In this paper, we propose an evolutionary algorithm for high dimensional global optimization, which makes use of correlation coefficients, cooperative coevolution and differential evolution (4CDE). The decision variables are associated in high correlated groups, that also change throughout generations, depending on the area being currently explored. Preliminary results are shown for 50 variables. The experiments are performed with unimodal, multimodal, separable and non-separable functions. The results obtained by 4CDE are generally better than those obtained by differential evolution alone.