We demonstrate how surprise and affirmationmotives naturally emerge from the utility-maximizing behavior of agents. We fully characterize thedynamics of the news spread and uncover the mechanisms that lead to a sharing cascade. We furtherinvestigate the impact of the network structure, heterogeneity of priors, and precision levels of news onthe ex-ante probability of the news going viral. In particular, we show that as individual perspectivesbecome more diverse, a wider range of news precision levels cause a cascade. Finally, we elucidatean association between the news precision levels that maximize the probability of a cascade and theprior wisdom of the crowd. Our results complement the empirical findings that support wider spreadof inaccurate/false news compared to accurate information on social networks, providing a theoreticalmicro-foundation for utility-based news-sharing decisions.