Different patterns of brain activity are seen in different subjects across a wide functional domains, however the meaning of these individual differences are not fully understood and these individual differences are often neglected through the group average. Based on the fundamental assumption that human behavior is rooted in the underlying brain function, we speculate that individual differences in brain function are also reflected in the individual differences in behaviors. Adopting 98 behavioral measures and the brain activity induced at 7 task fMRI states, we demonstrated that the individual differences in brain activity could be used to predict behavioral measures of unseen subjects with high accuracies using the partial least square regression model. We also revealed that the behavior-relevant individual differences transferred between different task states and could be used to reconstruct the individual brain activity. The reconstructed individual brain activity retained some individual differences that were lost in the group average and could be served as an individual functional localizer. Therefore, our results suggest that the individual differences in brain activity contain behavior-relevant information and should not be neglected in the group average. Moreover, the reconstructed individual brain activity also shows the potential use in precise and personalized medicine.