From repulsion to attraction: species- and spatial context-dependent threat sensitive response of the spider mite Tetranychus urticae to predatory mite cues.

Prey perceiving predation risk commonly change their behavior to avoid predation. However, antipredator strategies are costly. Therefore, according to the threat-sensitive predator avoidance hypothesis, prey should match the intensity of their antipredator behaviors to the degree of threat, which may depend on the predator species and the spatial context. We assessed threat sensitivity of the two-spotted spider mite, Tetranychus urticae, to the cues of three predatory mites, Phytoseiulus persimilis, Neoseiulus californicus, and Amblyseius andersoni, posing different degrees of risk in two spatial contexts. We first conducted a no-choice test measuring oviposition and activity of T. urticae exposed to chemical traces of predators or traces plus predator eggs. Then, we tested the site preference of T. urticae in choice tests, using artificial cages and leaves. In the no-choice test, T. urticae deposited their first egg later in the presence of cues of P. persimilis than of the other two predators and cue absence, indicating interspecific threat-sensitivity. T. urticae laid also fewer eggs in the presence of cues of P. persimilis and A. andersoni than of N. californicus and cue absence. In the artificial cage test, the spider mites preferred the site with predator traces, whereas in the leaf test, they preferentially resided on leaves without traces. We argue that in a nonplant environment, chemical predator traces do not indicate a risk for T. urticae, and instead, these traces function as indirect habitat cues. The spider mites were attracted to these cues because they associated them with the existence of a nearby host plant.