Kea show three signatures of domain-general statistical inference

Crows make optimal choices based on relative probabilities

A recent study by Johnston, Brecht, and Nieder (2023, Current Biology, 33, 3238-3243) finds that carrion crows associate varying rates of reinforcement with novel arbitrary stimuli and make optimal decisions when they must later choose between stimulus pairs. These results demonstrate that crows are capable of not only storing information about reward probabilities in their memory but also making optimal choices based on this information even a month later.

Probabilistic inference and Bayesian-like estimation in animals: Empirical evidence

Animals often make decisions without perfect knowledge of environmental parameters like the quality of an encountered food patch or a potential mate. Theoreticians often assume animals make such decisions using a Bayesian updating process that combines prior information about the frequency distribution of resources in the environment with sample information from an encountered resource; such a process leads to decisions that maximize fitness, given the available information. I examine three aspects of empirical work that shed light on the idea that animals can make such decisions in a Bayesian-like manner. First, many animals are sensitive to variance differences in behavioral options, one metric used to characterize frequency distributions. Second, several species use information about the relative frequency of preferred versus nonpreferred items in different populations to make probabilistic inferences about samples taken from populations in a manner that results in maximizing the likelihood of obtaining a preferred reward. Third, the predictions of Bayesian models often match the behavior of individuals in two main approaches. One approach compares behavior to models that make different assumptions about how individuals estimate the quality of an environmental parameter. The patch exploitation behavior of nine species of birds and mammals has matched the predictions of Bayesian models. The other approach compares the behavior of individuals who learn, through experience, different frequency distributions of resources in their environment. The behavior of three bird species and bumblebees exploiting food patches and fruit flies selecting mates is influenced by their experience learning different frequency distributions of food and mates, respectively, in ways consistent with Bayesian models. These studies lend support to the idea that animals may combine prior and sample information in a Bayesian-like manner to make decisions under uncertainty, but additional work on a greater diversity of species is required to better understand the generality of this ability.

Feathered Lectures-Evidence of Perceptual Factors on Social Learning in Kea Parrots (Nestor notabilis)

Social learning describes the acquisition of knowledge through observation of other individuals, and it is fundamental for the development of culture and traditions within human groups. Although previous studies suggest that Kea (Nestor notabilis) benefit from social learning, experimental evidence has been inconclusive, as in a recent two-action task, all perceptual factors were ignored. The present study attempts to address this by investigating social learning in Kea with a focus on social enhancement processes. In an experiment with a captive group of Kea, we investigated whether individuals that had the opportunity to observe a conspecific performing a simple task subsequently show better performance in that task than a control group without prior demonstration. This study provides a strong tendency of greater success in skill acquisition in Kea as a result of social learning. Kea that observed a conspecific solving a task showed clear evidence of perceptual factors drawing attention to the relevant parts of the experimental apparatus and manipulated these significantly more (100% of trials) than control birds (77.8% of trials). Combined with a strong trend (p = 0.056) of the test subjects solving the task more than the control subjects, this shows conclusively that Kea, at least when required to solve a task, do attend to perceptual factors of a demonstrated action.

Why birds are smart

Many cognitive neuroscientists believe that both a large brain and an isocortex are crucial for complex cognition. Yet corvids and parrots possess non-cortical brains of just 1-25 g, and these birds exhibit cognitive abilities comparable with those of great apes such as chimpanzees, which have brains of about 400 g. This opinion explores how this cognitive equivalence is possible. We propose four features that may be required for complex cognition: a large number of associative pallial neurons, a prefrontal cortex (PFC)-like area, a dense dopaminergic innervation of association areas, and dynamic neurophysiological fundaments for working memory. These four neural features have convergently evolved and may therefore represent 'hard to replace' mechanisms enabling complex cognition.

Animal behaviour: Darwin's mischievous hat stealers are innovative problem solvers

A new study shows a falcon species, the striated caracara, displays similar levels of behavioural innovation to tool-using parrots when solving a battery test in the wild.

Kea, bird of versatility. Kea parrots (Nestor notabilis) show high behavioural flexibility in solving a demonstrated sequence task

Social learning is an important aspect of dealing with the complexity of life. The transmission of information via the observation of other individuals is a cost-effective way of acquiring information. It is widespread within the animal kingdom but may differ strongly in the social learning mechanisms applied by the divergent species. Here we tested eighteen Kea (Nestor notabilis) parrots on their propensity to socially learn, and imitate, a demonstrated sequence of steps necessary to open an apparatus containing food. The demonstration by a conspecific led to more successful openings by observer birds, than control birds without a demonstration. However, all successful individuals showed great variation in their response topography and abandoned faithfully copying the task in favour of exploration. While the results provide little evidence for motor imitation they do provide further evidence for kea's propensity towards exploration and rapidly shifting solving strategies, indicative of behavioural flexibility.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10336-023-02127-y.

