Augmented Cognition:  New Paradigm for Future Interfaces

                          Prof. Misha Pavel

 Department of Biomedical Engineering Oregon Health & Science University Portland


Abstract

Augmenting human cognition is a novel approach to the design of future 
systems and devices that are expected to significantly enhance human 
cognitive abilities. The underlying idea is that the augmented 
cognition system would infer the context of the task, assess the 
cognitive state of the operator, and then optimize the input and output 
information requirements, as well as cognitive resource allocation. In 
this manner, it would be theoretically possible to mitigate information 
processing bottlenecks in human cognitive processes.  The success of 
this effort depends on our ability to develop sufficiently accurate 
models of human cognitive processes and of the reliability in assessing 
cognitive states of the operator. In this presentation, I will provide 
an overview of human cognitive limitations and introduce an approach to 
modeling information bottlenecks. I will then describe our preliminary 
work on developing novel state assessment and classification algorithms 
based on maximizing mutual information between EEG signals and the 
cognitive states of the operator. I will describe our pilot experiments 
and early results using these techniques, noting both successes and 
failures. Finally, I will give a couple of examples of mitigation 
strategies. One of the main points of this talk is to illustrate the 
importance of pattern classification and inference in these efforts