This paper aims to explore the analytical issues surrounding the graphical representation of causation in accident scenarios. Two types of graphical renderings, Petri Nets and Why-Because Graphs, are introduced and the different ways in which they denote accident causation are explored. The second part reports an experiment undertaken to explore whether graphical renderings of causality have any benefits over text, and if one of the two graphs is superior to the other. In so doing, this paper aims to draw out our visual understanding of graphical renderings and how these map from the original representation, the accident scenario, to the secondary representation.