Crows flexibly apply statistical inferences based on previous experience

Statistical inference, the ability to use limited information to draw conclusions about the likelihood of an event, is critical for decision-making during uncertainty. The ability to make statistical inferences was thought to be a uniquely human skill requiring verbal instruction and mathematical reasoning.1 However, basic inferences have been demonstrated in both preliterate and pre-numerate individuals,2,3,4,5,6,7 as well as non-human primates.8 More recently, the ability to make statistical inferences has been extended to members outside of the primate lineage in birds.9,10 True statistical inference requires subjects use relative rather than absolute frequency of previously experienced events. Here, we show that crows can relate memorized reward probabilities to infer reward-maximizing decisions. Two crows were trained to associate multiple reward probabilities ranging from 10% to 90% to arbitrary stimuli. When later faced with the choice between various stimulus combinations, crows retrieved the reward probabilities associated with individual stimuli from memory and used them to gain maximum reward. The crows showed behavioral distance and size effects when judging reward values, indicating that the crows represented probabilities as abstract magnitudes. When controlling for absolute reward frequency, crows still made reward-maximizing choices, which is the signature of true statistical inference. Our study provides compelling evidence of decision-making by relative reward frequency in a statistical inference task.

Cognition: Crows are natural statisticians

A new study shows that carrion crows use memorized reward associations in a combinatorial way to apply relative probabilistic information to optimize reward outcome. This for the first time shows that a corvid species can flexibly apply statistical inference during decision making.

Keep numbers in view: red-eared sliders (Trachemys scripta elegans) learn to discriminate relative quantities

The ability to discriminate relative quantities, one of the numerical competences, is considered an adaptive trait in uncertain environments. Besides humans, previous studies have reported this capacity in several non-human primates and birds. Here, we test whether red-eared sliders (Trachemys scripta elegans) can discriminate different relative quantities. Subjects were first trained to distinguish different stimuli with food reward. Then, they were tested with novel stimulus pairs to demonstrate how they distinguished the stimuli. The results show that most subjects can complete the initial training and use relative quantity rather than absolute quantity to make choices during the testing phase. This study provides behavioural evidence of relative quantity discrimination in a reptile species and suggests that such capacity may be widespread among vertebrates.

Giraffes make decisions based on statistical information

The ability to make inferences based on statistical information has so far been tested only in animals having large brains in relation to their body size, like primates and parrots. Here we tested if giraffes (Giraffa camelopardalis), despite having a smaller relative brain size, can rely on relative frequencies to predict sampling outcomes. We presented them with two transparent containers filled with different quantities of highly-liked food and less-preferred food. The experimenter covertly drew one piece of food from each container, and let the giraffe choose between the two options. In the first task, we varied the quantity and relative frequency of highly-liked and less-preferred food pieces. In the second task, we inserted a physical barrier in both containers, so giraffes only had to take into account the upper part of the container when predicting the outcome. In both tasks giraffes successfully selected the container more likely to provide the highly-liked food, integrating physical information to correctly predict sampling information. By ruling out alternative explanations based on simpler quantity heuristics and learning processes, we showed that giraffes can make decisions based on statistical inferences.

Recent developments in parrot cognition: a quadrennial update

Psittacines, along with corvids, are commonly referred to as 'feathered apes' due to their advanced cognitive abilities. Until rather recently, the research effort on parrot cognition was lagging behind that on corvids, however current developments show that the number of parrot studies is steadily increasing. In 2018, M. L. Lambert et al. provided a comprehensive review on the status of the most important work done so far in parrot and corvid cognition. Nevertheless, only a little more than 4 years after this publication, more than 50 new parrot studies have been published, some of them chartering completely new territory. On the 25th anniversary of Animal Cognition we think this warrants a detailed review of parrot cognition research over the last 4 years. We aim to capture recent developments and current trends in this rapidly expanding and diversifying field.

