Research on the interaction between emotion, cognition and language in the field of Artificial Intelligence has become particularly active along the last years. Lots of computational models of emotion have been developed. There are accounts stressing the role of canonical and mirror neurons as underlying the use of nouns and verbs. At the same time, neuropsychology is developing new approaches for modeling language, emotion and cognition inspired on the insights gained from robotics. The current landscape is thus a promising collaboration between several approaches: Social Psychology, Neuropsychology, Artificial Intelligence (mainly embodied), and even Philosophy, so that each field provides useful cues for the common goal of understanding social interactions (including the interactions with machines).The aim of this paper is to analyze and asses the current trends in psychology and neuroscience for studying the mechanisms of the neurocomputational cognitive-affective architecture related to the conceptualization and use of language.