In the analysis of eye movements, a lofty goal of many researchers is to deduce from fixation locations what a person is paying attention to at any given moment. Traditionally, the object or region of interest (ROI) that is closest to each fixation is assumed to be the focus of attention. When this process is repeated over a sequence of fixations the resulting pattern is often referred to as a scanpath. Algorithms for calculating scanpaths typically require that ROIs be simple geometric shapes (like circles or rectangles). This restriction has not been much of an issue historically as most tasks used in eye movement research involve static displays of relatively simple stimuli. However, eye movement research in complex tasks (like video games) is becoming ever more popular and traditional scanpath algorithms have no way to handle dynamic ROIs that can move, overlap with one another, change size, change shape or even disappear completely. Below is a link to the slides of talk I presented at the European Conference on Eye Movements (ECEM) 2015 which discusses many of the challenges associated with calculating scanpaths in dynamic environments. The talk also introduces a novel scanpath algorithm developed at the CogWorks Lab specifically designed to address the issues discussed.