Objective We explored the development of complexity in brain endogenous oscillations accompanied by normal aging by using resting-state fMRI, which could provide a new characteristic parameter for aging assessment. Methods Firstly, the regions with significant effect on Hurst exponent in whole brain were obtained from elderly and young groups, which were defined as regions of interest (ROI). Secondly, we implemented a correlation analysis between the mean Hurst exponent of each ROI and age. Results The regions with significant difference were localised to middle frontal gyrus, putamen, thalamus, middle occipital gyrus, precuneus, precentral and postcentral gyrus (P<0.05, k="">24, Alphasim correction). The mean Hurst exponents in the left precentral and postcentral gyrus displayed strongly negative association with age and the correlation coefficients were -0.76 and -0.82 respectively. The Hurst exponent in right putamen showed a high positive correlation with R=0.73. In addition, the other regions were demonstrated to be moderate correlations. Conclusions The Hurst exponent within specific brain functional regions has a close linear relationship with age, which can be regarded as a characteristic parameter for normal aging.