Abstract
Functional lateralization is established for various cognitive functions, but was hardly ever investigated for arithmetic processing. Most neurocognitive models assume a central role of the bilateral intraparietal sulcus (IPS) in arithmetic processing and there is some evidence for more pronounced lefthemispheric activation for symbolic arithmetic. However, evidence was mainly obtained by studies in righthanders. Therefore, we conducted a functional nearinfrared spectroscopy (fNIRS) study, in which IPS activation of lefthanded adults was compared to righthanded adults in a symbolic approximate calculation task. The results showed that lefthanders had a stronger functional rightlateralization in the IPS than righthanders. This finding has important consequences, as the bilateral IPS activation pattern for arithmetic processing seems to be shaped by functional lateralization and thus differs between left and righthanders. We propose three possible accounts for the observed functional lateralization of arithmetic processing.
Introduction
One important indicator of functional lateralization in the brain is handedness. Handedness describes the preferential use of one of the hands with high accuracy and motor speed while performing skilled (i.e., culturally influenced activities such as handwriting) and unskilled actions based on spontaneous activities (e.g., picking up items)^{1}. Both right and lefthanders represent the normal range of human diversity, but they differ in processing of various types of information and in the functional lateralization of the brain^{2}.
The role of handedness as a behavioural reflection of functional lateralization was already investigated in neurocognitive domains such as language processing^{3} and spatial processing^{4}. Language processing is leftlateralized in about 95% of righthanders, while lefthanders either also show a leftlateralized (75%) or a rightlateralized or bilateral representation of language^{5,6,7}. On the other hand, spatial processing is generally rightlateralized in righthanders and bilaterally or leftlateralized represented in lefthanders^{4,8}. Thus, functional lateralization is already established in other domains such as language and spatial processing, but was rarely studied in the domain of arithmetic processing so far.
The Neural Representation of Arithmetic Processing
Brain activation associated with arithmetic processing relies on a frontoparietal brain network^{9}. Thereby, the core of number magnitude processing is considered to be the bilateral intraparietal sulcus (IPS) located in the parietal cortex between the superior parietal lobule (SPL) and the inferior parietal lobule (IPL) ^{for dominant models and their extension see}^{10,11,12}^{; for important data see also}^{13,14}. However, while the bilateral IPS is associated with arithmetic processing^{15}, some studies reveal a more pronounced left hemispheric activation^{16,17,18,19,20,21}. Brain stimulation studies confirmed the functional role of the bilateral IPS in arithmetic processing and more basic number processing^{22,23,24} with a larger involvement of the left IPS^{22,25,26} compared to the right IPS^{27}.
This more pronounced left than righthemispheric IPS activation especially holds for symbolic arithmetic in Arabic notation compared to nonsymbolic arithmetic^{28}, which is supported by the basic differentiation of symbolic and nonsymbolic number processing across modalities^{29}. Different involvement of the left and right IPS were further found for approximate compared to exact calculation, while approximate calculation was found to be associated with stronger bilateral IPS activation than exact calculation^{28,30,31}.
These findings lead to the question whether there might be a functional lateralization of arithmetic processing within the IPS. This means that the bilateral or predominantly leftlateralized IPS involvement in arithmetic processing might be true for righthanded individuals, who form about 90% of the human population^{32}, but not for the human population as such. In particular, this might not hold for lefthanded individuals, who are usually discarded from neuroscientific studies as nuisance factor and might show an opposite lateralization. Therefore, the current study addresses the issue, whether lefthanders differ from righthanders in the lateralization of IPS activation during symbolic approximate calculation.
The Role of Handedness in Arithmetic Processing
Behaviourally, there is no clear evidence for general performance differences between marked righthanders and lefthanders in math^{33}, although both right and lefthanders usually perform better than for example mixedhanders^{34,35,36}. However, somewhat contrary to the above findings, professional mathematicians show on average a lower degree of handedness^{37,38}. Besides, left and righthanders were shown to differ in some basic numerical effects, as for example in the markedness effect^{39,40} and some spatialnumerical associations^{41}^{; but see}^{39,40}, but not in the distance effect or compatibility effect^{42}. In sum, there is no consistent empirical support for differences between righthanders and lefthanders in math performance, because some findings also seem to depend on age, sex and task^{34}.
Neurally, effects of handedness have not been systematically examined in the field of numerical cognition so far. Some neuroimaging studies, however, provide hints towards a difference in neural activation between left and righthanders, although only very small samples of lefthanders were analyzed (n = 8^{16}, n = 7^{43}, n = 3^{44}). According to this weak evidence, in the prefrontal cortex righthanders show a stronger leftlateralization, while lefthanders might show a bilateral or rightlateralized activation pattern^{16,43,44}. In the parietal cortex, lateralization aspects are even less clear, because a leftlateralization for arithmetic processing might hold for both right and lefthanders^{43} or it might be just stronger in righthanders^{44}. Moreover, lateralization seems to decrease with increasing arithmetic complexity^{43}^{; see also}^{45}. Because of the essential role of the IPS in arithmetic processing, the relation of handedness to the functional lateralization of arithmetic processing needs to be resolved in this brain region.
Objective
Functional lateralization of the brain was previously shown for other cognitive processes; however, it was not systematically investigated for arithmetic processing. Although arithmetic processing is considered to be bilaterally represented in the IPS, there are some indications of different functional roles of the two hemispheres. Therefore, the aim of the current study is to test functional lateralization of arithmetic processing in the IPS. As a wellestablished indicator of functional lateralization, we focus here on the relation between handedness and arithmetic processing, since handedness was found to be associated with some numerical effects^{39,40}. Hence, we will investigate whether lefthanders reveal a difference from righthanders in activation within the IPS during a symbolic approximate calculation task. As there are mostly no clear performance differences between left and righthanders on the behavioural level, the current study addresses the issue whether functional lateralization can be detected on the neural level. In righthanders, symbolic arithmetic was found to be associated with activation of the left IPS and approximate calculation with bilateral IPS activation. The question of the current exploratory study is whether neuroimaging findings from righthanders can be generalized to human arithmetic processing in general or, as we expect, whether lefthanders show less pronounced activation of the left IPS but instead more activation of the right IPS compared to righthanders because of differences in functional lateralization.
