A well-known problem faced by Multi-Objective Particle Swarm Optimization Algorithms (MOPSO) is the deterioration of its search ability when the number of objectives scales up. In the literature some techniques were proposed to overcome these limitations, however, most of them focuses on alternatives to the non-domination relation. In this work, a different direction is explored, and some specific aspects of MOPSO as the selection of the leaders to guide the search are investigated. The work presents a comparison of several approaches of leader selection to find which of them presents the better results in terms of convergence and diversity in many-objective scenarios. Also, a new method, called Opposite method, is proposed. The results are analyzed through different quality indicators and statistical tests.