In a cooperative network, users share information with each other to achieve a common target. Due to the concerns of privacy and cost, users may be reluctant to share genuine information with each other, which incurs the information trustworthiness problem. Most of the existing research attempts have proposed various mechanisms targeting to enhance the information credibility in various scenarios. However, the users' information sharing inclinations have not been well considered and modeled, which closely determine the information credibility in the network. In this paper, we study a trustworthy situation in cooperative networks and utilize the concept of reputation to model users' behaviors. Specifically, we propose two reputation learning methods based on both the peer-to-peer Bayesian learning and the social non-Bayesian learning models, and also propose a trustworthiness learning method based on the posterior estimation. Finally, simulations are conducted to verify the correctness and effectiveness of our theoretical analysis.