Methods
Participants
Seventy adults participated in the study, who were native German speakers with no history of neurological or mental disorders. Participants were excluded from analysis if they had an overall error rate exceeding 25% (n = 3) or because of missing neural data (n = 2) or noisy data (n = 10). Handedness was evaluated by the EdinburghHandedness Inventory, where the laterality index of handedness (LI_{handedness}) was calculated according to the formula \((RL)\div(R+L)\times 100\)^{46}. Individuals are considered as marked lefthanders in the range of −100 to −40, as marked righthanders in the range of + 40 to + 100, and as mixedhanders in the range of −40 to +40. The resulting sample included 23 marked lefthanders, 23 marked righthanders, and 9 mixedhanders (cf. Table 1). The handedness groups did not differ significantly in age, gender, or final math grade at school (cf. Table 1). Each participant gave written informed consent and received monetary compensation or student credits. The study was approved by the Ethics Committee of the Medical Faculty of the University of Tuebingen and conducted according to the ethical guidelines and principles of the international Declaration of Helsinki.
Material
The approximate calculation task consisted of twodigit addition and subtraction problems. In a choice reaction paradigm each arithmetic problem was presented simultaneously with two solution probes^{47}, whereby none of them represented the correct result of the arithmetic problem (cf. Fig. 1). This kind of task triggers approximate rather than exact calculation and relies on number magnitude processing^{30,47}. The task was to choose the solution probe that was closer to the correct answer (target) by pressing the left or right Ctrl key on a standard computer keyboard with the left or right index finger, respectively. This procedure assured that motor activity should not differ between left and righthanders. The target had a distance of ±1–3 to the correct result within the same decade and the distractor distance was either small (±4–8) or large (±14–18), whereby the direction of the distance from target and distractor to the correct result was the same.
For each combination of addition/subtraction with small/large distractor distance a stimulus set consisting of 25 arithmetic problems was created. Full decades (e.g., 30), ties within and between all numbers of an arithmetic problem (e.g., 55 + 16 or 25 + 45) were not included in the stimulus set. Addition problems with small and large distractor distance were matched for various stimulus properties^{47}: the numerical size and parity of the operands, target, distractor and correct result; overall problem size; relative and absolute distances between target, distractor and correct result; need of a carry operation (or borrow operation in case of subtraction); decade crossing between target and distractor; the positions of the smaller operand and of the target. Subtraction problems were constructed as the inverse problems of addition (e.g., 46 + 38→84 − 38).
Procedure
This study was part of a larger project; we focus here only on the data of the approximate calculation task. During the functional nearinfrared spectroscopy (fNIRS) measurement, each participant was sitting in front of a computer in a dimly lit room. After fNIRS preparation and receiving instructions, the participant performed computerized tasks including the approximate calculation task during the fNIRS measurement.
In the approximate calculation task, all problems and solutions were presented in white on a black screen using Presentation software (Neurobehavioral Systems, Inc., Berkeley, CA, USA). The problems were embedded in a block design with a block length of 35 s and an interblock interval of 20 s when the screen remained black (cf. Fig. 1). Blocks for each combination of addition/subtraction with small/large distractor distance were presented in randomized order in each of 5 runs (20 blocks in total). Each block started with 5 critical trials chosen from the respective stimulus set and was added up with additional filler trials chosen from an additional matched stimulus set. Trial order within each stimulus set was randomized for each participant. Each trial was terminated by button press or when the time limit of 6.5 s was reached. Termination was followed by an intertrial interval of 0.5 s. Participants were encouraged to solve math problems as quickly and accurately as possible. Participants did not receive feedback as to the correctness of their response. Prior to the experimental trials, the participants solved 6 practice trials in order to become familiar with the task. The duration of the task was 20 min.
fNIRS data acquisition
fNIRS recordings during the approximate calculation task were performed using the mobile nearinfrared spectroscopy device NIRSport Model 88 (NIRx Medical Technologies, LLC, NY, the USA). Eight sources and eight detectors were mounted in an fNIRS cap according to the 10/20 system covering the parietal lobes of both hemispheres (9 channels per hemisphere) with centres in CP3 – P3 and CP4 – P4, respectively (cf. Figure S1 in the Supplementary Material). Two nearinfrared laser beams with wavelengths of 760 and 850 nm were emitted. The sampling rate was 7.8125 Hz.
Data analysis
All statistical analyses were conducted by using SPSS (IBM SPSS Statistics, version 25; IBM Corp., Chicago, IL, USA) and effect sizes were calculated according to Lakens^{48}. For behavioural data analysis, only critical trials (and not filler trials) were entered into the analyses. Response time (RT) was regarded as the time interval from problem presentation on the screen to participants’ pressing one of two possible response keys. Only correct trials were included in RT analysis (exclusion of 11.56%), RTs beyond 3 SD of the participant’s M were repeatedly excluded (exclusion of 0.79%), and finally mean RT was calculated for each participant. The error rate (ER) was regarded as the proportion of incorrect or timeout responses to the total number of trials included in the analysis. RT and ER were compared between lefthanders and righthanders by ttests for independent samples. Note that the behavioural data of one participant could not be analyzed because of wrong button use (neural data of this participant was nevertheless included).
The relative concentration changes of oxygenated (O_{2}Hb) and deoxygenated haemoglobin (HHb) were extracted from the fNIRS signal for each channel. fNIRS data preprocessing was performed by using custom MATLAB (The MathWorks, Inc., USA, version R2013a) scripts. Data preprocessing included interpolating noisy fNIRS channels by neighbouring channels (9.29%), excluding blocks with uncorrectable artefacts (3.09%), and bandpass filtering of 0.010.2 Hz. The signal was further corrected by correlationbased signal improvement (CBSI) according to the assumption that cortical activation is reflected by simultaneous increases in O_{2}Hb and decreases in HHb^{49}. Afterwards, the amplitudes of all blocks of 35 s were corrected to the baseline of 5 s before each block and averaged across all blocks resulting in the mean amplitude for each channel and participant.
