Hybrid metaheuristics have received considerable interest in recent years. Since several years ago, a wide variety of hybrid approaches have been proposed in the literature including the new GA-EDA approach. We have design and implemented an extension to this GA-EDA approach, based on statistical significance tests. This approach had allowed us to make an study of the bal- ance of diversification (exploration) and intensification (exploitation) in Genetic Algorithms and Estimation of Distribution Algorithms. Over the last years, interest in hybrid metaheuristics has risen considerably among re- searchers. The best results found for many practical or academic optimization problems are obtained by hybrid algorithms. Combination of algorithms such as descent local search (15), simulated annealing (10), tabu search (6) and evolutionary algorithms have provided very powerful search algorithms. Two competing goals govern the design of a metaheuristic (19): exploration and exploitation. Exploration is needed to ensure every part of the search space is searched thoroughly in order to provide a reliable estimate of the global optimum. Exploitation is important since the refinement of the current solution will often produce a better solution. Population-based heuristics (where genetic algorithms (9) and estimation of distribution algorithms (12) are found) are powerful in the exploration of the search space, and weak in the exploitation of the solutions found. With the development of our new approach, GA-EDA, a hybrid algorithm based on genetic algorithms (GAs) and estimation of distribution algorithms (EDAs), we aim to improve the explorations power of both techniques. This hybrid algorithm has been tested on combinatorial optimization problems (with discrete variables) as well as real-valued variable problems. Results of several experi- ments show that the combination of these algorithms is extremely promising and com- petitive.