychol... L.). 278–282. 13. number word processing. J. G. M. Bow-Thomas.H. Fias. Dewulf. Bossini. 2257–2273. Corbin.. Cogn. and Brugger. M. Pesenti. 953–962.. Top. W. A. each other and asking questions about which is the preferred map.. 119. The hands of time: temporal gestures in English horizontal SNARC effect. with situated and specific contexts in which we use these numbers.). Cooperrider. D. Behav.S. 2010.. In: Landau. 2013. 386–392. The spatial mapping of numbers—its origin and flexibility. Winter et al. 2011.. 372–391. pp. Mast.H. Timo Röttger and Tyler Marghetis Psychol. A hierarchical view of grounded. 342–382. Bull. Neuropsychologia 36. 249–268. C. S. M. J. Extracting parity and magnitude not against one’s will. A. C. J. Fuson. 2005.. S. 1988. our review fits a view of numerical cognition that holds 384–398. Front.L. The parietal cortex and the representation of time.A. space. 869–874. M. M.. 370–381... 2015.B. sometimes equally strong or stronger than the more established Casasanto. 1831–1840. Jasmin.. from Arabic numerals: developmental changes in number processing and Galton.. S.. 1880a.D. Conscious. Q. 91–97. 1880b. effects is strong. J. M. There is more than “more of theoretical accounts and empirical findings.. horizontal SNARC effects can we begin to address the questions Collewijn.H. 15. 404. 141. Cogn. J. K. Neargarder. 2. We have reviewed the potential sources of human vertical saccadic eye movements. 2003a. Gesture 9..H. Spelt. And despite common belief to the contrary. 260. 1381. 1998.. 1–19. 59. van der Stigchel. Finger counting habits modulate spatial–numerical ping. A. J. Baumüller. 2011. Campens. Kingstone. 1982). D. The effect of orientation on is currently somewhat limited. 1962. Fischer.L. J. Lakoff and Núñez. embodied.. Sci. 18. 2007. Brugger. Sci. 1972. J..H.. 2014. Washington. Fischer. Cognitive representation of negative numbers.D. 243–250. M. J... E. Shaki and Fischer. H.. Fischer. 1–4 [online publication ahead of References print]. Psychol. Fischer. M. Brain Res. 2. G. mental representation.. 1990. Counting on the motor system: rapid action planning vast—including all kinds of response setups and representation reveals the format.. . Chapman. 19. A. Winter floor: bisecting horizontal and vertical mental number lines in neglect. 1947. 2004. Winter et al. Foley.. and situated numerical cognition.. E. Philos. 2009. However. a Exp.. E. Rubinsten.. from number-form synesthesia.. 254–262. 2008. Psychol. Meier. Gen. A featural analysis of preschoolers’ counting Geary.. 122. C. Front. J. P. of associations depends to a large extent on the tasks considered.P..C. M. Binocular co-ordination of raised in this paper. L. 2014...C.. 2007. the presence of multiple spatial mappings cognition.. / Neuroscience and Biobehavioral Reviews 57 (2015) 209–219 217 just along the horizontal axis but also along the vertical and sagi. Psychol. associations point to finger counting as prime examples of embodied tions. R. Neurosci.. When time is space: evidence for a mental ttal ones. 2989–3000. Knops. (Amst. S.. In: cognition and mathematical practice (cf. Robinson. Visualised numerals. Vis. Cognition 108. other research points to reading. Galfano. like all other knowledge.. pp. 13.. Wasner et al.and magnitude-dependent extraction of numerical formats—the range of tasks used for vertical and sagittal effects quantity. A. J. Trends Cogn. 2006. American Psychological Association Books.... 2009. Dev. J. reflection of our experience in a three-dimensional world.. 2014. 371–396.. M.. Fischer. Perception and discrimination as a function of stimulus number perception. M. 2014. 2006. Fischer. Here. New representational space. Walsh.. Warlop.J. Counting knowledge and skill in knowledge. Cogn. rience.. et al. vertical and sagittal SNARC effects. Cortex 44. Volkmann. M. Springer. M. Representing exact number visually using mental 266–278. F. A. When 9 is not on the right: implications asymptomatic individuals with HIV. Henik.. 