Interactive highlighting is a common component of many techniques used in visualization such as brushing and dynamic queries. Sometimes these may be used in combination necessitating that two different highlighting methods be simultaneously applied. The challenge of effective highlighting is to design methods that make a subset of the items on a display stand out clearly without overly interfering with other information on a display. This is especially difficult when more than one subset of displayed symbols must be simultaneously highlighted. Three experiments are reported that investigate four different highlighting methods: 3D vs 2D symbols, encirclement, oscillatory motion and blinking. These are applied to the nodes in node-link diagrams. The first experiment was designed to evaluate the highlighting methods used separately and the results showed all four techniques to be effective. The second experiment evaluated combinations of highlighting methods. E.g. can we easily find a node that is both moving AND 3D in a set of nodes some of which are 3D and some of which are moving. The results showed that combinations including motion were the most effective. The third experiment was designed to determine which highlighting methods, used both separately and in combination supported the rapid counting of small numbers of targets. Again, combinations using motion were the most effective.