Inferences about food location in three cercopithecine species: an insight into the socioecological cognition of primates

Contrasting two versions of the 4-cup 2-item disjunctive syllogism task in great apes

Chimpanzees excel at inference tasks which require that they search for a single food item from partial information. Yet, when presented with 2-item tasks which test the same inference operation, chimpanzees show a consistent breakdown in performance. Here we test a diverse zoo-housed cohort (n = 24) comprising all 4 great ape species under the classic 4-cup 2-item task, previously administered to children and chimpanzees, and a modified task administered to baboons. The aim of this study is to delineate whether the divergent results reported from the literature are taxonomic differences or artefacts of their methodologies, while extending the literature to cover the remaining great ape species. We find that apes adaptively adjust their choice behaviour in both variants of the task, but that they perform better in trials where the information provided rules out a location rather than removes one of the food items. In a second experiment involving those subjects who passed the first, along with a group of naïve subjects, we test whether subjects were able to apply the logical operation selectively by including control trials where the correct response is reversed. Performance in standard trials breaks down with the addition of control trials, meaning that if apes did solve the first experiment logically, they are not capable of applying that logic flexibly. Considering this finding, we conclude that a 4-cup 2-item task may not be a suitable test of logical reasoning in great apes.

Meta-analyses reveal support for the Social Intelligence Hypothesis

The Social Intelligence Hypothesis (SIH) is one of the leading explanations for the evolution of cognition. Since its inception a vast body of literature investigating the predictions of the SIH has accumulated, using a variety of methodologies and species. However, the generalisability of the hypothesis remains unclear. To gain an understanding of the robustness of the SIH as an explanation for the evolution of cognition, we systematically searched the literature for studies investigating the predictions of the SIH. Accordingly, we compiled 103 studies with 584 effect sizes from 17 taxonomic orders. We present the results of four meta-analyses which reveal support for the SIH across interspecific, intraspecific and developmental studies. However, effect sizes did not differ significantly between the cognitive or sociality metrics used, taxonomy or testing conditions. Thus, support for the SIH is similar across studies using neuroanatomy and cognitive performance, those using broad categories of sociality, group size and social interactions, across taxonomic groups, and for tests conducted in captivity or the wild. Overall, our meta-analyses support the SIH as an evolutionary and developmental explanation for cognitive variation.

Both sheep and goats can solve inferential by exclusion tasks

Despite the domestication of sheep and goats by humans for several millennia, we still lack comparative data on their cognitive capacity. Comparing the cognitive skills of farm animals can help understand the evolution of cognition. In this study, we compared the performances of sheep and goats in inference by exclusion tasks. We implemented two tasks, namely a cup task and a tube task, to identify whether success in solving the task could be attributed to either low-level mechanisms (avoiding the empty location strategy) or to deductive reasoning (if two possibilities A and B, but not A, then it must be B). In contrast to a previous study comparing goats and sheep in a cup task, we showed that both species solved the inferential condition with high success rates. In the tube task, performances could not be explained by alternative strategies such as avoiding the empty tube or preferring the bent tube. When applying a strict set of criteria concerning responses in all conditions and controlling for the potential effects of experience, we demonstrate that two individuals, a goat and a sheep, fulfil these criteria. This suggests that sheep and goats are able to make inferences based on deductive reasoning.

Come with me: experimental evidence for intentional recruitment in Tonkean macaques

Recruitment is a process by which animals can initiate collective movements: the action of an individual prompts conspecifics to follow. Although it has been hypothesized that animals may be able to intentionally recruit others, there is no experimental evidence of this to date. We tested this hypothesis in two pairs of Tonkean macaques in a situation requiring the subjects to find a food site in a 2800 m² area, and approach the site together to release rewards. Each subject was informed of the location of either highly or little-valued rewards. We recorded attention-action sequences in which an individual checked that his partner was attending to him before moving, and also simple departures (i.e., not preceded by eye contact). Analyses showed that sequences were more often followed by recruitment and leading the partner to a baited site than simple departures were. Moreover, subjects used attention-action sequences more frequently when informed of the location of the highly valued reward. This may be explained by the fact that the more motivated they were by the expected rewards, the more likely they were to actively recruit their partner. No such effect was found when subjects performed simple departures. We conclude that Tonkean macaques are capable of intentional recruitment because the subjects voluntarily behaved with the goal of influencing their partner's movement: they checked that the partner was paying attention to them and prompted him to follow by moving. Such performances can be accounted for either by associative learning or by intentional communication.

