Due to the limited brain capacity to process information, anticipatory mechanisms unfold to predict others’ behavior and adapt one’s own responses. To do this, the social brain follows cues like movements, facial expressions, and gaze direction to predict others’ mental states. With the rise of artificial intelligence and the imminent introduction of artificial agents into our society, robots will evolve from mere tools to become teammates or even companions. In order to understand how this might impact social interactions, we must understand how social brain networks respond to these technologies in real-life interactions. Although it is reasonable to assume that interacting with robots with human-like features (e.g., behavior, appearance) might elicit activation of these networks similar to human-human interactions, current research has not directly answered this query. In the current project, we aim to identify neural markers of the activation of social cognitive mechanisms with artificial agents using supervised machine learning (ML) models based on brain activity measured with functional near-infrared spectroscopy (fNIRS). Combining neuroscience methods with ML techniques has the potential to reveal whether interactions with robots distinctively evoke social-cognitive mechanisms (e.g., related to action-perception, mentalizing, face perception) or general cognitive mechanisms (e.g., attention, cognitive control). The goals of the project are to (i) gain an understanding of the social-cognitive mechanisms involved in human-robot interaction, (ii) characterize the brain circuitry engaged in human-robot interactions, and (iii) define the presence or absence of neural markers that reflect the recruitment of social cognitive mechanisms.