2012. A. Tversky. M.. Ansari. Ryan. 1984. Freeman.S... Fischer. C. Dehaene. Granà. Fischer. A.. 78. Besides num. D. Neurosci.C. Exp. the picture we are left with does not feature one Exp.. We thank Matthias Hartmann.. 1401–1414. (Eds..J. 2011. J.. dominant spatial representation but rather a multitude of flexi. Briars. 225–242. 366–374. 2014. 286–308. Castel. 2015).. Fischer. Gashaj. M. 245–266. Perceiving numbers causes spatial shifts of attention. 14. 1992.H. Biobehav. speakers.. 74.. Erkelens.K. 33. Bull. S. The illustration of the horizontal–vertical illusion. 2011).... Anderson. Fung. 192.. Kadosh. Res. Fischer... S. C. M.W. Semenza. Mapping mental number children. Psychology Press. 2014. 161–164.. Núñez. Siegler. Sci. R. 2008. F.K. London. Nature 22. 555–556. L.. 2003. Neuron 56. When digits help digits: spatial–numerical It also helps us approach the same problem from multiple direc.... these more recently discovered effects are Metaphor: Examining its Influence on Social Life. Hill. Cohen.) 124. 1993. York. Spatial representations in number processing: evidence from Rather than viewing multiple spatial mappings as competing with a pointing task.. Child Psychol. Gen. Q. Rev. M. We have number and other magnitudes.C.. P. Psychol. Barner. Beller. C. N.. 2014. E. D.H. compatibility in Arabic numeral processing... 1999. Fischer Attentional SNARC: there’s something special about numbers (let us count the ways). Trans. S. S. Umiltà. 1648–1663. Dodd.. M. Exp. brain. Seron. 493–508. Coello.

.)..J..J. Top. J.. L. D.. S. L. TX. K.. McShane. K. T. Dehaene. McShane. M. 5. J. Bello. D. Karnath. Bull. Time after time: the psychological reality of Hubbard. M.O.F. Chicago.. P. Farnè.M..W. Vision Res. Hung.. R. Youngstrom.. Tzelgov. horizontal–vertical illusion. T. A. Q.E. Gibbs. Turner. Fischer. Brugger... 2011. Kognitive Pinel.. 1992. Previc. D. Shipley. Percept.. pp. H. e64500. Vogel.218 B. Relating numeric cognition number-space synesthesia on the automaticity of numerical processing. L. Fire. A. Counting is easier Grade. Hölscher. Psychol.. 470–478. 2008. R. Nat.. pp. Bull. line: operational momentum in nonsymbolic arithmetic.H. M. Exp. D. M.. Dehaene. spatial biases in symbolic arithmetic. New York.. Guarini. Girelli. J. Functional specialization in the lower and upper visual fields in Knops. 1992. 2005.. induced by uninformative number symbols modulate neural activity in human 566–578. Goebel... Exp... Proceedings of the 33rd Annual Conference of the Cognitive Science hands? The task set influences the nature of the SNARC effect. 2009b. Cogn. Hoffmann. 161–176. Biol... Martin.. Psychol. 2012.. 21. 83.. Pesenti. J.. 1975. Kaup. J. Glenberg. W.. B. 519–542. Sci. 28. Percept. 67. 2007. P. A. 2005. Cognitive ecology. B... S. Austin. Holmes.... Dormal. 33.. M. 18. Grabherr. Heed the signs: operation signs have spatial processing. Hartmann.. Psychol. Heilman. Head turns bias the cultural and linguistic influences on the development of number processing. Lindemann. J. Cogn.. M. Levine. N.. calculation account for operational momentum in addition and subtraction? Q. 2011. Influence of fovea load on the functional visual field. . 2006.. 71. Gevers. E.. T.D. Masson.. 60. 1980.. D. Nicholls. left: effect of spatial orientation on political attitudes. Curr. 221–233. The University of Chicago Gibbs. Cognitive Science Society. 67. Violations of listing’s law after large eye and head gaze visual fields. Pierced by the number-line: integers are Guan. G. 45. Cross-Cult. Metaphor Symb. Schubiger. Fischer. Mennemeier. R.. Trail.. 1992.W.. R. M. Exp. Cogn. G. M. Exp.. G. pp. Implications of Lachmair. 2008. Is there an internal association of numbers to (Eds... Gibbs.. M. Exp. F. mental number line: simulating ‘spatial’ neglect in healthy subjects. New York. M. Louwerse. F. Hartmann.. S. J. 18. H.... Zimmerhackl. 