The work presented in this thesis collects three fMRI studies mainly focusing on empathy, i.e. the capacity to understand and/or share the emotional state of others. Empathy is central to human sociality, as it allows us to resonate with others’ positive and negative feeling, and consequently adjust our behavior. Despite recent research has shed light on many feature of empathic responses, we still ignore many other aspects: for instance, which kind of computational processes are executed by empathy ́s neural substrates, how empathic responses vary according to the type of observed experience, which neurochemical mechanisms are at the core of empathic responses, or also what is the link between empathic responses and the tendency to behave altruistically (usually referred to as ‘prosocial behavior’). The purpose of the work presented in this thesis is providing answers to some of the open questions. In Study 1 we aimed at understanding what are the neural substrates of empathy for social pain, a kind of pain that is constantly grabbing increasingly attention among social neuroscientists, and to which extent they overlap with the ones coding for physical pain. In Study 2 we investigated brain correlates of prosocial behavior by exploring functional connectivity within brain networks of participants who exhibited either a self-benefit behavior or an altruistic one in a life-threatening situation simulated in a virtual environment. In Study 3 we used a placebo manipulation on a group of participants undergoing first- hand and vicarious painful stimulations in order to observe how the supposed enhancement of endogenous opioids release would affect their behavioral and neurophysiological responses to the painful experience. Overall, the work presented in this thesis advances the knowledge on both empathy and prosociality mechanisms and opens the way for new investigations aiming at clarifying key aspects of social behavior.