Dysfunctional neural network of spatial working memory contributes to developmental dyscalculia

Abstract

The underlying neural mechanisms of developmental dyscalculia (DD) are still far from being clearly understood. Even the behavioral processes that generate or influence this heterogeneous disorder are a matter of controversy. To date, the few studies examining functional brain activation in children with DD mainly focus on number and counting related tasks, whereas studies on more general cognitive domains that are involved in arithmetical development, such as working memory are virtually absent. There are several studies showing a close relationship between DD and spatial working memory [Camos, V. (2008). Low working memory capacity impedes both efficiency and learning of number transcoding in children. Journal of Experimental Child Psychology, 99(1), 37–57; McLean, J. F., & Hitch, G. J. (1999). Working memory impairments in children with specific arithmetic learning difficulties. Journal of Experimental Child Psychology, 74(3), 240–260; Rosselli, M., Matute, E., Pinto, N., & Ardila, A. (2006). Memory abilities in children with subtypes of dyscalculia. Developmental Neuropsychology, 30(3), 801–818; Siegel, L. S., & Ryan, E. B. (1989). The development of working memory in normally achieving and subtypes of learning disabled children. Child Development, 60(4), 973–980]. The relationship between these two mechanisms is still matter of debate, but this study follows the assumption that poor spatial working memory capacity may hinder the acquisition of spatial number representations in children with DD [Geary, D. C. (1993). Mathematical disabilities: Cognitive, neuropsychological, and genetic components. Psychological Bulletin, 114(2), 345–362; von Aster, M., & Shalev, R. S. (2007). Number development and developmental dyscalculia. Developmental Medicine and Child Neurology, 49(11), 868–873].

Using functional MRI the current study compares brain activity associated with spatial working memory processes in 8–10-year-old children with DD and normally achieving controls. Both groups showed significant spatial working memory related activity in a network including occipital and parietal regions. Children with DD showed weaker neural activation compared to the control group during a spatial working memory task in the right intraparietal sulcus (IPS), the right insula and the right inferior frontal lobe. Performance tests outside the scanner showed impaired working memory proficiency in children with DD. Bringing behavioral performance and neural activity together we found significant correlations of right IPS activity with performance on the verbal digit span forward and the spatial Corsi Block Tapping test.

Our findings demonstrate for the first time an involvement of spatial working memory processes in the neural underpinnings of DD. These poor spatial working memory processes may inhibit the formation of spatial number representations (mental numberline) as well as the storage and retrieval of arithmetical facts.

Introduction

Developmental dyscalculia (DD) is characterized by difficulties representing and manipulating numerical information nonverbally and visuo-spatially, in learning and remembering arithmetic facts and in executing arithmetic procedures. DD in children has a prevalence of 3–6% in the school aged population, what is comparable to dyslexia, and high rates of comorbidities, such as ADHD (Koumoula et al., 2004, Shalev et al., 2000). Yet little is known about the underlying deficits. The question of whether this difficulty in learning mathematics is due to a single impairment of a basic number specific core competence (‘number sense’) or a combination of impairments in a more general cognitive system is still open (Butterworth, 2005, Mix and Sandhofer, 2007). One impediment to research on DD is the complexity of the numerical domain that includes verbal, visuo-spatial, memory, and executive functions (Ardila et al., 1998, Geary et al., 2000, von Aster, 2000). This wide array of cognitive factors that could contribute to DD poses a special challenge to investigate this disorder.

Deficits in working memory systems have been argued to substantially contribute to specific deficits in building cognitive representations of number, the formation of concepts and procedures as well as arithmetic fact retrieval in children with DD (Geary, 1993, von Aster and Shalev, 2007). Working memory refers to the mental capacity responsible for the temporary processing and storage of information (Rosselli, Matute, Pinto, & Ardila, 2006). It requires both the simultaneous processing of incoming and the retrieval or manipulation of retained information (Siegel & Ryan, 1989). This capacity for information processing is limited, since higher demands on the former will negatively influence the access to the latter, and vice versa. Therefore, variation in this capacity is related to performance in any cognitive activity (Camos, 2008), including arithmetics.

Several studies investigated the role of working memory in typically achieving children and children with DD (Bull et al., 2008, D’Amico and Guarnera, 2005, Geary et al., 2000, Geary and Hoard, 2001, McLean and Hitch, 1999, Rosselli et al., 2006, Siegel and Ryan, 1989). Van der Sluis, van der Leij, and de Jong (2005) showed that children with arithmetic disabilities performed worse on a task requiring the memorization of dynamic visual information—these results are consistent with other findings (McLean & Hitch, 1999), reporting lower performance of children with arithmetic disabilities on the Corsi Block Tapping task. Concerning the different aspects of working memory, children with poor arithmetic performance generally appear to have normal phonological working memory (McLean and Hitch, 1999, Siegel and Ryan, 1989), although their capacity of spatial working memory is impaired.

Spatial working memory is modulated by a broad network of regions including predominantly frontal and parietal regions. Children performing visuo-spatial working memory tasks show the same, but decreased activation pattern compared to adults, especially the dorsolateral prefrontal cortex is less recruited (Klingberg et al., 2002, Kwon et al., 2002, Nelson et al., 2000, Scherf et al., 2006).

To date, only a few studies have investigated children with DD by means of anatomical or functional brain functions. Kucian et al. (2006) showed that children with DD have weaker brain activation in the IPS and the middle and inferior frontal gyrus of both hemispheres for approximate calculation than typically achieving children. Evidence of parietal dysfunction in DD has also been reported by Price, Holloway, Räsänen, Vesterinen, and Ansari (2007). In a recent study investigating structural brain volume in children with and without DD Rotzer et al. (2008) found reduced grey matter volumes in frontal and parietal regions. Hence, the question arose whether these differences are related to specific number processes or whether they may be attributed to more domain general factors such as working memory and attention.

The current study aims, for the first time, to compare the functional neuroanatomy of children with and without DD while performing a spatial working memory task. We hypothesize that children with DD show weaker activation in brain areas related to spatial working memory, such as the frontal and parietal cortex, since children with DD seem to have impaired spatial working memory capacities that are modulated by frontal and parietal brain regions.