PhD Defense: The Next Generation of Implicit Brain-Computer Interfaces: Enhancing User Learning and Creativity.
There is a fundamentally limited bandwidth of implicit and explicit communication between the human and computer. However, both humans and computers are complex machines, capable of sophisticated functions. If we could provide the computer with more implicit information about the human, without any additional effort on the part of the user, the computer could then respond more intelligently in return.
In this thesis, I expand upon the next generation of implicit, brain-computer interfaces by building two, real-time adaptive brain-computer interfaces (BCIs) based on users' cognitive workload in the previously unexplored fields of learning and creativity. I demonstrate that users can learn with increased speed and accuracy with a BCI that guides learners' progress by measuring when they can cognitively handle more information and providing them with the next stage of learning at the right moment. I also build and evaluate an adaptive BCI that addresses the no-less tricky task of increasing user creativity in a musical task. Finally, I address the topic of whether users will actually trust such intelligent systems by building a trust evaluation system that reveals that building trust in human-computer interaction can be just as wiley as human-human interaction.
I suggest that measuring affective state in conjunction with cognitive state in the future would provide more accurate insight into the user's state, allowing for more personalized and intelligent adaptations by the next generation of implicit, brain- computer interfaces.