The recent surge of interest in advanced multipolar stimulation strategies for cochlear implants has led to a marked increase in the demand for computational tools capable of exploring increasingly complex electrode configurations [1]. Recent research has introduced a state-of-the-art modelling framework that integrates high-fidelity finite element simulations with a hybrid multi-objective evolutionary algorithm [2]. This enables the discovery of Pareto optimal stimulation patterns that jointly reduce power consumption while enhancing spatial focusing. However, the formulation of the focusing objective remains a critical bottleneck. The sensitivity of this objective to the chosen error norm and weighting scheme can substantially influence the resulting optimization landscape and, consequently, the quality of the electrode designs obtained. The present study systematically investigates alternative mathematical formulations for the focusing objective. The methodology is demonstrated across multiple electrode configurations in a representative cochlear implant test case, with results contextualised against the latest findings in computational neurostimulation research. REFERENCES [1] Hernández-Gil, M., Ramos-de-Miguel, A., Greiner, D., Benítez, D., Montero, G., Escobar, J.M.: A FEM-ANN framework to estimate the on-diagonal elements of the impedance matrix in a cochlear implant. Engineering Science and Technology, an International Journal, 73 (2026), 102273. https://doi.org/10.1016/j.jestch.2025.102273 [2] Hernández-Gil, M., Ramos-de-Miguel, A., Greiner, D., Benítez, D., Ramos-Macías, A., Escobar, J.M.: A computational model for multiobjective optimization of multipolar stimulation in cochlear implants. Expert Systems with Applications 280, 1-20 (2025), 127472. https://doi.org/10.1016/j.eswa.2025.127472