2011
| Action Number: 2102
Towards Autonomous, Adaptive, and Context-Aware Multimodal Interfaces. Theoretical and Practical Issues
This volume brings together, through a peer-revision process, the advanced research results obtained by the European COST Action 2102: “Cross Modal Analysis of Verbal and Nonverbal Communication”. The research published in this book was discussed for the first time at the 3rd jointly EUCOGII-COST 2102 International Training School on “Towards Autonomous, Adaptive, and Context-Aware Multimodal Interfaces: Theoretical and Practical Issues ” held in Caserta, Italy, March 15-19, 2010.
The school afforded a change of perspective in verbal and nonverbal communication, where the research focus moved from “communicative tools” to “communicative instances” and asked for investigations that, in modeling interaction, will take into account not only the verbal and nonverbal signals but also the internal and external environment, the context, and the cultural specificity in which communicative acts take place.
The consequences, in Information Communication Technology (ICT) research, should result in the development of autonomous, adaptive, and context-aware multimodal interfaces able to act by exploiting instantiated contextual and environmental signals and process them by combining previous experience (memory) adapted to the communicative instance. This new approach will foster artificial cognitive research by creating a bridge between the most recent research in multimodal communication (taking into account gestures, emotions, social signal processing etc) and computation models that exploit these signals being aware of the context in which these signals are instantiated and of the internal and external environmental background. Human behavior exploits this information and adapts. Artificial cognitive systems must account of this human ability for implementing a friendly and emotionally colored human machine interaction. In order to do this, investigations in cognitive computation must move from purely data driven systems to behavioral systems able to “interact with human to achieve their goals” (see the first chapter of this book by V.C. Müller), i.e. able to manifest intentions and goals through “resistance” to other intentions and goals. In summary, cognitive models of “resistance” must be developed such that the current interactive dialogue systems, robots, and intelligent virtual avatars graphically embodied in a 2D and/or 3D interactive virtual world, are able to interact intelligently with the environment, other avatars, and particularly with human users.
The themes of the papers presented in this book emphasize theoretical and practical issues for modelling cognitive behavioural systems, ranging from the attempts to describe Brain Computer Interface (BCI) applications, context-based approach to the interpretation and generation of dialogue acts, close synchronization among both speakers and listeners, mutual ratification, interaction and resistance, embodiment, language and multimodal cognition, timing effects on perception, action, and behaviours.