The interaction between human beings and robotic agents, and the interest towards such topics, have been exponentially growing in the recent years. The purpose of this thesis project is to identify a relation between the behaviours of a humanoid robot placed in a social context, and the emotional responses of a subject interacting with it. In particular, through the use of Brain-Computer Interface (BCI) and gaze tracking technologies, it has been investigated on the relation between the trust towards a robotic agent and the effects it has on the brain signals. In order to evaluate this relation, the framework makes use of the acquired brain signals to extract biometric features, such as attention, stress, and mental workload, along with the visual focus. In order to investigate towards this direction, an interactive game session has been set up for the human-robot interaction. In particular, an instance of the well-known Rock-Paper-Scissors game has been used. The experimental results have been shown a correlation between the behaviours of a robotic agent and the effect of trust on the brain signals of the human user. In particular, the emotional response varies depending the type of behaviours expressed by the robotic agent.