The computational model plays a vital role in the design and optimization of microelectrodes for efficient electrical stimulation and recording in the retinal prosthesis. Moreover, the material choice acts decisively in ensuring that the electronic device delivers sufficient stimulating current to the retina without harming the neighboring tissue. Recently, due to the enhanced electrical and electrochemical properties of graphene, it has become a viable material in biomedical applications. In this study, we analyzed the computational model for the retinal prosthesis by the novel use of graphene-based microelectrodes. For this, different topologies of the electrode arrangement were investigated. The most suitable configuration involves the arrangement of electrodes serving as the ground in a hexagonal fashion around the central stimulating electrode. We observed that the performance of graphene as the stimulating electrodes is comparable to the existing noble metal-based electrodes. Moreover, we found that optimizing the microelectrode design resulted in uniform electric potential distribution, and this eventually led to an increased electric field intensity at the desired activation point. Additionally, we analyzed the crosstalk phenomenon, and we observed from the results that it is better to implant such an electrode array in the vicinity of the targeted volume to minimize the effect of crosstalk.Clinical relevance- The present study can help in the improvement of the retinal prosthesis by analyzing the quantitative and qualitative effects of graphene-based microelectrodes on producing the threshold electric field in the retina tissue. The results obtained can be used to optimize the implantable microelectrode design and thus is a step forward in finding a cure to vision impairment diseases.