As the region of interest (ROI), we focused only on the bilateral IPS being located between the SPL and IPL, i.e., channel L8 corresponding to the left IPS and channel R17 corresponding to the right IPS (for the location of the channels see Figure S1 and for the results of all channels see Figure S2 in the Supplementary Material). The positions of the channels are labelled by the corresponding brain region according to the automated anatomic labelling (AAL) atlas^{50} based on virtualheadsurface landmark measurements^{51}. In the first analysis, a 2 × 2 ANOVA with the betweensubject factor handedness (left vs. right) and the withinsubject factor hemisphere (left vs. right) was conducted on the mean amplitudes.
In the second analysis, the laterality index of functional brain activation (LI_{brain}), which should not be confused with the laterality index of the handedness questionnaire (LI_{handedness}), was calculated for each participant according to the formula \((RL)\div(abs(R)+abs(L))\times 100\), whereby L and R depict the mean amplitude within one channel (IPS) on the left (L8) and right (R17) hemisphere, respectively^{52,53}. Thereby, negative values for LI_{brain} indicate a lateralization towards the left hemisphere and positive values indicate a functional lateralization towards the right hemisphere. LI_{brain} was compared between lefthanders and righthanders by a ttest for independent samples. In the third analysis, in order to evaluate the relation between brain lateralization and the degree of handedness, LI_{brain} was correlated with LI_{handedness} in the whole sample, i.e., including left, right and mixedhanders. Note that while in the first two analyses only left and righthanders were compared categorically, the third analysis, which used handedness as a continuous variable, could be conducted on all participants including mixedhanders.
Results
Behavioural results
Lefthanders (M = 3110 ms, SD = 614) and righthanders (M = 2896 ms, SD = 629) did not differ significantly in RT [t(43) = 1.16, p = 0.253, d = 0.35]. Regarding ER, compared to lefthanders (M = 0.10, SD = 0.04), righthanders (M = 0.13, SD = 0.06) made significantly more errors [t(36.72) = 2.41, p = 0.021, d = 0.74]. Additionally, a correlation analysis was conducted to check whether this difference in ER was associated with lateralization in the brain. The results yielded no significant correlation between ER and LI_{brain} of left and righthanders [r(43) = −0.15, p = 0.326], suggesting that the difference for LI_{brain} reported below does not seem to be influenced by the difference in performance between the groups.
fNIRS results
The ANOVA for mean amplitudes revealed a significant interaction between handedness and hemisphere [F(1,44) = 4.06, p = 0.050, \({\eta }_{p}^{2}\) = 0.08], indicating that in lefthanders activation was higher in the right IPS and in righthanders activation was higher in the left IPS (cf. Fig. 2A). The main effects of handedness [F(1,44) = 0.05, p = 0.830, \({\eta }_{p}^{2}\) < 0.01] and hemisphere [F(1,44) = 0.72, p = 0.401, \({\eta }_{p}^{2}\) = 0.02] were not significant.
The LI_{brain} for IPS activation differed significantly between lefthanders and righthanders [t(44) = 2.21, p = 0.033, d = 0.67] (cf. Fig. 2B). This indicates that in comparison to righthanders [t(22) = −0.71, p = 0.485, d = 0.15], lefthanders show a significant rightlateralization in the IPS [t(22) = 2.27, p = 0.033, d = 0.47], as revealed by posthoc onesample ttests.
To further investigate the relation between handedness and lateralization in the IPS, a correlation analysis was conducted between LI_{brain} and the degree of handedness in terms of LI_{handedness} in the whole sample (including mixedhanders, because handedness was used as a continuous variable). A significant negative correlation was observed between LI_{brain} and LI_{handedness} [r(53) = −0.27, p = 0.044], indicating that a higher degree of lefthandedness corresponds to a higher rightlateralization of IPS activation (cf. Fig. 2C), thus corroborating the results of the categorical analysis.
Discussion
This study set out to test functional lateralization of arithmetic processing in the parietal cortex. In the domain of numerical cognition, weak evidence was found for functional lateralization of basic number processing^{54} and our finding adds to this by showing functional lateralization of arithmetic processing. In this way, the functional lateralization findings of these two studies seem to converge, even though different aspects of numerical processing (basic processing vs. arithmetic) were examined. This finding in the field of numerical cognition is not entirely surprising since functional lateralization was detected for other domains, such as motor function in terms of handedness, language processing^{3} and spatial processing^{4}.
Most neurocognitive models of arithmetic processing assume a pivotal functional role of the bilateral IPS. However, virtually all these models and their supporting data have been based on studies in which predominantly righthanders were tested. Henceforth, the current study aimed at testing the functional lateralization of arithmetic processing by comparing the lateralization of the IPS during symbolic approximate calculation between left and righthanders. In line with our hypothesis, we found a stronger rightlateralization in the IPS in lefthanders than in righthanders. This supports the view that lefthanders differ from righthanders in the lateralization of arithmetic processing in the IPS, so that there is an association between handedness and the neural representation of arithmetic. This result suggests that previous findings derived from righthanders cannot be readily generalized to the human population as such, and, in particular, not to lefthanders. Consequently, existing models of arithmetic processing need to be extended to account for functional lateralization associated with handedness. Our data seem to indicate that the bilateral IPS activation for arithmetic processing seems to be shaped by functional lateralization: the importance and the roles of the left and the right IPS for arithmetic seem to be different in righthanders and lefthanders.