36th Annual Knops. Percept. Hatta. W.. (Amst.. 479–487.. Brain Res. Where Mathematics Comes From: How the Embodied 434–458. Looking for the answer: the mind’s Göbel.P. and semantic size in comparison tasks. Spatial structure of quantitative representation of numbers: hands: a spatial representation effect in digits handwriting. (Ed. F. Psychol. Yao. Calabria. R. S. movements in a natural task. M. Berlin. Núñez. F. Schwarz. movements during exact arithmetic reveal systematic.. A. O. R.... Wiley. Müller. Operational momentum affects Pitt. Psychol. R. Can approximate mental Conference of the Cognitive Science Society. Hornung. 761–771.W. Y.. occipital cortex.H.. Zorzi. Activity. 1–20. M. Brugger. 1514–1526. Mem. 2011. E. 139.). 42. University of Chicago Press. J. Tanne. size.. G. 376–385. Speed of adding and comparing numbers. Attentional bias induced by solving simple and Henik. W.. Rev.. T. A. 205. performance of speeded binary classification tasks. number generation. I. 1982.M.... D. de Hevia. McAnany. Q. 1949.. Eliuk. 2011. Psychophys... Pinel. Cross-Cult.. Oppenheimer. Moving along the mental time line influences the Marghetis..) 148. Loetscher. 1151–1161. Top.W....-O.. Y. 2. I. 2012..W. symbolic and nonsymbolic calculation: evidence from the operational Proctor. abstract situations. what number you have in mind. Horizontal trumps vertical in the spatial 37–50. 132.. 2008b. 107–114. M. Psychol. Why leaning to the left makes you lean to the Hutchins. Conscious. Psychophys. Oxford. M. Cho. Moeller.. 3419–3428. Thompson. Psychol. (Eds. and simulation: Lechelt. 860–868. Pelz. B. 2697–2704. M. T.H. Psychophys. E. Baroni. C. Oxford.. In: Bello. Neurophysiol.. R60–R62. Neglect. J. Cognitive Science Society.. Loetscher. New York.. R. W. K. 2013. Scassellati. Cortex and language processing: do numbers and words share a common 49.. 2001. The motor system contributes associated with back-to-front sagittal space. G. Child Neuropsychol. Holmes.W. T..-O. C. to comprehension of abstract language. Q. The motion behind the symbols: a vital role for line: interactions between whole-body motion and numerical cognition. Chicago. R. O. E.E.. S. mathematics.B. and the moon illusion. New York. Nicoletti. A.and time-reference-point distinction in metaphorical construals of number and space in parietal cortex. Wertman..A.. Society. 67.. Thirion.. J. LEAC.. Furlong. G. 2011. M. performance of preschoolers. Bergen. T. Q. 2006. Parietal rTMS distorts the eye in number space. Johnson. Wood. Núnez. 6 (6). 1044–1051. 21. Doing arithmetic by hand: hand processing of future related words. D. K. Moving along the mental number Marghetis.. Matlock. Rossetti. 2010. Acta Psychol. Dehaene. 1993... A. 2000. K. 2014.. 463–469. In: Gibbs. de la Vega.. Schocke. Q... C.M. Oxford University Press. Hubbard. Hutchinson. Schiltz. time. G. 1441–1450. S... Bockisch. Piazza.. Neuroscience 151. M. Lugli.M. Metaphor: A Practical Introduction. The cultural number line: a review of Loetscher. Cognition 109. K. 2006. E. M. 1416–1427. Psychol..H. Moving along the number Percept.... J.. J. Metaphor Symb. Gibbs. Neglect of near peripersonal Exp. 13. Srinivasan. A developmental fMRI study of nonsymbolic numerical and Pinhas. Nuerk... S. imagination. Holmes. S.. G. Z. U.Q. Women.. Austin. The coordination of eye. 2013.. Kremser.. (Eds.E. G. Cortex 44. J. Recruitment of humans: its ecological origins and neurophysiological implications. S. Biol. M.. Y. 1980. R. 775–791. Figurative Language. Interactions between the ego. Shaki. Metaphor. 21 (2). B. 1970. Mem. dynamism in the conceptualization of limits and continuity in expert Psychol. Psychophys.. V.. 1987. T. A. C.. Behav. 2010. 1174–1179. 38. 1324–1333. Verguts. J. 133–146. E. 32. Science 324. Front. Reznik. P.J. Cohen Kadosh. Numbers in space: differences between concrete and spatial–numerical association of response codes. TX. Bockisch.M. 2820–2830. 