While collective decision-making is recognised as a significant contributor to fitness in social species, the opposite outcome is also logically possible. We show that collective movement decisions guided by individual bison sharing faulty information about habitat quality promoted the use of ecological traps. The frequent, but short-lived, associations of bison with different spatial knowledge led to a population-wide shift from avoidance to selection of agricultural patches over 9 years in and around Prince Albert National Park, Canada. Bison were more likely to travel to an agricultural patch for the first time by following conspecifics already familiar with agricultural patches. Annual adult mortality increased by 12% due to hunting of bison on agricultural lands. Maladaptive social behaviour accordingly was a major force that contributed to a ~50% population decline in less than a decade. In human-altered landscapes, social learning by group-living species can lead to fitness losses, particularly in fusion-fission societies.

When group members possess differing information about the environment, they may disagree on the best movement decision. Such conflicts result in group break-ups, and are therefore a fundamental driver of fusion-fission group dynamics. Yet, a paucity of empirical work hampers our understanding of how adaptive evolution has shaped plasticity in collective behaviours that promote and maintain fusion-fission dynamics. Using movement data from GPS-collared bison, we found that individuals constantly associated with other animals possessing different spatial knowledge, and both personal and conspecific information influenced an individual's patch choice decisions. During conflict situations, bison used group familiarity coupled with their knowledge of local foraging options and recently sampled resource quality when deciding to follow or leave a group - a tactic that led to energy-rewarding movements. Natural selection has shaped collective behaviours for coping with social conflicts and resource heterogeneity, which maintain fusion-fission dynamics and play an essential role in animal distribution.

In a world increasingly dominated by human demand for agricultural products, we need to understand wildlife's ability to survive in agricultural environments. We studied the interaction between humans and Javan slow lorises (Nycticebus javanicus) in Cipaganti, Java, Indonesia. After its introduction in 2013, chayote (Sechium edule), a gourd grown on bamboo lattice frames, became an important cash crop. To evaluate people's use of this crop and to measure the effect of this increase on slow loris behaviour, home ranges, and sleep sites, we conducted interviews with local farmers and analysed the above variables in relation to chayote expansion between 2011-2015. Interviews with farmers in 2011, 2013 and 2015 confirm the importance of chayote and of bamboo and slow lorises in their agricultural practises. In 2015 chayote frames covered 12% of land in Cipaganti, occupying 4% of slow loris home ranges, which marginally yet insignificantly increased in size with the increase in chayote. Slow lorises are arboreal and the bamboo frames increased connectivity within their ranges. Of the sleep sites we monitored from 2013-2016, 24 had disappeared, and 201 continued to be used by the slow lorises and processed by local people. The fast growth rate of bamboo, and the recognition of the value of bamboo by farmers, allow persistence of slow loris sleep sites. Overall introduction of chayote did not result in conflict between farmers and slow lorises, and once constructed the chayote bamboo frames proved to be beneficial for slow lorises.

Little attention has been given in scientific literature to how introduced species may act as a new host for native infectious agents and modify the epidemiology of a disease. In this study, we investigated whether an introduced species, the Siberian chipmunk (Tamias sibiricus barberi), was a potentially new reservoir host for Borrelia burgdorferi sensu lato, the causative agent of Lyme disease. First, we ascertained whether chipmunks were infected by all of the B. burgdorferi sensu lato genospecies associated with rodents and available in their source of infection, questing nymphs. Second, we determined whether the prevalence and diversity of B. burgdorferi sensu lato in chipmunks were similar to those of a native reservoir rodent, the bank vole (Myodes glareolus). Our research took place between 2006 and 2008 in a suburban French forest, where we trapped 335 chipmunks and 671 voles and collected 743 nymphs of ticks that were questing for hosts by dragging on the vegetation. We assayed for B. burgdorferi sensu lato with ear biopsy specimens taken from the rodents and in nymphs using PCR and restriction fragment length polymorphism (RFLP). Chipmunks were infected by the three Borrelia genospecies that were present in questing nymphs and that infect rodents (B. burgdorferi sensu stricto, B. afzelii, and B. garinii). In contrast, voles hosted only B. afzelii. Furthermore, chipmunks were more infected (35%) than voles (16%). These results may be explained by the higher exposure of chipmunks, because they harbor more ticks, or by their higher tolerance of other B. burgdorferi sensu lato genospecies than of B. afzelii. If chipmunks are competent reservoir hosts for B. burgdorferi sensu lato, they may spill back B. burgdorferi sensu lato to native communities and eventually may increase the risk of Lyme disease transmission to humans.

The foraging decisions involved in acquiring a meal can have an impact on an animal’s spatial distribution, as well as affect other animal species and plant communities. Thus, understanding how the foraging process varies over space and time has broad ecological implications, and optimal foraging theory can be used to identify key factors controlling foraging decisions. Optimality models are based on currencies, options and constraints. Using examples from research on free-ranging bison (Bison bison), we show how variations in these model elements can yield strong spatio-temporal variation in expected foraging decisions. First, we present a simple optimal foraging model to investigate the temporal scale of foraging decisions. On the basis of this model, we identify the foraging currency and demonstrate that such a simple model can be successful at predicting animal distribution across ecosystems. We then modify the model by changing (1) the forager’s option, from the selection of individual plants to the selection of food bites that may include more than one plant species, (2) its constraints, from being omniscient to having incomplete information of resource quality and distribution and (3) its currency, from the maximisation of energy intake rate (E) to the maximisation of the ratio between E and mortality risk (u).We also show that, where the maximisation of E fails, the maximisation of E/u can explain the circadian rhythm in the diet and movements of bison. Simple optimal foraging-theory models thus can explain changes in dietary choice of bison within a foraging patch and during the course of a day.