Animals that cache food risk having their stored food pilfered by conspecifics. Previous research has shown that a number of food-caching species of corvid use strategies that decrease the probability of conspecifics pilfering their caches. In this experiment, we investigated whether Eurasian jays (Garrulus glandarius) would choose between caching behind an opaque and caching behind a transparent barrier whilst being observed by a conspecific. If caching in out-of-sight locations is a strategy to prevent conspecifics from pilfering these caches, then the jays should place a greater proportion of caches behind the opaque barrier when being observed than when caching in private. In accordance with this prediction, jays cached a greater proportion of food behind the opaque barrier when they were observed than when they cached in private. These results suggest that Eurasian jays may opt to cache in out-of-view locations to reduce the likelihood of conspecifics pilfering their caches.Electronic supplementary materialThe online version of this article (doi:10.1007/s10071-014-0743-2) contains supplementary material, which is available to authorized users.

SummaryMany corvid species accurately remember the locations where they have seen others cache food, allowing them to pilfer these caches efficiently once the cachers have left the scene [1]. To protect their caches, corvids employ a suite of different cache-protection strategies that limit the observers’ visual or acoustic access to the cache site 2, 3. In cases where an observer’s sensory access cannot be reduced it has been suggested that cachers might be able to minimise the risk of pilfering if they avoid caching food the observer is most motivated to pilfer [4]. In the wild, corvids have been reported to pilfer others’ caches as soon as possible after the caching event [5], such that the cacher might benefit from adjusting its caching behaviour according to the observer’s current desire. In the current study, observers pilfered according to their current desire: they preferentially pilfered food that they were not sated on. Cachers adjusted their caching behaviour accordingly: they protected their caches by selectively caching food that observers were not motivated to pilfer. The same cache-protection behaviour was found when cachers could not see on which food the observers were sated. Thus, the cachers’ ability to respond to the observer’s desire might have been driven by the observer’s behaviour at the time of caching.

Humans' predictions of another person's behaviour are regularly influenced by what they themselves might know or want. In a previous study, we found that male Eurasian jays (Garrulus glandarius) could cater for their female partner's current desire when sharing food with her. Here, we tested the extent to which the males' decisions are influenced by their own current desire. When the males' and female's desires matched, males correctly shared the food that was desired by both. When the female's desire differed from their own, the males' decisions were not entirely driven by their own desires, suggesting that males also took the female's desire into account. Thus, the male jays' decisions about their mates' desires are partially biased by their own desire and might be based upon similar processes as those found in humans.

In the Sandbox Task (e.g. Sommerville, Bernstein & Meltzoff, 2013), participants indicate where a protagonist who has a false belief about the location of an object will look for that object in a trough filled with a substrate that conceals the hidden object's location. Previous findings that participants tend to indicate a location closer to where they themselves know the object to be located have been interpreted as evidence of egocentric bias when attributing mental states to others. We tested the assumption that such biases occur as a result of reasoning about mental states specifically. We found that participants showed more egocentric bias when reasoning from a protagonist's false belief than from their own memory, but found equivalent levels of bias when they were asked to indicate where a false film would depict the object as when they were asked about a protagonist's false belief. Our findings suggest that that egocentric biases found in adult false belief tasks are more likely due to a general difficulty with reasoning about false representations than a specialised difficulty with reasoning about false mental states. Introduction Successfully comprehending what someone else thinks, knows, sees or believes, known as Theory of Mind (ToM) or 'mentalizing', is a critical ability in our day-today interactions with other people (e.g.

An important part of our Theory of Mind—the ability to reason about other people's unobservable mental states—is the ability to attribute false beliefs to others. We investigated whether processing these false beliefs, as well as similar but nonmental representations, is reliant on language. Participants watched videos in which a protagonist hides a gift and either takes a photo of it or writes a text about its location before a second person inadvertently moves the present to a different location, thereby rendering the belief and either the photo or text false. At the same time, participants performed either a concurrent verbal interference task (rehearsing strings of digits) or a visual interference task (remembering a visual pattern). Results showed that performance on false belief trials did not decline under verbal interference relative to visual interference. We interpret these findings as further support for the view that language does not form an essential part of the process of reasoning online (“in the moment”) about false beliefs.