Three Tests To Measure Sloth Bears’ Cognitive Skills

The main aim of comparative psychology is to test cognitive skills across different species, and from their distribution gain better insight into the socio-ecological factors that may have led to the evolution of specific or specifically-enhanced cognitive skills. Despite this laudable purpose, studies of animal cognition are still very unequally distributed across taxa. In the last years, such a situation has been slightly improving, with a series of studies on “unusual” species – whatever this term exactly means.

Thanks to the brilliant work of several researchers around the world (including Holekamp and Volk’s research groups), for instance, species within the Carnivore order are now known to perform amazingly in a series of cognitive tasks, spanning from cooperation to innovation, just to name a few. Crucially, this also shows that Carnivores can be extremely valuable candidates for studies of comparative psychology. Given their huge variability in terms of social systems (ranging from solitary to fission-fusion), ecological characteristics (from low to very high dietary complexity), and domestication levels, indeed, this order is an ideal one to explore the evolutionary origins of specific cognitive skills.

With this in mind, my co-authors and I decided to test one Carnivore species, sloth bears (Melursus ursinus), whose cognitive abilities had never been previously explored. Like other bears, this species has not been domesticated and shows very limited social interactions (especially in the adult life-stage). Nonetheless, sloth bears have large brains, which thus probably evolved as a result of mainly ecological pressures. As nothing was known on sloth bears at the time we ran the study (see Call for a following work on the same species), we decided to use some classic tasks (i.e. tasks that have already been successfully used across a variety of taxa) to assess some basic cognitive skills in the physical domain.

In the first task, we tested sloth bears’ object permanence skills. Having object permanence skills basically means that one knows that objects keep on existing in space and time, despite being temporarily not visible. Easy as it may seem, this ability is not innate, and develops through the first months of age even in humans.

To test object permanence, we first familiarized individual bears with a tree trunk having three holes. During the test, the bears watched a highly-preferred object (i.e. cereals or fruit dipped in honey) disappear in one of the three holes. While the food remained hidden inside the trunk, the experimenter pushed the trunk closer to the bear, who could try and get the food by selecting only one of the holes (pointing, touching it, or trying to reach for it). In this test, bears retrieved the food in the correct hole significantly above chance, suggesting that they understand that objects continue to exist even when they are not visible.

In the second task, we tested bears’ short-term memory, to understand whether they keep memories of food location even after short time delays. For this reason, we used exactly the same set-up as above, but a delay between 30 seconds and 2 minutes elapsed between the moment food disappeared in the trunk, and after the moment subjects could make their choice. In this task, bears performed at chance levels in all conditions, suggesting important limits in their short-term memory.

In a third task, we tested whether bears can use acoustic cues (or their absence) to locate hidden food. Depending on the condition, we, therefore, shook one of two identical containers (only one of which had been previously baited) and then allowed subjects to choose one of the two containers and obtain its contents. If the full container was shaken, bears heard the noise and had to reach for the shaken container. If the empty container was shaken, bears heard no noise and had to infer per exclusion that the food was thus hidden in the opposite container. In both cases, bears selected the baited container significantly above chance.

Finally, to control that bears in our tasks were not locating food through inadvertent olfactory cues, we repeated the procedure of the first task, with the only exception that food was positioned in one of the three holes when the subject was not watching. In this case, bears selected at chance, confirming that they did not rely on olfactory cues to successfully locate food in the first and last tasks.

Can these results be explained in terms of sloth bears’ socio-ecological characteristics? Sloth bears largely feed on termites and honeybee colonies, which mostly consist of hidden prey; therefore, it comes as no real surprise that they show object permanence.

Failure in the short-term memory test, instead, may suggest that memory for short-term punctual events has little evolutionary significance for bears, who do not specifically need short-term memory to keep track of prey when they are out of view, as bears do not individually follow and eat single insects and do not specifically need it to keep track of conspecifics, as their social life is very limited. In contrast, acoustic cues may play an important role in this species: solitary individuals communicate with distant conspecifics through a large vocal repertoire, and relying on acoustic cues may be especially advantageous when foraging at night.

These findings are described in the article entitled Understanding of object properties by sloth bears, Melursus ursinus ursinus, recently published in the journal Animal Behaviour. This work was conducted by Federica Amici from the University of Leipzig and the Max Planck Institute for Evolutionary Anthropology, Trix Cacchione from the University of Bern and University of Applied Sciences Northwestern Switzerland, and Nereida Bueno-Guerra from the Universidad Pontificia de Comillas.