Inferential Communication: Bridging the Gap Between Intentional and Ostensive Communication in Non-human Primates

Communication, when defined as an act intended to affect the psychological state of another individual, demands the use of inference. Either the signaler, the recipient, or both must make leaps of understanding which surpass the semantic information available and draw from pragmatic clues to fully imbue and interpret meaning. While research into human communication and the evolution of language has long been comfortable with mentalistic interpretations of communicative exchanges, including rich attributions of mental state, research into animal communication has balked at theoretical models which describe mentalized cognitive mechanisms. We submit a new theoretical perspective on animal communication: the model of inferential communication. For use when existing proximate models of animal communication are not sufficient to fully explain the complex, flexible, and intentional communication documented in certain species, specifically non-human primates, we present our model as a bridge between shallower, less cognitive descriptions of communicative behavior and the perhaps otherwise inaccessible mentalistic interpretations of communication found in theoretical considerations of human language. Inferential communication is a framework that builds on existing evidence of referentiality, intentionality, and social inference in primates. It allows that they might be capable of applying social inferences to a communicative setting, which could explain some of the cognitive processes that enable the complexity and flexibility of primate communication systems. While historical models of animal communication focus on the means-ends process of behavior and apparent cognitive outcomes, inferential communication invites consideration of the mentalistic processes that must underlie those outcomes. We propose a mentalized approach to questions, investigations, and interpretations of non-human primate communication. We include an overview of both ultimate and proximate models of animal communication, which contextualize the role and utility of our inferential communication model, and provide a detailed breakdown of the possible levels of cognitive complexity which could be investigated using this framework. Finally, we present some possible applications of inferential communication in the field of non-human primate communication and highlight the role it could play in advancing progress toward an increasingly precise understanding of the cognitive capabilities of our closest living relatives.

Exclusion in the field: wild brown skuas find hidden food in the absence of visual information

Inferential reasoning by exclusion allows responding adaptively to various environmental stimuli when confronted with inconsistent or partial information. In the experimental context, this mechanism involves selecting correctly between an empty option and a potentially rewarded one. Recently, the increasing reports of this capacity in phylogenetically distant species have led to the assumption that reasoning by exclusion is the result of convergent evolution. Within one largely unstudied avian order, i.e. the Charadriiformes, brown skuas (Catharacta antarctica ssp lonnbergi) are highly flexible and opportunistic predators. Behavioural flexibility, along with specific aspects of skuas' feeding ecology, may act as influencing factors in their ability to show exclusion performance. Our study aims to test whether skuas are able to choose by exclusion in a visual two-way object-choice task. Twenty-six wild birds were presented with two opaque cups, one covering a food reward. Three conditions were used: 'full information' (showing the content of both cups), 'exclusion' (showing the content of the empty cup), and 'control' (not showing any content). Skuas preferentially selected the rewarded cup in the full information and exclusion condition. The use of olfactory cues was excluded by results in the control condition. Our study opens new field investigations for testing further the cognition of this predatory seabird.

Reasoning Through the Disjunctive Syllogism in Monkeys

The capacity for logical inference is a critical aspect of human learning, reasoning, and decision-making. One important logical inference is the disjunctive syllogism: given A or B, if not A, then B. Although the explicit formation of this logic requires symbolic thought, previous work has shown that nonhuman animals are capable of reasoning by exclusion, one aspect of the disjunctive syllogism (e.g., not A = avoid empty). However, it is unknown whether nonhuman animals are capable of the deductive aspects of a disjunctive syllogism (the dependent relation between A and B and the inference that "if not A, then B" must be true). Here, we used a food-choice task to test whether monkeys can reason through an entire disjunctive syllogism. Our results show that monkeys do have this capacity. Therefore, the capacity is not unique to humans and does not require language.

Non-human primates use combined rules when deciding under ambiguity

Decision outcomes in unpredictable environments may not have exact known probabilities. Yet the predictability level of outcomes matters in decisions, and animals, including humans, generally avoid ambiguous options. Managing ambiguity may be more challenging and requires stronger cognitive skills than decision-making under risk, where decisions involve known probabilities. Here we compare decision-making in capuchins, macaques, orangutans, gorillas, chimpanzees and bonobos in risky and ambiguous contexts. Subjects were shown lotteries (a tray of potential rewards, some large, some small) and could gamble a medium-sized food item to obtain one of the displayed rewards. The odds of winning and losing varied and were accessible in the risky context (all rewards were visible) or partially available in the ambiguous context (some rewards were covered). In the latter case, the level of information varied from fully ambiguous (individuals could not guess what was under the covers) to predictable (individuals could guess). None of the species avoided gambling in ambiguous lotteries and gambling rates were high if at least two large rewards were visible. Capuchins and bonobos ignored the covered items and gorillas and macaques took the presence of potential rewards into account, but only chimpanzees and orangutans could consistently build correct expectations about the size of the covered rewards. Chimpanzees and orangutans combined decision rules according to the number of large visible rewards and the level of predictability, a process resembling conditional probabilities assessment in humans. Despite a low sample size, this is the first evidence in non-human primates that a combination of several rules can underlie choices made in an unpredictable environment. Our finding that non-human primates can deal with the uncertainty of an outcome when exchanging one food item for another is a key element to the understanding of the evolutionary origins of economic behaviour. This article is part of the theme issue 'Existence and prevalence of economic behaviours among non-human primates'.

