Myeloarchitecture of cerebral cortex has crucial implication on the function of cortical columnar modules. Based on the recent development of high-field magnetic resonance imaging (MRI), it was demonstrated that it is possible to individually reconstruct such intracortical microstructures. However, there is a scarcity of publicly available frameworks to perform group-wise statistical inferences on high resolution data. In this paper, we present a novel framework that parameterizes curved brain structures in order to construct correspondences across subjects without deforming individual geometry. We use the second Laplace-Beltrami eigenfunction to build such a parameterization, which is known to monotonically increase along the longest geodesic distance on an arbitrary manifold. To demonstrate our framework, a study on the lateralization of Heschl's gyrus is presented with multiple comparison correction.