Effects of frequency sorting towards finding optimal organizations of hierarchal file structures
Browsing through a hierarchy of information is a common task, but the relationship between human performance and parameters of the hierarchy are still incompletely understood. Based on Cockburn, Gutwin, & Greenberg (2007), we developed a model of this task involving two components at each level of the hierarchy: visual search time and motor movement time. A Monte Carlo simulation of the model shows how the optimal breadth of a hierarchal structure depends on both the size of the structure and how efficiently users perform the visual search. We fit a function to the Monte Carlo data which produced an optimum breadth consistent with the literature, and also shows analytically how properties like sorting can affect performance. The model indicates that the optimal hierarchy breadth depends heavily on the efficiency of the users' visual search; if the search can be sub-linear, then very broad hierarchies can be highly efficient.
