The Personal Equation of Interaction for Categorization of Composite Glyphs

In this paper we advance the "Personal Equation of Interaction", examining individual differences between users that can be used to predict the accuracy of analysts' reasoning during visual analysis. We report 2 studies which expand the use of the Personal Equation of Interaction (PEI) beyond its current scope of predicting the accuracy of interface learning tasks to predicting the outcome of visio-cognitive object categorization. These studies extend the research of Yamauchi & Markman [3] using the dual learning theory of Ashby & Maddox [17]. Because visual reasoning is ubiquitous to the human experience and integral to big data visualization analysis, this research bridges the domain divide between psychological categorization theory and visual data analysis using composite glyphs. We define a psychometric measure that predicts the accuracy of composite glyph categorization and discuss its impact going forward.