This study examines the effect of specific factors (including user features, such as propensity for conspiracy thinking, and news item features, such as news frame and news source) on the accuracy of social media users in fake news recognition. Being a part of a larger research on fake news perception, this study uses the data from an online experiment that asks social media users from three countries (Russia, Ukraine and Kazakhstan) to evaluate a set of news items constructed with specific conditions. Namely, the users receive true and fake news about the neighboring countries framed differently and ascribed to either domestic or foreign sources. We then assess users' accuracy in detecting fake news. The results of the study confirm the important role of conspiracy thinking style in false news recognition (leading to a decrease in accuracy) and users' capability for deliberation on social media more broadly. However, the influence of contextual factors is mixed. While news sources exhibit no influence on the accuracy of fake or true news detection, dominant framing tends to increase the accuracy of true news only. More predictors of news recognition accuracy are discussed in the paper. As a result, this research contributes to the theory of fake news susceptibility by revealing a rich set of individual factors and interaction effects that influence human judgment about news truthfulness and impact deliberation possibilities in socially mediated environments.