Animal Cognition

Is human language necessary for logical reasoning? Multiple studies address this by showing animals a bait hidden in either of two inverted cups, briefly lifting one exposing its content and then letting subjects choose. Consistent preference for the baited container is interpreted as evidence of inferential reasoning by exclusion ("if the bait is not here, it must be there"). More complex variations invoke metacognitive revision of hypotheses. We show that these protocols do not demonstrate logical reasoning. Instead, each event-stimulus elicits different actions and this suffices to choose correctly between them. We demonstrate this with experimental starlings and argue that other species may show similar behaviour, highlighting the need for careful controls when attributing high-level mental capacities such as reasoning or causal understanding.

Two related methods have been widely used to test animals capacity to reason by exclusion: Calls (2004) 2-cups task, where subjects choose between two cups in which one reward was hidden and the empty cup is revealed; and Premack & Premacks task (1994) where one reward goes in each of the two cups and one reward is removed. In both cases, success is identifying the cup with the reward. It has been suggested that among corvids, the demands associated with caching may foster exclusion abilities by reinforcing the experience of relevant contraries (e.g., empty vs. full), or by supporting domain general capacities that support caching and exclusion inferences. However, Shaw et al. (2013) found that Eurasian jays, proficient cachers, failed a version of Calls 2-cups task. To further test the caching hypothesis, we amended the procedure using invisible displacement to hide rewards, removing aspects of Shaw et al.s procedure that may have masked competence of Eurasian jays. We tested seven Eurasian jays, who showed spontaneous success on both Calls 1-reward/show-empty (80% correct) and Premack and Premacks 2-reward/remove-1 (85% correct). These results align with the hypothesis that corvids exclusion reasoning is related to caching behaviour.

Cognitive adaptations for processing causal information are fundamental to flexible problem-solving across species. Patterned-string tasks offer a classic means-end problem, requiring subjects to identify the functional connection between an action and its distal outcome. This study investigated patterned-string pulling in East Asian hornbills, a cognitively underexplored avian taxon, to assess their problem-solving abilities and underlying cognitive mechanisms. Nine hornbills were presented with three string configurations of increasing complexity: parallel (baseline), broken (contact/no-contact), and crossed strings. Subjects reliably selected the correct string in the baseline and contact conditions, demonstrating sensitivity to physical continuity cues. In contrast, accuracy declined significantly in the crossed condition, which required recognition of spatial displacement between the baited and accessible string ends. Reaction time (RT) analyses revealed prolonged latencies in the crossed condition, likely reflecting increased cognitive demands or limited habituation. Individuals who failed often adopted suboptimal strategies, such as proximity or side biases. We also observed lateralisation effects, with a left-side RT advantage, suggesting hemispheric specialisation. These findings support the hypothesis that task complexity modulates both decision accuracy and response speed, highlight RT as a useful proxy for interim cognitive processing, and establish hornbills as a valuable comparative model for studying causal cognition in birds.

The extraordinary olfactory capabilities in detection and rescue dogs are well-known. However, the olfactory performance varies by breed and search environment (Jezierski et al., 2014), as well as by the quantity of training (Horowitz et al., 2013). While detection of an olfactory cue inherently demands a judgment regarding the presence or absence of a cue at a given location, olfactory discrimination requires an assessment of quantity, a task demanding more attention and, hence, decreasing reliability as an informational source (Horowitz et al., 2013). This study aims at gaining more clarity on detection and discrimination of olfactory cues in untrained dogs and in a variety of dog breeds. Using a two-alternative forced choice (2AFC) paradigm, we assessed olfactory detection scores by presenting a varied quantity of food reward under one or the other hidden cup, and discrimination scores by presenting two varied quantities of food reward under both hidden cups. We found relatively reliable detection performances across all breeds and limited discrimination abilities, modulated by breed. We discuss our findings in relation to the cognitive demands imposed by the tasks and the cephalic index of the dog breeds.

