FRIENDs: Brain-monitoring agents for adaptive socio-technical systems

Alexis Morris, Mihaela Ulieru
2013 Multiagent and Grid Systems  
Brain-monitoring is quickly becoming an important field of research, with potentially significant impacts on how people will interact with technology. As understandings of the inner-workings of the brain become more accurate technologies are becoming more advanced, smaller, cheaper, and ubiquitous. It is expected that new forms of computing that take advantage of brain states will be developed. This will enable systems to be highly aware of user mental contexts (emotions, intentions, and
more » ... These systems would display higher autonomic behavior and would streamline user-interaction while managing the use of brain context data for applications and services. There are few studies of how to develop and make use of agent architectures in this new domain. Current approaches target a single user and application situation. To be ubiquitous it is unrealistic for applications to have specialized overhead for individual users. Personalizable, but distributed approaches are needed. To realize a general purpose agent for brain-monitoring and management of brain context is the goal of this work. This involves the selection of a brain-monitoring paradigm, the selection of an agent architecture paradigm, an inferencing mechanism, and the combination of the three towards a unified framework. Core motivations are discussed, and an early agent framework design (FRIEND) is presented, along with proposed proof-of-concept applications for using brain context. A. Morris and M. Ulieru / FRIENDs: Brain-monitoring agents for adaptive socio-technical systems Making sense of brain context on the fly (states and patterns of activity) requires adequate measurement methods, inferencing techniques, and a control architecture that captures these signals and makes use of them. The measurement methods are available, as well as inferencing methods [7] , but robust control architectures that are flexible, fast, autonomous, and applicable in multiple scenarios remain to be explored. At present individual researchers are developing custom solutions for handling brain measurements [24, 55, 56] but no general-purpose architecture has emerged. This research aims to fill this gap by exploring architectures and developing a novel agent-oriented approach to brain-monitoring and brain context management systems. Motivation Socio-technical systems gaps This work aims to seamlessly bridge the gap between people and technology from a techno-centric viewpoint, through smarter, human-context-aware computing. Its purpose is to augment social systems with more dynamic and flexible technology. At a high level there are three key issues. First, there are significant failures due to the misfit between technology and social context (i.e., the socio-technical gap between users and technologies). Second is that although these gaps can be minimized, there is only so much that can be done by modifying social systems to fit better with technologies. Third, there is a real challenge in designing, developing, and testing technologies that minimize the socio-technical gap. Failures due to socio-technical disharmony have been discussed by Vicente [75], as stemming from five core layers (physical, psychological, social-team, organizational, and political), and range from being simply annoying to catastrophic. These result from flaws in both human social systems and supporting technologies, in particular computing technologies. On the human side there are limits, task overloads, attention overloads, and mismatched mental models [39] , each contributing to potential failure points wherein people make mistakes due to excess demands, improper understanding of system states, tasks, and situations. On the technological side these problems are compounded by interface complexity and inefficiency [10], tasks out of context with situations, and the sheer increase of speed, data volume, and ubiquity. Socio-technical gaps can be minimized through development of either smarter social systems, or smarter technology. Since social systems are limited in terms of human cognitive and physical capacities (such as stress, situational, or relational limits), it is realistic to focus on the development of smarter technologies to bridge the gap. Technologies like this would perform tasks related to monitoring human limits, predicting the state of the social (human) system, protecting the social system through useful interventions/actions, and assisting the social system through anticipatory actions where possible. There is a need for software systems, as technology progresses, to understand human social systems, which are not well defined, yet highly dynamic, in many contexts and multiple environments, involved in potentially large complex and also multi-dimensional social networks (see [47, 75] for more on dimensionality of social systems). In order to succeed in this task these systems must themselves be highly dynamic, autonomic (self-*, [36]), and context-sensitive (especially for human-context, device-context, and situation-contexts). In this work these capabilities are considered as "human-awareness". Brain cognitive context awareness Developing high human-awareness systems is a big challenge, and remains a novel problem related to the overarching goal of improving the socio-technical gap. Human-aware systems must understand and
doi:10.3233/mgs-120198 fatcat:7a3zclxz2rbnrgo362d6g7lddq