Thereby, we need to acknowledge that the power of the main finding in our study was not so high (power of 0.60). Nevertheless, when integrating the weak evidence of functional lateralization of basic number processing in general^{54} together with the current finding of functional lateralization of arithmetic processing, there is cumulative evidence for the relation of handedness and lateralization in the IPS in the domain of numerical cognition, which would be a very important finding for neurocognitive models of numerical processing and beyond. However, we also wish to make clear that future research on lateralization in numerical cognition needs to substantiate these two findings by using better powered study designs, because the observed effects were not large.
The crucial question, however, is why arithmetic processing might show such a functional lateralization within the IPS. We offer different accounts of this issue, which might guide future research on the understudied topic of functional lateralization in numerical and arithmetic neurocognition.
The embodiment account of functional lateralization in number processing
In the last years, a growing body of literature has suggested that even basic numerical cognition is embodied^{55,56,57}^{, for embodied trainings see}^{58}. In particular, the use of hands and fingers has been postulated to influence numerical cognition, such as spatialnumerical associations^{41,59,60,61}, magnitude comparison^{55}, or also mental arithmetic^{62}. Importantly, embodied cognition has even been suggested to influence the neural representation of numbers and operations on numbers^{63,64}. For example, counting, as one of the most basic numerical skills, is associated with the excitability of motor circuits for hands^{65}.
Against this background, it is conceivable that functional lateralization of arithmetic processing might be explained by the influence of the dominant hand used during the acquisition of symbolic arithmetic, which might lead to a contralateral hemispheric dominance to the preferentially used hand see also^{41}. The underlying mechanism might be a colateralization of the motor activities, such as handwriting or finger counting, preferably conducted with the dominant hand and therefore be related to the developing cognitive skills in terms of symbolic arithmetic^{66}. If the basic numerical representations on a neural level are indeed influenced by the preferentially used hand (as suggested for finger counting^{64}), then a differential functional lateralization for lefthanders and righthanders is not surprising. In particular, this account would explain the relatively stronger activation of the right IPS in lefthanders (compared to righthanders), because the right hemisphere is contralateral to the dominant hand in lefthanders.
Supporting the embodiment account of functional lateralization, the degree of handedness was correlated with functional lateralization of the IPS for arithmetic processing. Going beyond the group differentiation of left and righthanders, the degree of handedness reflects an important factor that can influence cognitive abilities^{3,8}. Here, with increasing degree of lefthandedness, a stronger rightlateralization of the IPS was observed, which is in line with the embodiment account of functional lateralization.
Colateralization of different neurocognitive functions: Associations of number with language and space
On the one hand, numbers are closely related to different dimensions of space like their spatial direction or spatial extension^{67,68,69}, for a review and special issue on the topic see^{70}. For instance, larger numbers are associated with the right side of space and smaller numbers with the left side of space in Western societies^{71}. Since human infants^{72} and even newborns^{73} as well as other species^{74,75} were found to express spacenumber associations, the close relation between numbers and space might be even innate.
On the other hand, numbers are also related to different types of linguistic attributes ^{for a review and special issue on this topic see}^{76}. For instance, the grammatical language structure in number words seems to determine how easy or difficult it is to acquire numbers early in development^{77}. Even in adulthood, language attributes like reading direction^{42,78} or the lexical composition of number words influence numerical cognition^{76,79}. In sum, number processing has multiple relations to spatial and language processing on a behavioural level and can be traced back early in development.
In general, a colateralization of spatial and language processing has been observed. For instance, language processing and spatial attention were shown to be controlled by opposite hemispheres^{80}. While language is considered to be leftlateralized in about 95% of righthanders and 75% of lefthanders, the rest 25% of lefthanders demonstrate either a rightlateralized or bilateral representation of language^{5,6,7}. Although spatial processing is generally considered to be lateralized to the right hemisphere in righthanders^{4}, it was shown that lefthanders with right hemispheric dominance for language had lefthemispheric dominance for spatial attention^{80}. If numerical cognition builds on spatial processing^{81} or linguistic processing^{77}, the functional lateralization of these basic cognitive functions may lead to a similar colateralization of number processing.
The colateralization of different cognitive functions requires further investigation in the future in order to determine the position of arithmetic processing in relation to the processing of language and space. The current study is limited in this regard, since language dominance was not assessed and the observed difference in handedness might reflect language lateralization to a certain extent^{82}. However, if the colateralization account also holds for arithmetic processing, a lateralization of arithmetic seems reasonable because of the interrelation of arithmetic with language and spatial processing.
Endpoint of different developmental trajectories
The different lateralization of arithmetic processing in righthanders and lefthanders might be further supported by developmental findings for the IPS. Namely, there might be different trajectories of development of the two hemispheres in left and righthanders. In righthanders, activation in the left IPS for arithmetic processing increases during development^{83,84,85,86}^{, but see}^{44,87}. On the other hand, there is much less evidence for a similar activation increase in the right IPS^{83,86} and it might even be that there is no such activation change in the right IPS during development^{88,89}. Since no literature on the development of the left and right IPS exists for lefthanders, we can only speculate and hypothesize the opposite pattern for lefthanders. This means an activation increase in the right IPS during arithmetic development for lefthanders (cf. Fig. 3), explaining our finding of a rightlateralization in lefthanders compared to a leftlateralization in righthanders for arithmetic processing. In sum, the functional lateralization of arithmetic processing might be a result of increased activation in the left IPS in righthanders and increased activation in the right IPS in lefthanders during arithmetic development (cf. Fig. 3).
It is important to note that all three accounts are not mutually exclusive. For instance, this proposed developmental account of functional lateralization for arithmetic processing is supported by both the embodiment account as well as the colateralization account. In regards to embodied cognition, due to the strong relation between motor activities (like handwriting for symbolic arithmetic or even numberrelated motor activities like finger counting) with the right hand and contralateral lefthemispheric activation, arithmetic processing might be specialized more to the left parietal cortex in righthanders during development, while in lefthanders arithmetic processing might be specialized to the right parietal cortex. In regards to colateralization, number processing was found to be initially represented in the right IPS^{90,91,92} and magnitude relates more to its visuospatial representation, which was shown to be rightlateralized as well. During development, arithmetic processing becomes more related to verbal representations and thus leftlateralization for righthanders and rightlateralization for lefthanders becomes more prominent because of the hemispheric dominance for language processing. The proposed developmental model (cf. Fig. 3) reflects a proposal for the development of functional lateralization of arithmetic processing, which needs to be empirically evaluated in future research.