Insight--that eureka moment when understanding emerges without trial-and-error--has traditionally been limited to great apes, corvids, and parrots (Byrne, 2016; Emery & Clayton, 2004; Lambert et al., 2019). The vertical string-pulling paradigm, a hallmark test of avian cognition (Jacobs & Osvath, 2015), requires birds to retrieve suspended food by incrementally pulling the string with their bill while securing it with their foot--demonstrating both motor coordination and causal understanding. However, Oriental pied hornbills (Anthracoceros albirostris) present an interesting case due to their unique anatomical constraints--their fused proximal phalanges prevent them from employing the typical foot-assisted string-pulling technique used by other birds. This study challenged six Oriental pied hornbills with two distinct vertical string-pulling problems, each requiring different solving strategies based on their causal properties. Despite their pedal limitations, five of the six subjects successfully solved both problems with remarkable efficiency, achieving solutions within an average of 17.6 seconds of their first encounter with the tasks. Most notably, the successful hornbills immediately employed task-specific biomechanical strategies that directly addressed the causal relationship between the string and the food reward: vigorous shaking movements to dislodge unsecured rewards, versus precise bill-and-tongue manipulation for rewards firmly attached to the string. The hornbills immediate application of causally appropriate, task-specific solutions without prior trial-and-error strongly supports the genuine insight hypothesis rather than associative learning explanations. These findings reveal sophisticated problem-solving capabilities in this previously understudied taxon and contribute valuable evidence to our understanding of convergent cognitive evolution across phylogenetically distant species.

Eurasian jays (Garrulus glandarius) do not habitually use tools, yet they can be trained to solve object-dropping tasks, i.e. to insert a tool into an apparatus to release a food reward. Previous research suggests the these jays can learn a preference toward functional tools - objects allowing to obtain a food reward placed inside an apparatus - according to their density (Cheke et al., 2011). However, it is not yet known whether they can also select functional tools (tool selectivity) according to other physical properties such as size and shape, and use different kinds of tools to solve a similar task. Here we conducted three object-dropping experiments aimed at exploring these abilities in Eurasian jays. In Experiment 1, jays tended to select large stones as tools irrespective of the diameter of the apparatus. However, jays progressively developed a preference for the small tool, which was functional with both the wide and the narrow apparatuses. In Experiment 2, only vertically-oriented long stones could fit into the narrow apparatus, whereas both long and round stones were functional with the wide apparatus. Jays showed a preference for the long stone and, with the narrow apparatus, tended to achieve the correct manipulation after one or more unsuccessful attempts. In Experiment 3, jays were able to use sticks and adopt a novel technique on the same object-dropping apparatus, thus providing the first evidence that Eurasian jays can use sticks as tools. Taken together, these results indicate that Eurasian jays may have limited tool selectivity abilities but nonetheless can use different kinds of tools to solve similar tasks.

Object-based tasks are widely used in rodent behavioral research, yet object selection remains largely unsystematic. We present a paired-object validation protocol in which objects differ along one researcher-defined feature, allowing assessment of whether that feature is salient to the animal. Using six object pairs varying in height, color, shape, or aperture presence, we tested two wild-caught mice species with contrasting ecologies. Wood mice (Sylvaemus uralensis) and striped field mice (Apodemus agrarius) showed equal preference for both objects in most pairs, indicating that color, apertures, and apex shape differences are not salient under the tested conditions and can be used interchangeably in object recognition tasks. Height, however, produced ecology-predicted responses: arboreal wood mice avoided the shortest object while open-habitat striped field mice did not. These results demonstrate that the protocol successfully detects feature salience when present and that ecological background predicts which features matter. Summary StatementA systematic paired-object protocol reveals that most researcher-defined features (color, holes, shape) do not affect rodent exploration, but height preferences emerge in ecology-predicted patterns, demonstrating that feature salience is species-specific.

Grouping sets of elements into smaller, equal-sized, subsets constitutes a perceptual strategy employed by humans and other animals to enhance cognitive performance. We hypothesized that asymmetrical grouping, a characteristic of prime numbers, could provide visual cues enabling discrimination of prime from non-prime numerosities. Newborn chicks were habituated to even numerosities (as sets of elements presented on a screen), and then tested for their spontaneous choice among a prime and a non-prime odd numerosity. Chicks discriminated and preferred the prime over the composite set of elements. We discuss this result in terms of novelty preference. By employing different contrasts (i.e., 7vs.9, 11vs.9, and 13vs.15) we investigated the limits of such mechanism showing that induced grouping positively affects chicks performance.

Recent studies showed that humans, regardless of age, education, and culture, can extract the linear trend of a noisy graph. Here, we examined whether such skills for intuitive statistics are confined to humans or may also exist in non-human primates. We trained Guinea baboons (Papio papio) to associate arbitrary geometrical shapes with the increasing or decreasing trends of noiseless and noisy scatterplots, while varying the number of points, the noise level, and the regression slope. Many baboons successfully learned this conditional match-to-sample task for both noiseless and noisy plots. Crucially, for successful baboons, accuracy varied as a sigmoid function of the t-value of the regression, the same statistical index upon which humans also base their answers, even after controlling for other variables. These results are compatible with the hypothesis that the human perception of data graphics is based on the pre-emption and recycling of a phylogenetically older competence of the primate visual system for extracting the principal axes of visual displays.