Contrafreeloading in kea (Nestor notabilis) in comparison to Grey parrots (Psittacus erithacus)

Contrafreeloading-working to access food that could be freely obtained-is rarely exhibited and poorly understood. Based on data from Grey parrots (Psittacus erithacus), researchers proposed a correlation between contrafreeloading and play: that contrafreeloading is more likely when subjects view the task as play. We tested that hypothesis by subjecting a relatively more playful parrot species, the kea (Nestor notabilis), to the same experimental tasks. Experiment 1 presented eight kea with container pairs holding more- or less-preferred free or enclosed food items, and examined three types of contrafreeloading: calculated (working to access preferred food over less-preferred, freely available food); classic (working to access food identical to freely available food); and super (working to access less-preferred food over preferred, freely available food). At the group level, the kea behaved similarly to the Greys: They significantly preferred calculated contrafreeloading, performed classic contrafreeloading at chance, and significantly failed to super contrafreeload. However, overall kea engaged in more contrafreeloading than Greys. Experiment 2 examined a potentially more ecologically relevant task, a choice between shelled and unshelled walnuts. No kea contrafreeloaded for nuts, whereas two of five Greys significantly preferred nut contrafreeloading and one chose at chance. We examine proximate and adaptive explanations for the performances of these differentially playful parrot species to further elucidate the role of play in contrafreeloading.

Experiencing statistical information improves children's and adults' inferences

How good are people's statistical intuitions? Recent research has highlighted that sequential experience of statistical information improves adults' statistical intuitions relative to situations where this information is described. Yet little is known about whether this is also the case for children's statistical intuitions. In a study with 100 children (8-11 years old) and 100 adults (19-35 years old), we found that sequentially experiencing statistical information improved both adults' and children's inferences in two paradigmatic reasoning problems: conjunction and Bayesian reasoning problems. Moreover, adults' statistical competencies when they learned statistical information through description were surpassed by children's inferences when they learned through experience. We conclude that experience of statistical information plays a key role in shaping children's reasoning under uncertainty-a conclusion that has important implications for education policy.

Direct Inference and Probabilistic Accounts of Induction

Schurz (2019, ch. 4) argues that probabilistic accounts of induction fail. In particular, he criticises probabilistic accounts of induction that appeal to direct inference principles, including subjective Bayesian approaches (e.g., Howson 2000) and objective Bayesian approaches (see, e.g., Williamson 2017). In this paper, I argue that Schurz' preferred direct inference principle, namely Reichenbach's Principle of the Narrowest Reference Class, faces formidable problems in a standard probabilistic setting. Furthermore, the main alternative direct inference principle, Lewis' Principal Principle, is also hard to reconcile with standard probabilism. So, I argue, standard probabilistic approaches cannot appeal to direct inference to explicate the logic of induction. However, I go on to defend a non-standard objective Bayesian account of induction: I argue that this approach can both accommodate direct inference and provide a viable account of the logic of induction. I then defend this account against Schurz' criticisms.

Self-care tooling innovation in a disabled kea (Nestor notabilis)

Tooling is associated with complex cognitive abilities, occurring most regularly in large-brained mammals and birds. Among birds, self-care tooling is seemingly rare in the wild, despite several anecdotal reports of this behaviour in captive parrots. Here, we show that Bruce, a disabled parrot lacking his top mandible, deliberately uses pebbles to preen himself. Evidence for this behaviour comes from five lines of evidence: (i) in over 90% of instances where Bruce picked up a pebble, he then used it to preen; (ii) in 95% of instances where Bruce dropped a pebble, he retrieved this pebble, or replaced it, in order to resume preening; (iii) Bruce selected pebbles of a specific size for preening rather than randomly sampling available pebbles in his environment; (iv) no other kea in his environment used pebbles for preening; and (v) when other individuals did interact with stones, they used stones of different sizes to those Bruce preened with. Our study provides novel and empirical evidence for deliberate self-care tooling in a bird species where tooling is not a species-specific behaviour. It also supports claims that tooling can be innovated based on ecological necessity by species with sufficiently domain-general cognition.

Are parrots naive realists? Kea behave as if the real and virtual worlds are continuous

Human psychology and animal cognition have increasingly used virtual stimuli to test cognitive abilities, with the expectation that participants are 'naive realists', that is, that they perceive virtual environments as both equivalent and continuous with real-life equivalents. However, there have been no attempts to investigate whether nonhuman subjects in fact behave as if physical processes in the virtual and real worlds are continuous. As kea parrots have previously shown the ability to transfer knowledge between real stimuli and both images on paper and images on touchscreens, here we test whether kea behave as naive realists and so expect physical processes to be continuous between the physical and virtual worlds. We find that, unlike infants, kea do not discriminate between these two contexts, and that they do not exhibit a preference for either. Our findings therefore validate the use of virtual stimuli as a powerful tool for testing the cognition of nonhuman animal species.

