Many animal species use stereotyped displays to attract the opposite sex and to intimidate same-sexed rivals. Research aimed at understanding display recognition, function, and usage can be aided through the use of animal robots that allow one side of signaler–receiver interactions to be controlled. Manipulation of displays in ways that do not occur in nature has the potential to determine the boundaries of display recognition, as well as to provide insights into the manner in which animal display contests are structured. We describe two experiments that extend previous work on display recognition in the lizard Anolis grahami. In the first experiment, we used robots to determine the relative importance of body coloration and headbob display structure for species recognition. The results showed that subjects responded more strongly to robots having both conspecific appearance and display structure than to robots that deviated in those characteristics from the conspecific stimulus. In the second experiment, we explored the effect of removing display components on subjects’ responses, where subjects witnessed a conspecific robot exhibiting a typical display (headbobs followed by dewlap pulses), or a deficient display consisting only of headbobs or dewlap pulses. Contrary to expectation, subjects in the headbobs-only treatment spent more time headbobbing than dewlapping in response; whereas, those in the dewlap-only treatment spent more time dewlapping than headbobbing. Interactive robots could be used in future investigations of the functions of different display components, as well as to examine the rules by which lizard display contests are conducted in nature.