Causal understanding in animal cognition can be divided into two broad categories (Woodward, 2011): learned associations between cause and effect (Le Pelley et al., 2017) and understanding based on underlying mechanisms (Johnson and Ahn, 2017). One experiment that gives insight to animals use of causal mechanisms is the stone-dropping task. In this, subjects are given an opportunity to push a platform to make it collapse and are then required to innovate dropping a stone tool to recreate the platform collapsing (von Bayern et al., 2009). We describe how 16/18 subjects of two species of macaw (n=18; Ara ambiguus (n=9) & Ara glaucogularis (n=9)) were able to innovate the solution in this task. Many of the subjects were able to innovate the behaviour through exploratory object combination, but it is also possible that a mechanistic understanding of the necessity for contact with the platform influenced some subjects behaviour. All the successful subjects were able to recreate their novel stone-dropping behaviour in the first or second trial after innovation (and all trials thereafter) and they were also able to do the behaviour increasingly faster. This suggests they also rely on learned associations of cause and effect. However, in a transfer task in which subjects had to guide a stick tool to make it touch a differently positioned platform, all but one of the subjects failed. This would suggest that the majority of the subjects were not using an understanding of platform contact to solve the task, although the subjects difficulty with using stick tools may have also affected their performance in this transfer.

The ability to regulate and withhold an immediate behaviour in pursuit of a more preferred or valuable, albeit delayed, outcome is regarded as an important cognitive ability enabling adaptive decision-making in both social and asocial contexts. Abilities to cope with a delay in gratification have been investigated in a range of species using a variety of experimental paradigms. The present study attempts a first systematic evaluation of available experimental data from non-human animals, which is an essential basis for quantifying biological and non-biological factors (e.g. socio-ecology versus experimental design) affecting performance in delay of gratification tasks. Data were sourced from 52 separate studies, and a comprehensive overview of the available literature on delay of gratification in non-human animals is presented. We present data from 21 species, spanning across eight taxonomic order, with 1-9 species tested per taxonomic group. We highlight variation in experimental paradigms used to study delay of gratification abilities in non-human animals, both with regard to reward type or experimental setup, and discuss the implications for comparative analyses. We conclude that, at present, cross-species comparisons of delay of gratification abilities are hindered by a lack of consistency in experimental designs and low number of species tested across taxonomic orders. We hope to stimulate research on a more diverse set of species, and that future studies consider potential social and ecological factors that cause intra-specific variation in test performances, that is repeatedly seen across species.

The ability to stop behaviour is essential for adapting to changes in the environment, a principle that holds true across various species. While traditionally considered a unitary psychological construct, recent studies indicate that this ability is multifaceted. Our research evaluates this multifaceted nature using three tasks that measure stopping in different contexts in two related gull species: herring gulls (Larus argentatus) and lesser black-backed gulls (L. fuscus). These species were selected for their distinct migration and foraging strategies, offering a unique lens through which to examine behavioural adaptations. Across tasks and species, we conceptualised stopping as a race between a go and a stop runner, and predicted correlations based on the type of stop stimulus, the relative timing of the go and the stop stimuli, and the type of action that needed to be stopped. We found correlations between measures of going across tasks, but there was less consistency in measures of stopping. Furthermore, we observed significant differences in going and stopping behaviours that were specific to each species, which may be linked to their migration and foraging strategies. These findings highlight the importance of considering the multifaceted nature of stopping in evolutionary and behavioural studies.

Most catarrhine primates are considered to be strongly visually oriented, obtaining information about conspecifics and their environment from a diversity of visual cues. Other sensory modalities may provide information that is redundant and/or complimentary to visual cues. When cues from multiple sensory modalities are available, these may reinforce or suppress each other, as shown in several taxa ranging from insects to humans. Here, we tested how the presence and ambiguity of visual information affects the use of olfactory cues when exploring food and non-food items in semi-free ranging Barbary macaques at Affenberg Salem, Germany. We presented monkeys with pipes containing food (peanuts, popcorn), non-food (stones, feces) or no items in transparent or opaque containers, and assessed whether animals looked, sniffed and/or grabbed into the pipes depending on visibility of the contents (experiment 1). Visual information had no robust effect on sniffing probability, but it did affect sniffing behavior with respect to the timing of olfactory inspections (i.e. whether sniffing occurred first). Both visual and olfactory information affected, whether or not monkeys attempted to retrieve the items from the pipes. Furthermore, we manipulated the visual appearance of familiar food items (popcorn) with food colorant (experiment 2), which resulted in substantially increased olfactory inspections compared to unmanipulated popcorn. Taken together, reliance on the olfactory sense was modulated by the available visual information, emphasizing the interplay between different sensory modalities for obtaining information about the environment.