For righthanders, we did not find evidence for a strong leftlateralization within the parietal cortex. This might be due to the point that approximate calculation in comparison to exact calculation was shown to be rather bilaterally represented in the parietal cortex in righthanders^{30,31}. Furthermore, the complexity of the arithmetic problems used in the current study was relatively high, so it could lead to less pronounced leftlateralization^{43,45}. However, these explanations would also hold for lefthanders and thus contradict the detected lateralization effect in lefthanders. Therefore, the explanation might be derived from the hypothetical model (cf. Fig. 3), where the degree of lateralization in the IPS is expected to be larger in lefthanders than in righthanders (given a similar slope of IPS development in righthanders and lefthanders).
Limitations
In the current study, we did not control for saccades during calculation so that a possible different pattern of saccades in left and righthanders might have had an impact on the lateralization of neural activation. On the one hand, it represents a general limitation for neuroimaging research that saccades might differ between groups or conditions (although the same stimuli were shown to both groups). On the other hand, this might represent a specific problem for research on arithmetic because of operational momentum effects^{93,94,95} (although the stimulus material consisted of both addition and subtraction problems in equal parts, so that this very effect is not a problem for our study). However, we cannot exclude that other sources of eyemovement behavior differ for left and righthanders, although we do not find strong indication in the literature that this should be the case in our paradigm. Nevertheless, a difference in saccades might be an alternative explanation for the lateralization effects observed in our study, which needs future investigations.
Conclusions
Functional lateralization, as indicated by handedness, – as previously shown for other domains like language and space – was demonstrated here for arithmetic processing. Our results refine the view that the bilateral IPS involvement for arithmetic processing, usually observed in empirical studies and postulated by the dominant models of numerical cognition, is similar for all humans. Namely, we observed lefthanders to have a stronger rightlateralization in the IPS for symbolic approximate calculation compared to righthanders. We proposed three different accounts for this functional lateralization:

(i)
The embodiment account: Because of the embodied influences even on basic numerical representations, the preferred use of the dominant hand in numberunrelated, but also numberrelated activities (finger counting) might determine the functional lateralization of arithmetic processing.

(ii)
The colateralization account: Because numbers are tightly linked to spatial and language processing, the functional lateralization of these cognitive functions may be associated with the functional lateralization of arithmetic processing.

(iii)
The developmental account: Because of individual differences in developmental trajectories, which determine lateralization in other behaviours such as handedness, language, and spatial processes, the functional lateralization of arithmetic processing at the endpoint of this development may also be different.
Note that these three accounts are not mutually exclusive. We proposed how they could function and also how they could be tested in future. We believe that this endeavour is of utter importance for several reasons. First, the phenomenon of lefthandedness cannot be neglected in research and models of arithmetic processing. Second, functional lateralization is considered to be beneficial for cognitive skills^{8} and thus understanding the mechanisms might be beneficial for arithmetic education. Finally, such research addresses general mechanisms of neurocognitive functioning (embodiment, coactivation, development), for which functional lateralization research provides a critical test, especially if it holds even for abstract representations such as number magnitude in the IPS. Since the power and design of our study were limited, we recommend more powerful studies investigating several aspects of arithmetic processing (e.g., symbolic vs. nonsymbolic, exact vs. approximate calculation) and functional lateralization (e.g., handedness, language dominance) to followup. In any case, we believe that this study on functional lateralization of arithmetic processing provides a good starting point for a future research line.
Data availability
The datasets generated during the current study are not publicly available due to ethical restrictions but are available from the corresponding author on reasonable request.
References
Steenhuis, R. E. & Bryden, M. P. Different dimensions of hand preference that relate to skilled and unskilled activities. Cortex 25, 289–304 (1989).
Willems, R. M., Der Haegen, L., Van, Fisher, S. E. & Francks, C. On the other hand: Including lefthanders in cognitive neuroscience and neurogenetics. Nat. Rev. Neurosci. 15, 193–201 (2014).
Somers, M. et al. On the relationship between degree of handpreference and degree of language lateralization. Brain Lang. 144, 10–15 (2015).
Vogel, J. J., Bowers, C. A. & Vogel, D. S. Cerebral lateralization of spatial abilities: A metaanalysis. Brain Cogn. 52, 197–204 (2003).
Bethmann, A., Tempelmann, C., De Bleser, R., Scheich, H. & Brechmann, A. Determining language laterality by fMRI and dichotic listening. Brain Res. 1133, 145–157 (2007).
Szaflarski, J. P. et al. Language lateralization in lefthanded and ambidextrous people: fMRI data. Neurology 59, 238–244 (2002).
Knecht, S. Handedness and hemispheric language dominance in healthy humans. Brain 123, 2512–2518 (2000).
Gotts, S. J. et al. Two distinct forms of functional lateralization in the human brain. Proc. Natl. Acad. Sci. 110, E3435–E3444 (2013).
Arsalidou, M. & Taylor, M. J. Is 2+2=4? Metaanalyses of brain areas needed for numbers and calculations. Neuroimage 54, 2382–93 (2011).
Dehaene, S., Piazza, M., Pinel, P. & Cohen, L. Three parietal circuits for number processing. Cogn. Neuropsychol. 20, 487–506 (2003).
Dehaene, S. & Cohen, L. Towards an anatomical and functional model of number processing. Math. Cogn. 1, 83–120 (1995).
