Socio-cognitive language processing for special user groups

Sociocognitive Language Processing (SCLP) is the idea of coping with everyday language, including slang and multi-lingual phrases and cultural aspects, and in particular, with irony/sarcasm/humour, and paralinguistic information such as the physical and mental state and traits of the dialogue partner (e.g., affect, age groups, personality dimensions), and social aspects. Additionally, multimodal aspects such as facial expression, gestures or bodily behaviour should ideally be included in the analysis wherever possible. At the same time, SCLP can render future dialogue systems more 'chatty' by not only appearing natural, but also by being truly emotionally and socially competent, ideally leading to a more symmetrical dialogue. To do this, the computer should itself have a 'need for humour', an 'increase of familiarity', etc., i.e., enabling computers to experience or at least better understand emotions and personality so that that they have 'a feel' for these concepts. Beyond these ideas, the broader idea of SCLP includes verbal behaviour analysis, a closer coupling between language understanding and generation, incorporating social and affective information, and new language resources to meet these ends. In this way, SCLP unites expertise from psychology, social sciences, and (natural) language processing. Here, we give a short introduction.