Eye-Tracking to Predict User Cognitive Abilities and Performance for User-Adaptive Narrative Visualizations

Magazine-style narrative visualizations or MSNVs, are a very common type of information visualization that combine both text and graphic components. Processing MSNVs can be challenging due to the need to split attention between the two information sources, which can lead to an increase in cognitive load and negative impact on comprehension. One way to address user's challenges in comprehension is to provide adaptations that facilitate processing of the visualization. However, recent work has shown that adaptations are useful only for users with specific levels of cognitive abilities, and if these adaptations are provided when unwanted they can be intrusive and distracting.\r\nIn this presentation we present our approach at using eye tracking data to predict task performance measures as well as user's cognitive abilities, which would help us identify when and to whom provide these adaptations, informing the design of personalized-user adaptive MSNVs.\r\n