Information presentation in multimodal interfaces

Oviatt (2003), in her recent state-of-the-art overview of multimodal interfaces, notes that current multimodal interfaces tend to be unbalanced: they use sophisticated multimodal input, and pay little attention to output. She considers the development of advanced output modules one of the main future directions for multimodal interfaces. The current project focuses on the development of such modules, combining language, speech and computer graphics. They will be based on empirical findings (how do humans do it?) and evaluated experimentally (how do humans interpret and evaluate the produced multimodal output?)