408–415. Huber. C.. Lefèvre. Mental movements without magnitude? A study of Kaufmann. 274–278... Basic books. Loeber. L.H. M. Brain Res. Neurosci.). Orienting numbers in mental space: horizontal Noël. Alipour. In: Carlson.H. 42. 413–430.. Viarouge.N. 2014. Núñez. Experiential origins of the mental number line. C.. Exp.. Koten. Numbers can move our Ito. Influence of gaze observation on random while experiencing a congruent motion. 2008. 2014.. 68.. 416–442. Dehaene.L. 2013. Le Bihan. T. Automatic Lakoff..W. 35. 705–715... Fias. 2004. Wiley. 67. 2010. Dev. Exp. The Organization of Behavior: A Neuropsychological Theory. Anelli.W. I.. 1614–1625. Dehaene. 2007. F.. N.. The relative capabilities of the upper and lower Glenn. T.. 2005. Proceedings of the 36th Annual Conference of the Haith. Neuropsychologia 50.. 2013. Cristina. S. Psychol. Psychon.. Fischer. shifts.. cerebral representations of number. 2010. M. and luminance during comparative Katz. 51. Dehaene-Lambertz.E.. Bricolo.. Percept. Exp. In: Karnath. T.M. Previc. Vilis.S. R. Teuscher. Exp. 651–661. 2276–2281. Boot. PLOS ONE 8. Potomac. Psychophys. University Press.. spatial associations. A. Lourenco.. 2014. Psychol. 260–271. Metaphors We Live By. processing... 116. R. M. Fischer. 13. 1994..M. Finger counting habits in middle Goffaux.. Cogn. D. Restle. New York. P.C.J. 803–821. PLOS ONE 8. Anisotropic perception of visual angle: implications for the influences our understanding of time. 20. 11.. J. brain’s internal random generator. E.. 2010. Austin. J... A.. 2014. I.. overconstancy of size. 2. 279–291.. M. U. C. S. 2013. an area involved in eye movements during mental arithmetic.. Flexible spatial mapping of spatial–numeric associations: evidence from spatial and quantitative different notations of numbers in Chinese readers. Mind Lang. J. McCrink. 725–729... Cogn. Cambridge University Press. organization of numerical magnitude.. Lourenco. T.L. 2012. 2006. Ramscar. and Dangerous Things: What Categories Reveal response activation of implicit spatial information: evidence from the SNARC About the Mind.. T. M. Oxford judgements. Exp. B. M.. S. / Neuroscience and Biobehavioral Reviews 57 (2015) 209–219 Gertner.. Cambridge Handbook of comparison of visual and haptic space. 662–673.H. e75183. effect.F.. The Poetics of Mind. S. 946–951. M. 389–395. Cogn... Perform. R. Marghetis.) 122. Language statistics explain the Pecher. Mast. 26. Núñez. W. Finger gnosia: a predictor of numerical abilities in children? organization trumps vertical.J. 2002. Eye position predicts Neuropsychologia 44. 1583–1585. 2014. Michel. R.M. Sci.. H. Vision Res. Early development of Hung. Metaphor interpretation as embodied simulation. C. pp.. C.... 2014. P. 2012. 1990..-C. 65.. 2014. 1998. 1976. 543–565. 1541–1556. L. Metaphor and Thought. Psychol. Howard. 2008a. Schiltz..L. Brain Sci. Klein.E. Opfer. space evidence for multidirectional attentional systems in humans. 1558–1562. E. 435–448.H. L.. B. Pesenti. Exp.. Lammertyn. Sci. Neuron 41.. representational platform? Acta Psychol. P.. D. K.-P. 2006.. Cambridge University Press... Atten. J.R. C. Soc.. evidence from the SNARC effect. Matlock. L. Is three greater than five: the relation between physical complex addition and subtraction problems. Perceptual orientational asymmetries: a psycholinguistic evidence. Blume. Curr. S. 299–316. Kövecses. B. A.. M. F. Tiher. Brain 115. J. Dehaene. A. Rules that Babies Look By: The Organization of Newborn Visual Cognitive Science Society. 2009a.). Rev. S. Developing number–space 69. M. Mind Brings Mathematics into Being. 2004.. pp. dynamic spatial Hebb. S. Dynamic representations underlying space. D. Martin. Meng.. Perrone.. Springer Verlag... associations: SNARC effects using a color discrimination task in 5-year-olds. Brugger. Y. Dudschig. C. Borghi. 