Klein, E. et al. Considering structural connectivity in the triple code model of numerical cognition: differential connectivity for magnitude processing and arithmetic facts. Brain Struct. Funct. 221, 979–995 (2016).
Ansari, D., Dhital, B. & Siong, S. C. Parametric effects of numerical distance on the intraparietal sulcus during passive viewing of rapid numerosity changes. Brain Res. 1067, 181–188 (2006).
Piazza, M., Izard, V., Pinel, P., Le Bihan, D. & Dehaene, S. Tuning curves for approximate numerosity in the human intraparietal sulcus. Neuron 44, 547–55 (2004).
Menon, V., Rivera, S. M., White, C. D., Glover, G. H. & Reiss, A. L. Dissociating prefrontal and parietal cortex activation during arithmetic processing. Neuroimage 12, 357–65 (2000).
Burbaud, P. et al. Lateralization of prefrontal activation during internal mental calculation: a functional magnetic resonance imaging study. J. Neurophysiol. 74, 2194–200 (1995).
Rueckert, L. et al. Visualizing Cortical Activation during Mental Calculation with Functional MRI. Neuroimage 3, 97–103 (1996).
Rickard, T. C. et al. The calculating brain: an fMRI study. Neuropsychologia 38, 325–35 (2000).
Chochon, F., Cohen, L., Moortele, P. Fvande & Dehaene, S. Differential Contributions of the Left and Right Inferior Parietal Lobules to Number Processing. J. Cogn. Neurosci. 11, 617–630 (1999).
Zago, L. et al. Neural correlates of simple and complex mental calculation. Neuroimage 13, 314–27 (2001).
Pesenti, M., Thioux, M., Seron, X. & De Volder, A. Neuroanatomical aubstrates of Arabic number processing, numeral comparison, and simple addition: A PET study. J. Cogn. Neurosci. 12, 461–479 (2000).
Cappelletti, M., Muggleton, N. & Walsh, V. Quantity without numbers and numbers without quantity in the parietal cortex. Neuroimage 46, 522–9 (2009).
Klein, E. et al. Bilateral bicephalic tDCS with two active electrodes of the same polarity modulates bilateral cognitive processes differentially. PLoS One 8, e71607 (2013).
Andres, M., Pelgrims, B., Michaux, N., Olivier, E. & Pesenti, M. Role of distinct parietal areas in arithmetic: An fMRIguided TMS study. Neuroimage 54, 3048–3056 (2011).
Hauser, T. U., Rotzer, S., Grabner, R. H., Mérillat, S. & Jäncke, L. Enhancing performance in numerical magnitude processing and mental arithmetic using transcranial Direct Current Stimulation (tDCS). Front. Hum. Neurosci. 7, 244 (2013).
Andres, M., Seron, X. & Olivier, E. Hemispheric lateralization of number comparison. Cogn. Brain Res. 25, 283–290 (2005).
Artemenko, C., Moeller, K., Huber, S. & Klein, E. Differential influences of unilateral tDCS over the intraparietal cortex on numerical cognition. Front. Hum. Neurosci. 9 (2015).
Venkatraman, V., Ansari, D. & Chee, M. W. L. Neural correlates of symbolic and nonsymbolic arithmetic. Neuropsychologia 43, 744–53 (2005).
Vogel, S. E. et al. The left intraparietal sulcus adapts to symbolic number in both the visual and auditory modalities: Evidence from fMRI. Neuroimage 153, 16–27 (2017).
Dehaene, S., Spelke, E., Pinel, P., Stanescu, R. & Tsivkin, S. Sources of mathematical thinking: behavioral and brainimaging evidence. Science 284, 970–4 (1999).
StanescuCosson, R. et al. Understanding dissociations in dyscalculia: a brain imaging study of the impact of number size on the cerebral networks for exact and approximate calculation. Brain 123, 2240–55 (2000).
Coren, S. & Porac, C. Fifty Centuries of RightHandedness: The Historical Record Behavioral Choice: Neural Mechanisms in Pleurobranchaea. Science 198, 631–632, https://doi.org/10.1126/science.335510 (1977).
PapadatouPastou, M. et al. Hand preference and Mathematical Learning Difficulties: New data from Greece, the United Kingdom, and Germany and two metaanalyses of the literature. 1–99 (2019).
Sala, G., Signorelli, M., Barsuola, G., Bolognese, M. & Gobet, F. The relationship between handedness and mathematics is nonlinear and is moderated by gender, age, and type of task. Front. Psychol. 8, 1–14 (2017).
Whittington, J. E. & Richards, P. N. Mathematical ability and the rightshift theory of handedness. Neuropsychologia 29, 1075–1082 (1991).
Cheyne, C. P., Roberts, N., Crow, T. J., Leask, S. J. & GarcíaFiñana, M. The effect of handedness on academic ability: A multivariate linear mixed model approach. Laterality 15, 451–464 (2010).
Benbow, C. P. Physiological Correlates of Extreme Intellectual Precocity. Neuropsychologia 24, 719–725 (1986).
Annett, M. & Kilshaw, D. Mathematical Ability and Lateral Asymmetry. Cortex 18, 547–568 (1982).
Cipora, K., Soltanlou, M., Reips, U. D. & Nuerk, H. C. The SNARC and MARC effects measured online: Largescale assessment methods in flexible cognitive effects. Behav. Res. Methods. 51, 1676–1692 (2019).
Huber, S. et al. Embodied markedness of parity? Examining handedness effects on parity judgments. Psychol. Res. 79, 963–977 (2015).
Patro, K., Nuerk, H. C. & Cress, U. Does your body count? Embodied influences on the preferred counting direction of preschoolers. J. Cogn. Psychol. 27, 413–425 (2015).
Huber, S., Nuerk, H. C., Reips, U. D. & Soltanlou, M. Individual differences influence twodigit number processing, but not their analog magnitude processing: a largescale online study. Psychol. Res. 83, 1444–1464 (2019).
