This master’s thesis studies a nonlinear multistate voter model with imperfect copying as a model for flocking. The system is first analysed on a complete graph, where a mean-field description and linear stability analysis are developed. Numerical simulations confirm the theoretical prediction for the order-disorder transition, and finite-size analysis shows critical behaviour compatible with a continuous mean-field transition. The model is then extended by restricting the field of view of the particles. For narrow viewing angles, persistent angular fragmentation appears at low noise, while moderate copying noise there is global alignment. Finally, moving particles with both angular and distance restrictions are considered. Particle motion suppresses persistent angular fragmentation but generates spatial organization, including large-scale density variations and dense local groups. Overall, the results show that collective order emerges from the interplay between nonlinear copying, noise and interaction geometry.