Social decisions are among the most important choices in our life. They are often proposed to rely on functionally specialized neural circuitry, based on correlated neural activity observed with neuroimaging. However, neuroimaging studies usually do not allow conclusions about whether the identified neural activity causally controls behavior, rather than being a consequence of it. This gap is now being bridged by brain stimulation studies that test the causal relationship between neural activity and three different types of processes underlying social decisions: social emotions, social cognition, and social behavioral control. Here we critically review this evidence and propose future steps that may help to advance our understanding of how the brain implements social decisions.