This paper introduces a modified algorithm to partical swarm based optimization that significantly reduces the number of iterations required to reach good solutions and discusses the results of experimentally comparing its performance with the performance of several variants of the standard particle swarm optimizer. The variants of the modified algorithm and the most common variants of the particle swarm optimizers are tested using a set of multimodal functions commonly used as benchmark optimization problems in evolutionary computation. Results indicate that the algorithm is highly competitive and can be considered as aviable alternative to solve the optimization problems.