The Acquisition of Modal Concepts

Sometimes we accept propositions, sometimes we reject them, and sometimes we take propositions to be worth considering but not yet established, as merely possible. The result is a complex representation with logical structure. Is the ability to mark propositions as merely possible part of our innate representational toolbox or does it await development, perhaps relying on language acquisition? Several lines of inquiry show that preverbal infants manage possibilities in complex ways, while others find that preschoolers manage possibilities poorly. Here, we discuss how this apparent conflict can be resolved by distinguishing modal representations of possibility, which mark possibility symbolically, from minimal representations of possibility, which do not encode any modal status and need not have a logical structure.

Comparing physical and social cognitive skills in macaque species with different degrees of social tolerance

Contemporary evolutionary theories propose that living in groups drives the selection of enhanced cognitive skills to face competition and facilitate cooperation between individuals. Being able to coordinate both in space and time with others and make strategic decisions are essential skills for cooperating within groups. Social tolerance and an egalitarian social structure have been proposed as one specific driver of cooperation. Therefore, social tolerance is predicted to be associated with enhanced cognitive skills that underpin communication and coordination. Social tolerance should also be associated with enhanced inhibition, which is crucial for suppressing automatic responses and permitting delayed gratification in cooperative contexts. We tested the performance of four closely related non-human primate species (genus Macaca) characterized by different degrees of social tolerance on a large battery of cognitive tasks covering physical and social cognition, and on an inhibitory control task. All species performed at a comparable level on the physical cognition tasks but the more tolerant species outperformed the less tolerant species at a social cognition task relevant to cooperation and in the inhibitory control task. These findings support the hypothesis that social tolerance is associated with the evolution of sophisticated cognitive skills relevant for cooperative social living.

Reasoning by exclusion in the kea (Nestor notabilis)

Reasoning by exclusion, i.e. the ability to understand that if there are only two possibilities and if it is not A, it must be B, has been a topic of great interest in recent comparative cognition research. Many studies have investigated this ability, employing different methods, but rarely exploring concurrent decision processes underlying choice behaviour of non-human animals encountering inconsistent or incomplete information. Here, we employed a novel training and test method in order to perform an in-depth analysis of the underlying processes. Importantly, to discourage the explorative behaviour of the kea, a highly neophilic species, the training included a large amount of novel, unrewarded stimuli. The subsequent test consisted of 30 sessions with different sequences of four test trials. In these test trials, we confronted the kea with novel stimuli that were paired with either the rewarded or unrewarded training stimuli or with the novel stimuli of previous test trials. Once habituated to novelty, eight out of fourteen kea tested responded to novel stimuli by inferring their contingency via logical exclusion of the alternative. One individual inferred predominantly in this way, while other response strategies, such as one trial learning, stimulus preferences and avoiding the negative stimulus also guided the responses of the remaining individuals. Interestingly, the difficulty of the task had no influence on the test performance. We discuss the implications of these findings for the current hypotheses about the emergence of inferential reasoning in some avian species, considering causal links to brain size, feeding ecology and social complexity.

Inference by Exclusion in Goffin Cockatoos (Cacatua goffini)

Inference by exclusion, the ability to base choices on the systematic exclusion of alternatives, has been studied in many nonhuman species over the past decade. However, the majority of methodologies employed so far are hard to integrate into a comparative framework as they rarely use controls for the effect of neophilia. Here, we present an improved approach that takes neophilia into account, using an abstract two-choice task on a touch screen, which is equally feasible for a large variety of species. To test this approach we chose Goffin cockatoos (Cacatua goffini), a highly explorative Indonesian parrot species, which have recently been reported to have sophisticated cognitive skills in the technical domain. Our results indicate that Goffin cockatoos are able to solve such abstract two-choice tasks employing inference by exclusion but also highlight the importance of other response strategies.