Testing the hypothesis required separating two things that look alike from a distance: concealment among flowers, and attraction in their absence. Researchers presented mantises on plain green vegetation, away from any blossom, and counted insect approaches against those made to nearby real flowers.
Pollinators visited the mantis at least as often, and in some trials more often, than the flowers. Spectrometry showed the mantis body reflects light in a way bees and flies perceive as flower-like, including in the ultraviolet, and the lobed legs break the outline into petal-shaped units. The animal is not blending into a floral background; it is functioning as a false flower and being approached as one. This is aggressive mimicry, in which a predator advertises to draw prey in.
Two caveats hold. The effect has been measured in females, whose size and lobed legs are most pronounced, and males are much smaller and less flower-like. Whether the mimicry targets any specific plant is not supported: no particular orchid model has been identified.