The validity and reliability of quantifying hemispheric specialisation using fMRI: Evidence from left and right handers on three different cerebral asymmetries

Neuroimaging has tremendous potential for quantifying hemispheric specializations. However, the possibilities remain under-utilized, in part, given some of the complexities in quantifying any differences in a reliable, transparent fashion. A second issue with hemispheric asymmetries is that they are extremely one-sided in most people. This skew limits the generalisability of any findings to those participants with rarer forms of cerebral asymmetry. Here, we demonstrate usefulness of an approach developed by Wilke and Lidzba, (J Neurosci Meth, 163, 2007), which allows for threshold-independent estimates of cerebral asymmetry to be calculated in individual participants. We compared these estimates from two separate runs for three different cerebral asymmetries in the same participants. We circumvented the skewed nature of this type of data in two ways; first, we scanned a large number of non-right handed participants, and second, we included asymmetries that favour the right hemisphere in right handers, which we had reason to believe were less skewed than those related to speech and language. Verbal fluency and two visuoperceptual asymmetries were localized in a sample of 33 right handed and 60 non-right handed participants. Laterality indices (LIs), which quantify the direction and strength of an asymmetry, were calculated for BOLD activity relating to language, face perception, and body perception in each run separately. Run 1 - run 2 correlations were all statistically significant and surprisingly sizeable (r = 0.89 to r = 0.62), considering the relatively short amount of time on task within our particular localizers. This noteworthy success validates a number of useful ways that functional neuroimaging can be used to advance understanding of cerebral asymmetries.