Zhang, Y., Zhang, Q., Zhang, J. & Li, W. Laterality of brain areas associated with arithmetic calculations revealed by functional magnetic resonance imaging. Chin. Med. J. (Engl). 118, 633–638 (2005).
Artemenko, C., Soltanlou, M., Ehlis, A. C., Nuerk, H. C. & Dresler, T. The neural correlates of mental arithmetic in adolescents: A longitudinal fNIRS study. Behav. Brain Funct. 14, 1–13 (2018).
Soltanlou, M. et al. Increased arithmetic complexity is associated with domaingeneral but not domainspecific magnitude processing in children: A simultaneous fNIRSEEG study. Cogn. Affect. Behav. Neurosci 17, 724–736 (2017).
Oldfield, R. The assessment and analysis of handedness: The Edinburgh inventory. Neuropsychologia 9, 97–113 (1971).
Klein, E., Nuerk, H.C., Wood, G., Knops, A. & Willmes, K. The exact vs. approximate distinction in numerical cognition may not be exact, but only approximate: How different processes work together in multidigit addition. Brain Cogn. 69, 369–81 (2009).
Lakens, D. Calculating and reporting effect sizes to facilitate cumulative science: a practical primer for ttests and ANOVAs. Front. Psychol. 4, 1–12 (2013).
Cui, X., Bray, S. & Reiss, A. L. Functional near infrared spectroscopy (NIRS) signal improvement based on negative correlation between oxygenated and deoxygenated hemoglobin dynamics. Neuroimage 49, 3039–3046 (2010).
TzourioMazoyer, N. et al. Automated Anatomical Labeling of Activations in SPM Using a Macroscopic Anatomical Parcellation of the MNI MRI SingleSubject Brain. Neuroimage 15, 273–289 (2002).
Jurcak, V., Tsuzuki, D. & Dan, I. 10/20, 10/10, and 10/5 systems revisited: Their validity as relative headsurfacebased positioning systems. Neuroimage 34, 1600–1611 (2007).
Kennan, R. P., Kim, D., Maki, A., Koizumi, H. & Constable, R. T. Noninvasive assessment of language lateralization by transcranial near infrared optical topography and functional MRI. Hum. Brain Mapp. 16, 183–189 (2002).
Watanabe, E. et al. Noninvasive assessment of language dominance with nearinfrared spectroscopic mapping. Neurosci. Lett. 256, 49–52 (1998).
Goffin, C., Sokolowski, H. M. & Ansari, D. Does writing handedness affect neural representation of symbolic number? An fMRI Adaptation Study. Cortex 121, 27–43 (2019).
Domahs, F., Moeller, K., Huber, S., Willmes, K. & Nuerk, H.C. Embodied numerosity: implicit handbased representations influence symbolic number processing across cultures. Cognition 116, 251–66 (2010).
Moeller, K. et al. Learning and development of embodied numerosity. Cogn. Process. 13, 271–4 (2012).
Myachykov, A., Scheepers, C., Fischer, M. H. & Kessler, K. TEST: A tropic, embodied, and situated theory of cognition. Top. Cogn. Sci. 6, 442–460 (2014).
Dackermann, T., Fischer, U., Nuerk, H.C., Cress, U. & Moeller, K. Applying embodied cognition: from useful interventions and their theoretical underpinnings to practical applications. ZDM  Math. Educ. 49, 545–557 (2017).
Lindemann, O. & Tira, M. D. Operational Momentum in Numerosity Production Judgments of MultiDigit Number Problems. J. Psychol. 219, 50–57 (2011).
Fischer, M. H. Finger counting habits modulate spatialnumerical associations. Cortex 44, 386–392 (2008).
Wasner, M., Moeller, K., Fischer, M. H. & Nuerk, H.C. H. C. Aspects of situated cognition in embodied numerosity: The case of finger counting. Cogn. Process. 15, 317–328 (2014).
Klein, E., Moeller, K., Willmes, K., Nuerk, H.C. & Domahs, F. The influence of implicit handbased representations on mental arithmetic. Front. Psychol. 2, 197 (2011).
Andres, M., Michaux, N. & Pesenti, M. Common substrate for mental arithmetic and finger representation in the parietal cortex. Neuroimage 62, 1520–1528 (2012).
Tschentscher, N., Hauk, O., Fischer, M. H. & Pulvermüller, F. You can count on the motor cortex: finger counting habits modulate motor cortex activation evoked by numbers. Neuroimage 59, 3139–48 (2012).
Andres, M., Seron, X. & Olivier, E. Contribution of hand motor circuits to counting. J. Cogn. Neurosci. 19, 563–576 (2007).
Brookshire, G. & Casasanto, D. Motivation and motor control: Hemispheric specialization for approach motivation reverses with handedness. PLoS One 7, 4–8 (2012).
Patro, K., Nuerk, H. C., Cress, U. & Haman, M. How numberspace relationships are assessed before formal schooling: A taxonomy proposal. Front. Psychol. 5, 1–6 (2014).
Cipora, K., Patro, K. & Nuerk, H.C. Are SpatialNumerical Associations a Cornerstone for Arithmetic Learning? The Lack of Genuine Correlations Suggests No. Mind, Brain, Educ. 9, 190–206 (2015).
Cipora, K., Schroeder, P. A., Soltanlou, M. & Nuerk, H.C. More space, better math: Is space a powerful tool or a conerstone for understanding arithmetic? In Visualizing Mathematics. Research In Mathematics Education (eds Mix, K. S. & Battista, M. T.) 77–116 (Springer, Cham 2018).
Fischer, M. H. & Shaki, S. Spatial associations in numerical cognition  From single digits to arithmetic. Q. J. Exp. Psychol. 67, 1461–1483 (2014).
Dehaene, S., Bossini, S. & Giraux, P. The mental representation of parity and number magnitude. J. Exp. Psychol. Gen. 122, 371–396 (1993).
de Hevia, M. D. & Spelke, E. S. Numberspace mapping in human infants. Psychol. Sci. 21, 653–660 (2010).
