Genetic and environmental contributions to brain activation during calculation
Introduction
Mastering the ability to solve mathematical problems, even as simple as addition or subtraction, requires a long period of teaching and training and is reached only through several developmental stages (Butterworth, 2005, Von Aster and Shalev, 2007). Behavioral studies of young infants emphasized the importance of early numerical competences such as approximation, comparison or matching of numerical magnitude that exist prior to education (McCrink and Wynn, 2004, Xu and Spelke, 2000). This initial preverbal system is later refined by education and serves as a starting point for the subsequent acquisition of numerical symbols and emergence of exact calculation (Ansari and Dhital, 2006, Verguts and Fias, 2004). Neuroimaging studies of adults and infants converge to localize this preverbal numerical system within the parietal lobe (Cantlon et al., 2009, Castelli et al., 2006, Dehaene et al., 2003, Piazza et al., 2004, Pinel et al., 2001, Temple and Posner, 1998, Venkatraman et al., 2005). An event-related potential study showed that this parietal representation is present even in newborns (Izard et al., 2009) noticeably in the right hemisphere. These results are coherent with the existence of a core system for numerosities that maybe inherited and shared with other animal species (Cantlon and Brannon, 2007, Nieder and Miller, 2004), and forms a promising context for the search of genetic determinant of numerical skills.
Indeed, several experimental approaches consistently demonstrated the importance of genetic factors in explaining individual variability in arithmetic, both in normal and in pathological ranges. Familial studies of developmental dyscalculia, an impairment in the acquisition of arithmetical skills, showed a high prevalence of this developmental disease among siblings (Alarcón et al., 1997, Shalev et al., 2001). Quantitative genetic studies suggested a heritability estimate ranging from 0.2 to 0.9 (Oliver et al., 2004). These results were recently comforted by a genome-wide association analysis, performed on two distinct samples, which isolated ten genetic polymorphisms that may partially explain variation of individual performance in mathematics (Docherty et al., 2010). However, it is unclear whether this genetic contribution is directly related to the core numerical system or whether it reflects variations in parts of the arithmetical network related to language. Indeed, the same authors reported, based on a large twin study, that two-thirds of the genetic factors that contribute to variation in mathematic also affect reading performance (Plomin and Kovas, 2005).
No study to date directly explored the localization of the cerebral regions underlying this genetic contribution. The only available data come from clinical populations characterized by genetic anomalies and which also present impairments in numerical abilities. The most documented pathologies are the VeloCardioFacial Syndrome (VCF) caused by a micro-deletion in the chromosome 22q11, the Turner syndrome caused by the deletion of one X chromosome, and the fragile X syndrome, caused by a mutation of the FMR1 gene. Extensive investigation of VCF patients showed both functional (Eliez et al., 2001) and anatomical (Barnea-Goraly et al., 2005, Eliez et al., 2000) abnormalities in the left parietal region, nearby the supramarginal gyrus. A functional magnetic resonance imaging (fMRI) study of female subjects with fragile X syndrome also demonstrated a correlation between FMR1 protein expression and activation of the left fronto-parietal network during a mental calculation task, in particular around the supramarginal/angular gyrus region (Rivera et al., 2002). Finally, a neuroimaging study of Turner syndrome showed that patients presented an abnormal recruitment of the bilateral intraparietal sulci (IPS) during exact and approximate calculation task, as well as an abnormal anatomical organization of the right intraparietal sulcus (Molko et al., 2003). All of these studies point to an atypical development of the parietal lobe. However, the severity of these diseases, usually accompanied by other cognitive impairments, does not allow evaluating the specificity of the genetic impact onto numerical cognition. It remains unclear how genetic determinants interact with education at the cerebral level in a typically developed brain.
