Depression is associated with poor interpersonal and social functions. To improve the social dysfunction is the key component of successful intervention in depressed patients. Therefore, the review of this topic can further our understanding and promotes potential clinical practice. To understand depressed patients’ interpersonal and social dysfunction in the context of social decision-making, an advantage is that researchers can use game theory paradigms from neuroeconomics to examine interpersonal and social dysfunction in patients with depression. Ruff and Fehr (2014) proposed three classes of situations in social decision-making. First, it refers to assess how other individuals' valuation or feedback affect the agent’s own well-being (“social feedback” for short). This class of social decision-making includes situations in which the agent experiences social rewards or punishments from oth?ers, such as social acceptance or social rejection. The second class concerns situations in which an individual vicariously values choice options and outcomes for another individual, i.e., vicarious experience of social rewards and punishment, vicarious social decision-making, and vicarious social learning (“vicarious valuation” for short). The third class comprises situations in which an individual values his or her behavior in a reference frame of normative social principles (“social principle” for short). Based on the three classes of aforementioned situations, the current review summarized recent behavioral and fMRI studies on social decision-making in healthy adults as well as in patients with depression. For the situation of "social feedback", depressed individuals showed social anhedonia for positive social interaction and social reward, which may due to reduced response from the nucleus accumbens. In contrast, patients had increased sensitivity to negative social interactions, such as social rejection or exclusion, which is associated with increased activation in the neural network of social pain. For the situation of "vicarious valuation", although we found no direct deficits in depression, depressed individuals showed deficits in empathy (which is relevant to vicarious reward and punishment) and theory of mind (which is relevant to vicarious decision-making and vicarious learning). For the situation of "social principle", when depressed individuals guided their behavior to comply with normative social principles (e.g. fairness, cooperation, altruism), these individuals demonstrated maladaptive behaviors, such as hyper-altruism and reduced cooperative behavior. According to the previous results reviewed above, this paper proposed three unsolved problems and associated possible solutions. First, although depressed individuals showed multifaceted anhedonia, such as reduced rewarding experience of desire, anticipation, motivation, effort and pleasure, most of current studies used money as incentives. We argued that a shift from monetary to social stimuli is needed in clinical research to improve our understanding of social anhedonia in depressed individuals. Second, it is not clear whether the observed neural responses are causally necessary for social decision-mak?ing. The combination of non-invasive brain stimulation tools, such as repetitive transcranial magnetic stimulation (rTMS), and brain imaging methods can resolve this issue. In this context, researchers can investigate the causal effect of rTMS on the behavioral changes in the stimulated brain regions. Third, the hyperscanning technique which has a high ecological validity is suggested to be used in the future social decision-making studies; because it allows the inter–human experimental designs and can simultaneously record neural activities of interactive subjects.