Di Giorgio, E. et al. A mental number line in human newborns. Dev. Sci. 22, e12801 (2019).
Rugani, R., Vallortigara, G., Priftis, K. & Regolin, L. Numberspace mapping in the newborn chick resembles humans’ mental number line. Science 347, 534–536 (2015).
JohnsonUlrich, Z. & Vonk, J. Spatial representation of magnitude in humans (Homo sapiens), Western lowland gorillas (Gorilla gorilla gorilla), and American black bears (Ursus americanus). Anim. Cogn. 21, 531–550 (2018).
Dowker, A. & Nuerk, H.C. Editorial: Linguistic influences on mathematics. Front. Psychol. 7, 1–4 (2016).
Sarnecka, B. W. On the relation between grammatical number and cardinal numbers in development. Front. Psychol. 5, 1–4 (2014).
Shaki, S., Fischer, M. & Petrusic, W. Reading habits for both words and numbers contribute to the SNARC effect. Psychon. Bull. Rev. 16, 328–31 (2009).
Nuerk, H.C., Weger, U. & Willmes, K. Language effects in magnitude comparison: Small, but not irrelevant. Brain Lang. 92, 262–277 (2005).
Cai, Q., Van der Haegen, L. & Brysbaert, M. Complementary hemispheric specialization for language production and visuospatial attention. Proc. Natl. Acad. Sci. 110, E322–E330 (2013).
Patro, K., Nuerk, H.C. & Cress, U. Mental Number Line in the Preliterate Brain: The Role of Early Directional Experiences. Child Dev. Perspect. 10, 172–177 (2016).
Forrester, G. S. & Quaresmini, C. The Right Hand Man: Manual Laterality and Language. In Behavioral Lateralization in Vertebrates (eds Csermely, D. & Regolin, L.) 125–141 (Springer 2013).
Chang, T.T., Metcalfe, A. W. S., Padmanabhan, A., Chen, T. & Menon, V. Heterogeneous and nonlinear development of human posterior parietal cortex function. Neuroimage 126, 184–195 (2016).
Rivera, S. M., Reiss, A. L., Eckert, M. A. & Menon, V. Developmental changes in mental arithmetic: evidence for increased functional specialization in the left inferior parietal cortex. Cereb. cortex 15, 1779–90 (2005).
Kucian, K., von Aster, M., Loenneker, T., Dietrich, T. & Martin, E. Development of neural networks for exact and approximate calculation: a FMRI study. Dev. Neuropsychol. 33, 447–73 (2008).
RosenbergLee, M., Barth, M. & Menon, V. What difference does a year of schooling make? Maturation of brain response and connectivity between 2nd and 3rd grades during arithmetic problem solving. Neuroimage 57, 796–808 (2011).
Qin, S. et al. Hippocampalneocortical functional reorganization underlies children’s cognitive development. Nat. Neurosci. 17, 1263–1269 (2014).
Prado, J., Mutreja, R. & Booth, J. R. Developmental dissociation in the neural responses to simple multiplication and subtraction problems. Dev. Sci. 17, 537–552 (2014).
Emerson, R. W. & Cantlon, J. F. Continuity and change in children’s longitudinal neural responses to numbers. Dev. Sci. 18, 314–326 (2015).
Hyde, D., Boas, D., Blair, C. & Carey, S. Nearinfrared spectroscopy shows right parietal specialization for number in preverbal infants. Neuroimage 53, 647–652 (2010).
Edwards, L. A., Wagner, J. B., Simon, C. E. & Hyde, D. C. Functional brain organization for number processing in preverbal infants. Dev. Sci. 19, 757–769 (2016).
Izard, V., DehaeneLambertz, G. & Dehaene, S. Distinct cerebral pathways for object identity and number in human infants. PLoS Biol. 6, e11 (2008).
Knops, A., Dehaene, S., Berteletti, I. & Zorzi, M. Can approximate mental calculation account for operational momentum in addition and subtraction? Q. J. Exp. Psychol. 67, 1541–1556 (2014).
Klein, E., Huber, S., Nuerk, H.C. H.C. & Moeller, K. Operational momentum affects eye fixation behaviour. Q. J. Exp. Psychol. 67, 1614–1625 (2014).
Masson, N. & Pesenti, M. Attentional bias induced by solving simple and complex addition and subtraction problems. Q. J. Exp. Psychol. 67, 1514–1526 (2014).
Acknowledgements
This research was funded by the LEAD Graduate School & Research Network [GSC1028], which is funded within the framework of the Excellence Initiative of the German federal and state governments. This research was further funded by a grant from DAAD for research stays for university academics and scientists supporting M.A.S. M.S. was supported by the Deutsche Forschungsgemeinschaft grant [NU 265/31] to H.C.N. We want to thank all participants of the study, and also Anne Büsemeyer and Andra Coldea for assistance in the measurements and Zoë Kirste for language proofreading of the paper.
Author information
Authors and Affiliations
Contributions
M.A.S., H.C.N., C.A. and T.D. designed the study. M.A.S. and C.A. collected data. C.A., M.A.S. and M.S. analyzed data. M.S., C.A. and H.C.N. conceptualized the study. C.A. and M.A.S. wrote the first draft of the manuscript. All authors reviewed the manuscript.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Artemenko, C., Sitnikova, M.A., Soltanlou, M. et al. Functional lateralization of arithmetic processing in the intraparietal sulcus is associated with handedness. Sci Rep 10, 1775 (2020). https://doi.org/10.1038/s41598020584777
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598020584777
This article is cited by

Training causes activation increase in temporoparietal and parietal regions in children with mathematical disabilities
Brain Structure and Function (2022)

Space, time and number: common coding mechanisms and interactions between domains
Psychological Research (2022)

Frontal and parietal EEG alpha asymmetry: a largescale investigation of shortterm reliability on distinct EEG systems
Brain Structure and Function (2022)
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.