In this paper, the problem of forming opinions (through aggregation) is considered, wherein the initial opinions are generated by a set of independent individuals (or agents) and then these individuals share this information to form aggregated opinions simultaneously. This problem is challenging due to its decentralized nature and also agents won’t reveal their private opinions, but substitutes to their private opinions. To the best of our knowledge, decentralized nature of opinion formation (as information aggregation) problem is not considered in the literature and this setting arises in several real-life scenarios such as candidate shortlisting. This paper addresses the decentralized information aggregation problem by formulating it as a game, wherein the payoff of any agent depends on its own information (e.g., rank order) as well as other agents’ information (e.g., others’ rank orders). Then an efficient method is proposed for the best response dynamics of this game leading to an equilibrium that corresponds to information aggregation. It is observed that the disagreement among the rank orders of agents at equilibrium is significantly less compared to that of the initial information (rank orders). If there is no consensus among the equilibrium rank orders, then one can resort to any existing centralized algorithms to aggregate the rank orders at equilibrium. Using both synthetic and real Web search data, empirically it is shown that the proposed approach significantly outperforms several well-known centralized information aggregation algorithms.