2013. 2012. 122–130. V. 1527–1540. 255–260. Schwarz.. Winter et al. Cogn.. Mast. Wu. Lakoff. 309–318.. 121. 20. 1579–1596. Cognition 106. Shaki. Göbel. R..J.M. R.. Guarini... M. T. Child Psychol. P. Distributed and overlapping Neurowissenschaften. J. Berteletti.. Lakoff. and hand Ikeda. Visual search asymmetries in three-dimensional Knops. R264–R265. head. 2014. Human Visual Orientation. 67. (Eds.. (Amst. Mem. Hum. Percept. 21. 10. 266–277. Scassellati. M. B. R.E. Casasanto. 2008. M. M. Attentional shifts eastern and western individuals: an online survey. Cognitive Science Society. Press. 1982.). P. Motz.. Takeuchi. Piazza. TX. Polarity correspondence: a general principle for momentum effect. Psychol.M. M.. 218–230. M. Tzeng. Notebaert. Even abstract motion Higashiyama. Henik. Hayhoe. In: eye fixation behaviour.. 1352–1362. Pinhas.

. On the cognitive link Shaki. Psychon. Sci.. 5.. S. Q. W.... numbers contribute? Perception 41.. Hervás. W. Cognition 44. 561–569. Oxford.. Spatial associations in number-related tasks. 1991. 512–516. Schwarz. Wood. J. C... Perlman.). 232. Petrusic. Winter.... Random walks on the mental number line.J. [online publication ahead of print]. R. 67. X. Göbel. D. 2306–2309.. 2014.. Fischer. Ulrich. 2011. S. K. R. context-dependent SNARC effects with comparative judgments.. Q. Shaki.. Tversky. S. Bowers. S.. pp. M..A. The number space and neglect. T. Funes. J.. M.. Cortex Santens. Fischer. Brugger. Using space to talk and gesture about Seron. 1061–1067. 2006. Winter et al. Bull. Fias.P.. 994–1003. Shaki.. P. K. You. 275–281. K. pp.. A theory of magnitude: common cortical metrics of time. 2015. . 38. M. M.. Rev..W.. Matlock. Psychol. 192. 19.. 2009. Percept. C. numbers: evidence from the TV news archive. representation of numerical and non-numerical sequences: evidence from Spivey. 2008. M. Rev. mental number line and Winter. Brain between space and number: a meta-analysis of the SNARC effect. Fischer. cognition of time. F. Meneghello.. X. trends in graphic productions. 2015. and Shaki. Of magnitudes and metaphors: personality. S. Wasner...). 2008. S. S.... Walsh. B. 271–295. Proceedings of the 35th Annual Conference of the Cognitive Science Society. K. H. 2014. 931–937. J. 2012.. J. Keus. Weger. 2011.. Vázquez. Q. Seno.P. Walsh. Exp. Psychon. Multiple spatial mappings in numerical cognition. Fischer. K. Peripersonal and vertical neglect. 328–331.. Brain damage: neglect disrupts the mental Schemas and Abstract Thought. Kugelmass. Extending the mental number line—how do negative Brain 113. 2005. Lupáñez. A. Spatial schemas in depictions. MIT Press. Semenza. comparing SNARC effects with saccadic and manual responses.H. K. Psychol. Heilman. Moving the eyes along the mental number line: Oxford University Press.. influences memory retrieval: the case of finger counting habits. S. Cortex 64. time. The Continuity of Mind. M. (Ed. Number-induced shifts in spatial attention: a a comparative study of spatially directional counting biases in cultures with replication study. and illiterate Arabic speakers. N.C. Cooperrider.. 2014. T. 2013. Cross-cultural and developmental effects as a function of task demands.. The origin of number-space associations: Zanolie. Werner. 165–190. 14.P. W. Kadosh. M. 377–408. Lindemann.. M. Cortex 47.... Psychon. The SNARC effect does not imply a mental number 40. 2004..P. Q. W.. Cognitive Science Society. 2012. 263–270. Numeric comparison in a visually-guided manual Zorzi. Bull. Gevers. Rahona. 138–139. 53. 515–557. Ihaya. 7. Moeller. 2007. Sturt. C. Song.. Grammatical number elicits SNARC and MARC Tversky. Wachsmuth. A working Salillas. V. or “When 98 is upper left and 6 sky blue”.. Cogn Sci. Abrahamse. Exp.H.. Psychol. Umiltà. A. 651–664. 1992. Psychol. line.. 489–525. 