Inhibitory control is one of several cognitive mechanisms required for self-regulation, decision making and attention towards tasks. Linked to a variety of maladaptive behaviours in humans, inhibitory control is expected to influence behavioural plasticity in animals in the context of foraging, social interaction, or responses to sudden changes in the environment. One widely used cognitive assay, the detour task, putatively tests inhibitory control. In this task, subjects must avoid impulsively touching transparent barriers positioned in front of food, and instead access the food by an alternative but known route. Recently it has been suggested that the detour task is unreliable and measures factors unrelated to inhibitory control, including motivation, previous experience and persistence. Consequently, there is growing uncertainty as to whether this task leads to erroneous interpretations about animal cognition and its links with socio-ecological traits. To address these outstanding concerns, we designed a variant of the detour task for wild great tits (Parus major) and deployed it at the nesting site of the same individuals across two spring seasons. This approach eliminated the use of food rewards, limited social confounds, and maximised motivation. We compared task performance in the wild with their performance in captivity when tested using the classical cylinder detour task during the non-breeding season. Task performance was temporally and contextually repeatable, and none of the confounds had any significant effect on performance, nor did they drive any of the observed repeatable differences among individuals. These results support the hypothesis that our assays captured intrinsic differences in inhibitory control. Instead of throwing the detour task out with the bathwater, we suggest confounds are likely system and experimental-design specific, and that assays for this potentially fundamental but largely overlooked source of behavioural plasticity in animal populations, should be validated and refined for each study system.

Behavioral plasticity can be described as the ability to adjust behavior depending on environmental information. We used a food-storing (caching) paradigm, during which individuals either ate or cached food under different conditions, to investigate whether they could adjust their caching behavior when observed by conspecifics and heterospecifics, and which cues they used to elicit these behavioral changes. We examined the location and number of caches made by two corvid species differing in sociality, highly social pinyon jays (Gymnorhinus cyanocephalus) and less social Clarks nutcrackers (Nucifraga columbiana). Although pinyon jays cached a similar amount of food across conditions, they allocated more caches to areas less accessible to the observer. Nutcrackers, however, reduced the number of seeds cached when another nutcracker was present in comparison to when they cached alone. Both species relied on different social cues to elicit re-caching: pinyon jays responded to the amount of time the observer spent close to the caching locations, whereas nutcrackers responded to the amount of time the observer spent pilfering their caches. The differences in cache protection behaviors and the social cues eliciting them may be explained by the species social organization. Pinyon jays may only adjust their caching behavior when necessary, as they are often surrounded by other individuals. Clarks nutcrackers reduce their caching when observed, as they have more opportunities to cache alone, and may resort to additional cache protection when experiencing pilferage. Overall, our results provide insight into understanding how pressures associated with the social environment may influence foraging behaviors.

Episodic-like memory is considered as one of the most advanced cognitive capacities and could therefore have major implications on how animals with this capacity are treated in respect to their welfare. Yet, episodic-like memory has only been substantiated in a few vertebrate species. These species are quite diverse, originate from several taxa, and the question is raised whether episodic-like memory is indeed a rather specific capacity or whether this capacity turns out to be more common when further species are investigated. In general, intensively foraging animals that rely on some replenishing resources are promising species to investigate episodic-like memory. Here, we conducted a pilot study on minipigs (Sus scrofa f. domestica), replicating an approach used in dogs. Both, dogs and pigs are macrosmate species, making the comparison of their memory of olfactory stimuli promising. The minipigs learnt to differentiate the sequence of encountered location-smell combinations as easily as dogs. This makes minipigs a promising species and the use of smells a promising approach to investigate episodic-like memory in more detail. This study also hints towards a cognitive capacity of pigs, which raises further concerns regarding intensive and barren housing conditions.

