There are different stress-strain definitions to measure the mechanical properties of the brain tissue. However, there is no agreement as to which stress-strain definition should be employed to measure the mechanical properties of the brain tissue at both the longitudinal and circumferential directions. It is worth knowing that an optimize stress-strain definition of the brain tissue at different loading directions may have implications for neuronavigation and surgery simulation through haptic devices. This study is aimed to conduct a comparative study on different results are given by the various definitions of stress-strain and to recommend a specific definition when testing brain tissues. Prepared cylindrical samples are excised from the parietal lobes of rats' brains and experimentally tested by applying load on both the longitudinal and circumferential directions. Three stress definitions (second Piola-Kichhoff stress, engineering stress, and true stress) and four strain definitions (Almansi-Hamel strain, Green-St. Venant strain, engineering strain, and true strain) are used to determine the elastic modulus, maximum stress and strain. The highest non-linear stress-strain relation is observed for the Almansi-Hamel strain definition and it may overestimate the elastic modulus at different stress definitions at both the longitudinal and circumferential directions. The Green-St. Venant strain definition fails to address the non-linear stress-strain relation using different definitions of stress and triggers an underestimation of the elastic modulus. The results suggest the application of the true stress-true strain definition for characterization of the brain tissues mechanics since it gives more accurate measurements of the tissue's response using the instantaneous values.