Particle Swarm Optimization (PSO) algorithm has shown good performance in many optimization problems, but PSO suffers from the problem of early convergence into a local minima. Introduction of opposition based initialization and mutation operators have played an important role to overcome the convergence problem in function optimization. In this study we have reviewed different variants of PSO for function optimization. Researchers have proposed different modifications in PSO to prevent it from getting stuck in local optima. At the end, we have proposed a variant of PSO for better conversion.