Abstract: Visual working memory provides an essential link between perception and higher cognitive functions, allowing for the active maintenance of information about stimuli no longer in view1, 2. Research suggests that sustained activity in higher-order prefrontal, parietal, inferotemporal and lateral occipital areas supports visual maintenance3, 4, 5, 6, 7, 8, 9, 10, 11, and may account for the limited capacity of working memory to hold up to 3–4 items9, 10, 11. Because higher-order areas lack the visual selectivity of early sensory areas, it has remained unclear how observers can remember specific visual features, such as the precise orientation of a grating, with minimal decay in performance over delays of many seconds12. One proposal is that sensory areas serve to maintain fine-tuned feature information13, but early visual areas show little to no sustained activity over prolonged delays14, 15, 16. Here we show that orientations held in working memory can be decoded from activity patterns in the human visual cortex, even when overall levels of activity are low. Using functional magnetic resonance imaging and pattern classification methods, we found that activity patterns in visual areas V1–V4 could predict which of two oriented gratings was held in memory with mean accuracy levels upwards of 80%, even in participants whose activity fell to baseline levels after a prolonged delay. These orientation-selective activity patterns were sustained throughout the delay period, evident in individual visual areas, and similar to the responses evoked by unattended, task-irrelevant gratings. Our results demonstrate that early visual areas can retain specific information about visual features held in working memory, over periods of many seconds when no physical stimulus is present.

Abstract: Transitive inference (TI) involves using known relationships to deduce unknown ones (for example, using A > B and B > C to infer A > C), and is thus essential to logical reasoning. First described as a developmental milestone in children, TI has since been reported in nonhuman primates, rats and birds. Still, how animals acquire and represent transitive relationships and why such abilities might have evolved remain open problems. Here we show that male fish (Astatotilapia burtoni) can successfully make inferences on a hierarchy implied by pairwise fights between rival males. These fish learned the implied hierarchy vicariously (as 'bystanders'), by watching fights between rivals arranged around them in separate tank units. Our findings show that fish use TI when trained on socially relevant stimuli, and that they can make such inferences by using indirect information alone. Further, these bystanders seem to have both spatial and featural representations related to rival abilities, which they can use to make correct inferences depending on what kind of information is available to them. Beyond extending TI to fish and experimentally demonstrating indirect TI learning in animals, these results indicate that a universal mechanism underlying TI is unlikely. Rather, animals probably use multiple domain-specific representations adapted to different social and ecological pressures that they encounter during the course of their natural lives.

Abstract: The ant Temnothorax albipennis uses a technique known as tandem running to lead another ant from the nest to food--with signals between the two ants controlling both the speed and course of the run. Here we analyse the results of this communication and show that tandem running is an example of teaching, to our knowledge the first in a non-human animal, that involves bidirectional feedback between teacher and pupil. This behaviour indicates that it could be the value of information, rather than the constraint of brain size, that has influenced the evolution of teaching.

Abstract: All organisms interact with their environment, and in doing so shape it, modifying resource availability. Termed niche construction, this process has been studied primarily at the ecological level with an emphasis on the consequences of construction across generations. We focus on the behavioural process of construction within a single generation, identifying the role a robustness mechanism--conflict management--has in promoting interactions that build social resource networks or social niches. Using 'knockout' experiments on a large, captive group of pigtailed macaques (Macaca nemestrina), we show that a policing function, performed infrequently by a small subset of individuals, significantly contributes to maintaining stable resource networks in the face of chronic perturbations that arise through conflict. When policing is absent, social niches destabilize, with group members building smaller, less diverse, and less integrated grooming, play, proximity and contact-sitting networks. Instability is quantified in terms of reduced mean degree, increased clustering, reduced reach, and increased assortativity. Policing not only controls conflict, we find it significantly influences the structure of networks that constitute essential social resources in gregarious primate societies. The structure of such networks plays a critical role in infant survivorship, emergence and spread of cooperative behaviour, social learning and cultural traditions.