The capacity for animals to produce and comprehend vocalisations that provide referential acoustic cues to their eliciting cause (e.g. predator-specific alarm calls and food calls) is a highly adaptive means of maximising the benefits of group living, and has been widely studied in diverse species. However, an underexplored dimension of referentiality is the ability to process the direction in which a vocalisation is emitted as a cue towards its referent. Here, we replicated an experimental design previously applied to dogs, chimpanzees and human infants to investigate whether domestic goats (Capra hircus) can use human voices as a directional cue in a hidden-object task. Twenty-nine goats from a UK sanctuary participated in three experimental conditions. In each condition, goats were individually presented with a human experimenter obscured by a barrier, and two identical containers, one of which was baited with food. In the reward directed speech condition, the experimenter vocalised excitedly towards the baited container while sitting closer to the un-baited container, and then the goat was able to select which container to explore. While substantial inter-individual variation existed, on average, subjects chose the baited container at above chance level across the four trials. Two control conditions explored alternative explanations for this result: a "no speech" condition, in which the experimenter was silent but remained in the same location as in the test condition, and a "non-reward directed speech" condition, in which the experimenter directed their voice away from both containers. Subjects showed no evidence of choosing the baited container at above chance level in either of these control conditions. We conclude that goats, like dogs, but not chimpanzees, are capable of attending to the directional cues provided by human voices, and discuss the possible role of domestication in the taxonomic distribution of this ability.

Cognitive abilities allow animals to navigate through complex, fluctuating environments. In the present study, we tested the performance of a captive group of eight crows, Corvus corone, and ten domestic chickens, Gallus gallus domesticus, in the cylinder task, as a test of motor inhibitory control, and reversal learning as a measure of learning ability and behavioural flexibility. Four crows and nine chickens completed the cylinder task, eight crows and six chickens completed the reversal learning experiment. Crows performed better in the cylinder task compared to chickens. In the reversal learning experiment, species did not significantly differ in the number of trials until learning criterion was reached. Performance in the reversal learning experiment did not correlate with performance in the cylinder task in chickens. Our results suggest crows to possess better motor inhibitory control compared to chickens. In contrast, learning performance in a reversal learning task did not differ between the species, indicating similar levels of behavioural flexibility. Interestingly, we describe notable individual differences in performance. We stress the importance not only to compare cognitive performance between species but also between individuals of the same species when investigating the evolution of cognitive skills.

Extractive foraging is generally studied from the perspective of behavioral flexibility, cognitive ability, innovation and social learning. Despite its potential to elucidate synanthropic adaptation in species exploiting enclosed anthropogenic food, research on extractive foraging under urban conditions is limited. Since a large extent of anthropogenic food is packaged and contains highly processed food, processes of identification/extraction of food by nonhuman species become intriguing themes of research. We studied how processing status of embedded food determined extraction decisions across groups of a species differing in exposure and familiarity to the food. Further, we tested the generalizability of extraction methods. Experimenting with wild bonnet macaques (Macaca radiata), we found exposure- and form (native/shelled/peeled)-specific familiarity to peanuts, state (raw/boiled/roasted)-specific distinction in depeeling, and exposure- and state-specific differences in methods of depeeling. Group with the highest exposure to peanut differed in its propensity to use sophisticated extraction methods, e.g. depeeling by rubbing between palms (bimanual asymmetric action) and rubbing against horizontal substrata (unimanual action). The innovative methods were also extended to roasted peas and chickpeas by the urban group. Our study establishes a causal relationship between familiarity and processing status of food and shows the generalized extension of extraction methods based on food categorization. Summary StatementNonhuman species in cities face upheaval challenges of accessing enclosed and highly processed anthropogenic food. We studied the effects of minor processing of enclosed food on its extraction decisions.

Humans have the metacognitive capacity to be aware of what they do and do not know. While uncertainty monitoring has long been regarded as uniquely human, researchers in search of the polygenetic root of this ability have gathered evidence that primate species possess functional features parallel to humans. However, there were no systematic studies that quantitively take into account of extant data for these non-primate animals. Through a meta-analysis, we collected published data reported in 11 articles from 55 individual non-primate animals spanning over four species on the "opt-out" paradigm, the most prevailing paradigms used to test nonhuman animals uncertainty monitoring. We used chosen-forced advantage and opt-out rate to quantify animals performance results for computing the aggregated effect size for this literature. We found that these four NPA species process a significantly positive effect size for both scores and identified the moderators that have contributed to the inconsistencies across these studies. Implications for theories on metacognition are discussed.