Object categorization by wild-ranging birds in nest defence.

Despite object categorization being an important ability for the survival of wild animals, the principles behind this ability have been only scarcely studied using wild-ranging, untrained animals. Reiterating our previous study undertaken with wild-ranging titmice on winter feeders (Nováková et al. Behav Process 143:7-12, 2017), we aimed to test two hypotheses of object recognition proposed by animal psychology studies: the particulate feature theory and recognition by components in the methodological paradigm of nest defence. We tested whether the parents of the red-backed shrike (Lanius collurio) recognize the dummies of the common kestrel (Falco tinnunculus), which is a potential predator of large chicks or fledglings, as a threat in case when their body parts are scrambled. The kestrel dummy was presented with the head at the top, in the middle, and at the bottom of the body. We showed that the shrikes did not consider dummies of a kestrel with an inappropriately placed head as a threat to the nest and attacked it equally scarcely as the harmless control. These results support the theory of recognition by components, presuming that the mutual position of body parts is essential for appropriate recognition of the object. When the body parts were scrambled, most of shrikes were not able to identify the kestrel in such an object despite all local features (eye, beak, colouration, and claws) being present. Nevertheless, shrikes did not consider the scrambled dummies as completely harmless, because they fed their chicks in their presence significantly less often than in the presence of harmless control.

Leaf volatile and nonvolatile metabolites show different levels of specificity in response to herbivory.

Plants produce diverse chemical defenses with contrasting effects on different insect herbivores. Deploying herbivore-specific responses can help plants increase their defensive efficiency. Here, we explore how variation in induced plant responses correlates with herbivore species, order, feeding guild, and level of specialization. In a greenhouse experiment, we exposed 149 plants of Salix fragilis (Linnaeus, 1753) to 22 herbivore species naturally associated with this host. The insects belonged to four orders (Coleoptera, Lepidoptera, Hemiptera, and Hymenoptera), three feeding guilds (external leaf-chewers, leaf-tying chewers, and sap-sucking), and included both dietary specialists and generalists. Following herbivory, we quantified induced changes in volatiles and nonvolatile leaf metabolites. We performed multivariate analyses to assess the correlation between herbivore order, feeding guild, dietary specialization, chewing damage by herbivores, and induced responses. The volatile composition was best explained by chewing damage and insect order, with Coleoptera and Lepidoptera eliciting significantly different responses. Furthermore, we recorded significant differences in elicited volatiles among some species within the two orders. Variation in nonvolatile leaf metabolites was mainly explained by the presence of insects, as plants exposed to herbivores showed significantly different metabolites from controls. Herbivore order also played a role to some extent, with beetles eliciting different responses than other herbivores. The induction of volatile and nonvolatile leaf metabolites shows different levels of specificity. The specificity in volatiles could potentially serve as an important cue to specialized predators or parasitoids, increasing the efficacy of volatiles as indirect defenses. By contrast, the induction of nonvolatile leaf metabolites was largely unaffected by herbivore identity. Most nonvolatile metabolites were downregulated, possibly indicating that plants redirected their resources from leaves in response to herbivory. Our results demonstrate how diverse responses to herbivores can contribute to the diversity of plant defensive strategies.

Object categorization by wild ranging birds-Winter feeder experiments.

The object categorization is only scarcely studied using untrained wild ranging animals and relevant stimuli. We tested the importance of the spatial position of features salient for categorization of a predator using wild ranging birds (titmice) visiting a winter feeder. As a relevant stimulus we used a dummy of a raptor, the European sparrowhawk (Accipiter nisus), placed at the feeding location. This dummy was designed to be dismantled into three parts and rearranged with the head in the correct position, in the middle or at the bottom of the dummy. When the birds had the option of visiting an alternative feeder with a dummy pigeon, they preferred this option to visiting the feeder with the dummy sparrowhawk with the head in any of the three positions. When the birds had the option of visiting an alternative feeder with an un-rearranged dummy sparrowhawk, they visited both feeders equally often, and very scarcely. This suggests that the titmice considered all of the sparrowhawk modifications as being dangerous, and equally dangerous as the un-rearranged sparrowhawk. The position of the head was not the most important cue for categorization. The presence of the key features was probably sufficient for categorization, and their mutual spatial position was of lower importance.