Individual differences in reading skill and language lateralisation: A cluster analysis

Abstract

Individual differences in reading and cerebral lateralisation were investigated in 200 college students who completed reading assessments and divided visual field word recognition tasks, and received a structural MRI scan. Prior studies on this data set indicated that little variance in brain–behaviour correlations could be attributed to the effects of sex and handedness variables (Chiarello, Welcome, Halderman, & Leonard, 2009; Chiarello, Welcome, Halderman, Towler, et al., 2009; Welcome et al., 2009). Here a more bottom-up approach to behavioural classification (cluster analysis) was used to explore individual differences that need not depend on a priori decisions about relevant subgroups. The cluster solution identified four subgroups of college age readers with differing reading skill and visual field lateralisation profiles. These findings generalised to measures that were not included in the cluster analysis. Poorer reading skill was associated with somewhat reduced VF asymmetry, while average readers demonstrated exaggerated RVF/left hemisphere advantages. Skilled readers had either reduced asymmetries, or asymmetries that varied by task. The clusters did not differ by sex or handedness, suggesting that there are identifiable sources of variance among individuals that are not captured by these standard participant variables. All clusters had typical leftward asymmetry of the planum temporale. However, the size of areas in the posterior corpus callosum distinguished the two subgroups with high reading skill. A total of 17 participants, identified as multivariate outliers, had unusual behavioural profiles and differed from the remainder of the sample in not having significant leftward asymmetry of the planum temporale. A less buffered type of neurodevelopment that is more open to the effects of random genetic and environmental influences may characterise such individuals.

Keywords:

• Individual differences
• Cerebral asymmetry
• Reading skill
• Cluster analysis
• Corpus callosum
• Structural asymmetry
• Divided visual field
• Word reading
• Language lateralisation
• Planum temporale

Acknowledgements

This research was supported by NIH grant DC006957. We thank Dr Ronald Otto for facilitating scan acquisition and examination, and Laura K. Halderman, Janelle Julagay, Travellia Tjokro, and Stephen Towler for assistance with data collection and analysis.

Notes

¹Participant recruitment for this project was unrestricted for handedness so as to obtain a representative sample of the college age population. Hence there are a relatively small number of left handers (N=22, 11% of sample), precluding strong statistical comparisons between left- and right-handers. For this reason we have explored differences between mixed- (N=97) and consistent-handers (N=103) within our sample. There is an increasing amount of evidence for a variety of behavioural differences between these groups (e.g., Christman, Varalakshmi, & Jasper, 2009; Propper, Christman, & Phaneuf, 2005).

²All data are reported as z-scores to facilitate comparisons across measures. A z-score of 0 represents the sample mean, hence negative z-scores are those falling below the mean. As reported previously (Chiarello, Welcome, Halderman, Towler, et al., 2009) all VF tasks, with the exception of nonword naming RT, resulted in robust RVF/LH advantages. Therefore a z-score of 0 for the VF measures indicates the typical RVF/LH advantage for that task; small negative z-scores indicate a reduced RVF/LH advantage and large negative z-scores a reversed asymmetry.

³Inspection of the untransformed asymmetries indicated that, for individuals in Cluster 2, 11 had reversed or no asymmetry for masked word recognition, 19 had reversed or no asymmetry for verb generation, and 3 had reversed asymmetry for lexical decision.

⁴As noted in a prior publication using this sample (Welcome et al., 2009), two male participants had extremely large corpus callosa. This raises the question as to whether the current findings might have been influenced by two highly unusual cases. When these two individuals were dropped from the current analyses, all of the reported effects remained significant. In fact the main effect for the splenium was more reliable, F(3, 177)=4.11, p=.008, η²=.07, as were all post hoc contrasts (Cluster 2 vs 1, p=.01, d=.70, Cluster 2 vs 3, p=.006, d=.81, Cluster 2 vs 4, p=.04, d=.69).

⁵Cluster 4 individuals also had somewhat variable asymmetries. However, the average asymmetry standard deviations (consistency scores across all seven VF tasks) for these individuals (accuracy asymmetry SD=.858, RT asymmetry SD=.886) were similar to the other clusters and smaller than those observed for the outliers.