Planning and inhibition in corvids

Abstract: Executive functions (EF) denote the host of top-down mental processes that mediate self-regulation to achieve a goal. Among other things, EF support cognitive flexibility by allowing inhibition of automatic responses in favour of obtaining future goals. Corvids, the bird group that is comprised of ravens, crows and jays, are distinguished by their cognitive flexibility. This thesis investigates two components of EF – planning and inhibition – in corvids, in an attempt to understand the evolution of complex cognitive abilities.Planning is a cognitive ability that is at the core of human society. Human planning is flexible and domain-general: we can plan across different contexts, for example when planning to go to the moon or when planning to organize a party. It was long considered that flexible planning skills were restricted to the great ape lineage. Previous studies documented corvids can plan their food caching behaviour, but since corvids are habitual food-cachers, some argued their planning skill is restricted to the caching domain and thus may not be domain-general. We tested whether corvid planning is domain-general, by testing ravens on ecologically-invalid planning tasks that did not involve caching (tool-use task and bartering task; Paper I). This study mainly replicated but also extended the previous studies on great apes. The results suggest ravens perform remarkably well – often on a par with great apes - in such planning tasks. Thus, the study suggests that corvid planning abilities are domain-general, and that flexible planning skills evolved independently in corvids and great apes, warranting future investigations on possible similarities at the level of underlying mechanisms.Inhibitory control is a core EF component and it allows inhibiting prepotent responses in favour of more appropriate behaviour. Motor self-regulation is a basic inhibitory mechanism that allows stopping a prepotent motor response. Given it’s a fundamental cognitive mechanism that is necessary for higher-order skills, investigating motor self-regulation comparatively can inform on broader aspects on cognitive evolution. We tested motor self-regulation in three species of corvids – ravens, New Caledonian crows and jackdaws –using a detour task that involved inhibiting a dominant motor tendency of directly reaching for a visible reward behind a transparent barrier (Paper II). This study replicated a previous large-scale study that tested 36 species on the same task. This large-scale study found absolute brain size was the best predictor of task performance, with great apes being the best performers. We found that corvids perform excellently on this task, on a par with great apes, despite their vastly smaller absolute brain sizes. This finding provided counter evidence for the importance of absolute brain volume in predicting the cognitive performance. Instead, given the large neural density of the corvids, it was suggested that total number of pallial neurons might be a better predictor.We thus investigated parrots, another bird group that has a large number of pallial neurons, on the same task (Paper IV). However, parrots performed significantly worse than corvids and great apes on this task. Further analyses revealed some of their failures were caused not by inhibitory incompetency but by bodily exploration of the barrier. Such methodological issues are discussed further in Paper V, which provides a review of the detour paradigm, a widely used task in animal cognition. In this paper we review the cognitive skills measured by the detour paradigm and suggest methodological improvements to make the task meaningfully applicable to wide range of species. In Paper III, we compare motor self-regulation developmentally: we attempt to answer whether the similarity between ravens and great apes in cognitive performance by mature individuals entails similarity on their development. The results suggest raven chicks undergo a similar pattern of development when succeeding on the detour tasks to those of human and rhesus monkey infants, but at a remarkably faster rate. These findings have implications for the mechanism behind the independent evolution of complex cognition; that is, whether complex cognitive skills can be attained independently using completely different pathways. Cognitive zoology is still a relatively young field, and future studies testing more species will give a clearer picture on the proximate and ultimate factors behind the evolution of complex cognitive skills.