2003. neglect. H. 2001. Marghetis. C. B.. 2009. domains: from arithmetic to language and back. Domahs. Exp. Oxford University Press.. Tversky. F. M. Cult. Exp.A. J.. Exp. Psychol. 2011.H.. 1323–1335. Spatial Zorzi. 12.. Kaschak. Oxford. Marenzi.. 4–15. 43–49. 2015. perception) alters cognitive states. Res. Exp. k. Brain Lang. 16. Matlock. (Eds.. Moving your eyes to solution: effects of movement on Sell. Bekkering. Cooper. T. free-choice task..H.. Cogn.J. 2015... Austin. R. Bull. J. Cognition 57. C. P. 1231–1248. Zhang... M. . The spatial reaching task.. D. 209–224. 2014. Zebian. 987. 406–413.L. space Scheepers. Reading habits for both words and directionality of writing: the SNARC effect and the reverse SNARC effect in numbers contribute to the SNARC effect. T. T. S. J. 2004....... Yamada. Reading space into numbers: a cross-linguistic along two axes. M.. (Eds. 1990. M. Pecher.. 3789–3974.. J. 66. Time flies like an arrow: space–time compatibility Sell. The spatial representation of numbers: magnitude determines gaze direction: spatial–numerical association in a evidence from neglect and pseudoneglect. The Oxford Handbook of Numerical Cognition. S. 1557–1570. 112. 67. Psychol. E. 552–565. 2006. Pratt. right. 117. pp. 67. B.. The comprehension of sentences involving quantity the perception of a problem-solving task. A. Q.... D. G. Time (also) flies from left to temporal gestures. Y. 2014. Front.. I. J. Q. Perform.H.. Sunaga. English and Arabic monoliterates. Bull.). Priftis. 499–535. K. Cognitive Science Society... Matlock. Neuropsychologia 44. Sensory and cognitive processes of memory account of the interaction between numbers and spatial attention. Petrusic. B.. Exp. Acar. Cogn. 1995. Rev.. number line. 2008.. 317–328.. responses. S.. M. pp. Psychon. .. Cambridge. Process. The continuity of metaphor: evidence from Santiago. Winter. Raab.. Kaschak..J. 39–44. 2012. Rev. In: Gattis. 2014. R. Vuilleumier. Trends Cogn. 590–599.M. 159–196... Deloche. Ruiz Fernández. 617–620. 2012.. M.. T. TX. J. I. Cognition 108.H. K. Peebles. J. B.. Shelton. Processing time shifts affects the execution of motor effects suggest the use of a mental timeline. Y.. A representational analysis of numeration systems. Umiltà. 2007. O. Proceedings of the explaining cognitive interactions between space.. Fischer.M. A. comparison of the SNARC effect.. Nature 417. Cortex 44. shifts of spatial attention induced by numbers: an ERP study. J. M. K.. Nuerk.A... 2008.. Zorzi. J. 23.. 2008. M. arithmetic: evidence from the motion-arithmetic compatibility effect. C. D. TX.J. Cognitive Science Society. 79–111. W. F. Zhang. B.. G. 804–809. Exp.. E. Ito. Müller. M. Pauen. Noël.. Spatial interferences in mental 708–714.. 15. Gesture 13. S. M. van Dijck. B. A theory of magnitude: the parts that sum to number.. Sci. Norman.. More is up. S. M. 1500–1513. Cornet.. 2014. Hum. Fischer. Number Umiltà. (Eds. Taya. Linkages between number concepts.. N. 2005.H. Psychol 5.... In: Cohen 1643–1654.. J. Schwarz. Bull. Psychol. G. Austin. M. R. Sci.. Top. 68. How situated cognition Psychophys. 2008. 67. and right: random number generation Shaki.M. In: Miyake. E. / Neuroscience and Biobehavioral Reviews 57 (2015) 209–219 219 Roettger. K. Priftis. M. S. information affects responses on the up-down axis. Visualizing thought. Cogn. spatial thinking. 2012.M. Willmes.. 3. 191–205. 50. Bi-directional syntactic priming across cognitive and quantity. Psychon. Walker. Nakayama. V. 15. Vection (self-motion Psychol. El Yagoubi. B... Brain Res. In: Knauff. Exp. biliterates. Sebanz. W. Psychol. P. Pesenti.. Winter.. Shaki. J.). Priftis. Ortigue. J. On the mental representation of negative numbers: Cogn. Pérez. and number. Images of numbers. M. Dowker. Cognition 108. Rev. H. Child Psychol. different reading directions. Percept. M.M. 399–410. Cognition 106. 483–488. Fischer. Exp. Wiemers.H. 2014. D. 426–430. 1571–1578. Nuerk. 2002. U.C. H..P. Ketels. P. E.