Kea (Nestor notabilis) fail a loose-string connectivity task

Naïve individuals of some bird species can rapidly solve vertical string-pulling tasks with virtually no errors. This has led to various hypotheses being proposed which suggest that birds mentally simulate the effects of their actions on strings. A competing embodied cognition hypothesis proposes that this behaviour is instead modulated by perceptual-motor feedback loops, where feedback of the reward moving closer acts as an internal motivator for functional behaviours, such as pull-stepping. To date, the kea parrot has produced some of the best performances of any bird species at string-pulling tasks. Here, we tested the predictions of the four leading hypotheses for the cognition underpinning bird string-pulling by presenting kea with a horizontal connectivity task where only one of two loose strings was connected to the reward, both before and after receiving perceptual-motor feedback experience. We find that kea fail the connectivity task both before and after perceptual-motor feedback experience, suggesting not only that kea do not mentally simulate their string-pulling actions, but also that perceptual-motor feedback alone is insufficient in eliciting successful performance in the horizontal connectivity task. This suggests a more complex interplay of cognitive factors underlies this iconic example of animal problem-solving.

More than the sum of its parts: Exploring the development of ratio magnitude versus simple magnitude perception

Humans perceptually extract quantity information from our environments, be it from simple stimuli in isolation, or from relational magnitudes formed by taking ratios of pairs of simple stimuli. Some have proposed that these two types of magnitude are processed by a common system, whereas others have proposed separate systems. To test these competing possibilities, the present study examined the developmental trajectories of simple and relational magnitude discrimination and relations among these abilities for preschoolers (n = 42), 2nd-graders (n = 31), 5th-graders (n = 29), and adults (n = 32). Participants completed simple magnitude and ratio discrimination tasks in four different nonsymbolic formats, using dots, lines, circles, and irregular blobs. All age cohorts accurately discriminated both simple and ratio magnitudes. Discriminability differed by format such that performance was highest with line and lowest with dot stimuli. Moreover, developmental trajectories calculated for each format were similar across simple and ratio discriminations. Although some characteristics were similar for both types of discrimination, ratio acuity in a given format was more closely related with ratio acuities in alternate formats than to within-format simple magnitude acuity. Results demonstrate that ratio magnitude processing shares several similarities to simple magnitude processing, but is also substantially different.

Play behaviour, not tool using, relates to brain mass in a sample of birds

Play behaviour and tool using in birds, two well-delineated and amply researched behaviours, have generally been associated with cognitive abilities. In this study, these behaviours were related to relative brain mass in a sample of Australian native birds. Despite suggestive research results so far between cognition and tool using, this study found no significant difference in relative brain mass or in lifespan between tool-using birds and non-tool users. By contrast, in play behaviour, subdivided into social players and non-social players, the results showed statistically very clear differences in relative brain mass between social, non-social and non-players. Social play was associated with both the largest brain mass to body mass ratios and with the longest lifespans. The results show that play behaviour is a crucial variable associated with brain enlargement, not tool using. Since many of the tool using species tested so far also play, this study suggests that false conclusions can be drawn about the connection between tool using and cognitive ability when the silent variable (play behaviour) is not taken into account.

Do infants and preschoolers quantify probabilities based on proportions?

Most statistical problems encountered throughout life require the ability to quantify probabilities based on proportions. Recent findings on the early ontogeny of this ability have been mixed: For example, when presented with jars containing preferred and less preferred items, 12-month-olds, but not 3- and 4-years-olds, seem to rely on the proportions of objects in the jars to predict the content of samples randomly drawn out of them. Given these contrasting findings, it remains unclear what the probabilistic reasoning abilities of young children are and how they develop. In our study, we addressed this question and tested, with identical methods across age groups and similar methods to previous studies, whether 12-month-olds and 3- and 4-years-olds rely on proportions of objects to estimate probabilities of random sampling events. Results revealed that neither infants nor preschoolers do. While preschoolers' performance is in line with previous findings, infants' performance is difficult to interpret given their failure in a control condition in which the outcomes happened with certainty rather than a graded probability. More systematic studies are needed to explain why infants succeeded in a previous study but failed in our study.