To shed light on this question we collected behavioral and fMRI data on monozygotic (MZ) and dizygotic (DZ) twin adults performing mental calculation tasks. Because MZ twins share 100% of their genetic polymorphism while DZ twins share only 50%, any greater similarity in MZ compared to DZ should reflect a genetic contribution, given that both MZ and DZ pairs are supposed to share equally similar environments. We then aimed here to separate, within the calculation network, activations under genetic contribution from those under environmental contribution. This mapping should help us to understand if the numerical core system is effectively influenced by genetic factors. We also examined the correlations between individual fMRI activations and behavioral scores in arithmetic, an approach which may help isolate the components of the arithmetical network that contribute to the large heritability reported for mathematical skills. Finally, because it has been hypothesized that learning arithmetical rules is accompanied by the recycling of evolutionary older parietal functions, including the nearby parietal regions involved in eye movements (Knops et al., 2009, Simon et al., 2002), we also collected activations related to saccade to test whether they share any genetic contribution with the arithmetical network.
Section snippets
Subjects
19 pairs of monozygotic (MZ) twins (mean age = 23.2 years old) and 13 pairs of dizygotic (DZ) twins (mean age = 22.3 years old) participated to our study. They were all healthy right handed male adults. Zygosity was determined by genetic analysis of single nucleotide polymorphisms (SNPs) extracted from subjects' saliva (DNA collection kit from DNA Genotek/OG-250, DNA Genotek). DNA was collected in a small volume of 200 μL of TE10:1 and transferred to the French Centre National de Génotypage for
Results
Scores from the behavioral calculation task show that MZ and DZ performed equally well on average (mean MZ score = 0.73, standard deviation (sd) = 0.21; mean DZ score = 0.73, sd = 0.21). In agreement with previous behavioral studies using larger numbers of subjects (Kovas et al., 2007b, Loehlin and Nichols, 1976, Oliver et al., 2004, Wadsworth et al., 1995), calculation scores were highly correlated within MZ siblings (ICCMZ = 0.53, p = 0.007; ICCDZ = 0.12, p > 0.1), and both estimates of heritability (h2F =
Discussion
Comparing the functional correlation in the brain activation of MZ and DZ twins during a subtraction task, we decomposed the cerebral network activated during mental arithmetic into regions showing either a genetic or an environmental contribution. Although the activity evoked by the subtraction task covers a large set of cortical and subcortical areas, functional similarity in brain activation between twins was essentially restricted to the posterior part of the parietal lobe, while no
Conclusion
In conclusion, our results show that genetic determinants and shared environment have distinct and complementary impacts on the cerebral circuits for arithmetic. Both a superior frontoparietal set of areas, which also support eye movements, and the left angular gyrus, are subject to a genetic contribution. Meanwhile, the shared environment (most likely including education practices) primarily affects the functional lateralization of activation in the intraparietal sulcus. Future experiments,
Acknowledgments
We thank C. Lalanne, T. Bourgeron, F. Faucherau, E. Artiges, D. LeBihan, G. Dehaene-Lambertz and A. Jobert for their contribution to data acquisition or analysis. We gratefully acknowledge the CC-IN2P3 (Centre de Calcul de l'Institut National de Physique Nucléaire et de Physique des Particules) for providing a significant amount of the computing resources and services needed for this work, as well as P Calvat and S. Daligault and C. Delpuech for their help. This study was supported financially
References (99)
- et al.
Math learning disorder: incidence in a population-based birth cohort, 1976–82, Rochester, Minn
Ambul. Pediatr.
(2005) - et al.
Arithmetic ability and parietal alterations: a diffusion tensor imaging study in velocardiofacial syndrome
Cogn. Brain Res.
(2005) - et al.
Quantifying the heritability of task-related brain activation and performance during the N-back working memory task: a twin fMRI study
Biol. Psychol.
(2008) - et al.
A cognitive characterization of dyscalculia in Turner syndrome
Neuropsychologia
(2004) - et al.
The role of the left intraparietal sulcus in the relationship between symbolic number processing and children's arithmetic competence
Dev. Cogn. Neurosci.
(2012) - et al.
Mathematical disabilities in children with velo-cardio-facial syndrome
Neuropsychologia
(2007) - et al.
Cultural recycling of cortical maps
Neuron
(2007) - et al.
Cerebral activations during number multiplication and comparison: a PET study
Neuropsychologia
(1996) - et al.
Learning complex arithmetic—an fMRI study
Cogn. Brain Res.
(2003) - et al.
Common regions of the human frontal lobe recruited by diverse cognitive demands
Trends Neurosci.
(2000)