In this dissertation, a computational paradigm is proposed that adopts the particle swarm optimization within a cultural framework. The cultural algorithm adopted here consists of two space, population space and belief space including five sections, situational knowledge, normative knowledge, topographical knowledge, history knowledge, and domain knowledge. Several innovative paradigms have been proposed including diversity-based particle swarm optimization, cultural-based multiobjective particle swarm optimization, cultural-based constrained particle swarm optimization, and cultural-based dynamic particle swarm optimization. Comparative study between different proposed algorithm and the state-of-the-art algorithms in the corresponding field, demonstrated the potential existing in cultural-based particle swarm optimization that resulted in competitive results verifying the effectiveness and efficiency of the proposed cultural particle swarm optimization to solve different types of optimization problem from single objective optimization problems, multiobjective optimization problems, constrained optimization problems, and dynamic optimization problems.