In this paper, a strategy for multi-objective optimization based upon the behavior of a particle swarm with rotational and linear motion is presented. The strategy for multi-objective optimization is based upon the emulation of the linear and circular movements of a swarm (flock). Thus emerges the physical basis for the cognitive model, which in conjunction with exploration-exploitation results in the proposal of a cognitive algorithm, which is tested through several multi-objective optimization functions. The algorithm proposed is compared with standard particle swarm optimization multi-objective via statistical analysis.