Exploring the use of projected videos to test action matching from different perspectives in dogs.

Dogs trained with the Do as I Do method can imitate human actions upon request, but their ability to match actions observed from different perspectives remains unknown. The use of 2D video stimuli may enable researchers to systematically manipulate the perspective from which demonstrations are observed, thereby widening the range of methods available to study cognitive skills related to imitation. In this study, we explore the possibility of using 2D stimuli to test action matching in dogs, including when demonstrations are seen from different perspectives. We examined two dogs' imitative performance using videos projected on a screen; while, the owner interacted with the dog remotely through an online meeting software. The dogs were first trained to match human actions seen on a screen frontally, and then were tested when the projected demonstrations were seen frontally, from the side, and from above. Results revealed that both dogs matched the demonstrated actions from frontal and, notably, also from side perspectives, at least to some extent, consistent with familiarity of their daily interactions with humans. However, action matching from an above perspective presented challenges, indicating the potential influence of observational experience and highlighting the importance of perspective manipulation when investigating imitation abilities. These findings show that it is possible to use 2D videos to test imitation in dogs, thereby expanding the potential methodologies to study imitation and other related cognitive skills.

A citizen science model turns anecdotes into evidence by revealing similar characteristics among Gifted Word Learner dogs.

Dogs that have a vocabulary of object labels (Gifted Word Learner dogs-GWL dogs) have great potential as a comparative model for studying a variety of cognitive mechanisms. However, only a handful of studies, with a small sample size of 1 or 2 dogs, have examined this phenomenon. GWL dogs appear to share many of the same distinctive characteristics, but due to their rarity, it is not clear if these similarities are only anecdotal or indeed reflect characteristics that are similar in these rare individuals. Here we present the first study conducted on a relatively large sample of 41 GWL dogs that were recruited and tested using a citizen science model. After testing the dogs' receptive vocabulary of toy names, we asked the owners to complete a questionnaire about their and their dog's life experiences. Our findings highlight several characteristics that are shared among most GWL dogs, such as their learning speed, their large vocabulary, and that they learned the names of the toys spontaneously, without the explicit intent of their owners. Our findings validate previous anecdotal evidence on common characteristics of GWL dogs and supply additional support to the hypothesis that these dogs represent a unique group of dogs.

Spontaneous action matching in dog puppies, kittens and wolf pups.

We investigated the spontaneous tendency of dog puppies, kittens and wolf pups to match their behaviour to actions demonstrated by a human, in the absence of food reward. Based on dogs' inherent sociality and domestication history, we predicted that the tendency to match human actions is more pronounced in this species than in the other two. To test this, we exposed N = 42 dog puppies, N = 39 kittens and N = 8 wolf pups to ostensive human demonstrations of an object-related action. We found that dog puppies paid more attention to the demonstration than kittens and wolf pups. Dog puppies and wolf pups matched the demonstrated actions in more trials than kittens. Dog puppies also tended to reproduce the human demonstration that differed from the action they typically preformed in the absence of demonstration. These results support that dog puppies show a tendency to attend to humans and conform their behaviour to human demonstrations in the absence of extrinsic food rewards. This spontaneous tendency is also relevant for practical applications, by providing the basis to devise puppy-training methods that rely less on food rewards, and instead exploit puppies' natural predisposition for social learning.

Investigating responses to object-labels in the domestic dog (Canis familiaris).

Since the dawn of comparative cognitive research, dogs were suspected to possess some capacity for responding to human spoken language. Neuroimaging studies have supported the existence of relevant mechanisms, but convincing behavioral performance is rare, with only few exceptional dogs worldwide demonstrating a lexicon of object-labels they respond to. In the present study we aimed to investigate if and how a capacity for processing verbal stimuli is expressed in dogs (N = 20), whose alleged knowledge of verbal labels is only backed-up by owner reports taken at face value, and concerning only a few words (on average 5). Dogs were tested in a two-choice paradigm with familiar objects. The experiment was divided into a cue-control condition (objects visible to the owner vs. shielded by a panel, thereby controlling the owner's ability to emit cues to the dog) and a response type condition (fetching vs. looking). Above chance performance in fetching and looking at the named object emerged on the level of the sample as a whole. Only one individual performed reliably above chance, but the group-level effect did not depend on this data point. The presence of the panel also had no influence, which supports that performance was not driven by non-verbal cues from the owners. The group-level effect suggests that in typical dogs object-label learning is an instable process, either due to the animals primarily engaging in contextual learning or possibly analogous to the early stages of implicit, statistical learning of words in humans and opposed to the rapid mapping reported in exceptional dogs with larger passive vocabulary.

Multisensory mental representation of objects in typical and Gifted Word Learner dogs.

Little research has been conducted on dogs' (Canis familiaris) ability to integrate information obtained through different sensory modalities during object discrimination and recognition tasks. Such a process would indicate the formation of multisensory mental representations. In Experiment 1, we tested the ability of 3 Gifted Word Learner (GWL) dogs that can rapidly learn the verbal labels of toys, and 10 Typical (T) dogs to discriminate an object recently associated with a reward, from distractor objects, under light and dark conditions. While the success rate did not differ between the two groups and conditions, a detailed behavioral analysis showed that all dogs searched for longer and sniffed more in the dark. This suggests that, when possible, dogs relied mostly on vision, and switched to using only other sensory modalities, including olfaction, when searching in the dark. In Experiment 2, we investigated whether, for the GWL dogs (N = 4), hearing the object verbal labels activates a memory of a multisensory mental representation. We did so by testing their ability to recognize objects based on their names under dark and light conditions. Their success rate did not differ between the two conditions, whereas the dogs' search behavior did, indicating a flexible use of different sensory modalities. Little is known about the cognitive mechanisms involved in the ability of GWL dogs to recognize labeled objects. These findings supply the first evidence that for GWL dogs, verbal labels evoke a multisensory mental representation of the objects.

An exploratory analysis of head-tilting in dogs.

Little is known about head-tilts in dogs. Based on previous investigations on the head turning and the lateralised brain pattern of human speech processing in dogs, we hypothesised that head-tilts may be related to increased attention and could be explained by lateralised mental functions. We observed 40 dogs during object-label knowledge tests and analysed head-tilts occurring while listening to humans requesting verbally to fetch a familiar toy. Our results indicate that only dogs that had learned the name of the objects tilted their heads frequently. Besides, the side of the tilt was stable across several months and tests. Thus, we suggest a relationship between head-tilting and processing relevant, meaningful stimuli.

Acquisition and long-term memory of object names in a sample of Gifted Word Learner dogs.

Dogs with a vocabulary of object names are rare and are considered uniquely gifted. In a few cases, these Gifted Word Learner (GWL) dogs have presented cognitive skills that are functionally similar to those of human infants. However, the acquisition rate of new object names and the ability of GWL dogs to form long-term memories of those is unknown. In this study, we examine the ability of six GWL dogs to acquire the names of new objects in a short period and to retain those in their long-term memory without post-acquisition exposures. In Experiments 1 and 2, the dogs were tested on their ability to learn, during social interactions with their owners, the names of 6 and 12 new toys respectively, in one week. In Experiments 3 and 4, the dogs' memory of these objects was tested after one and two months. GWL dogs typically learned the names of the new objects and remembered those. We suggest that dogs with knowledge of object names could be a powerful model for studying mental mechanisms related to word acquisition in a non-human species.

Word learning dogs (Canis familiaris) provide an animal model for studying exceptional performance.

Exceptional performance is present in various human activities but its origins are debated and challenging to study. We report evidence of exceptional performance and qualitative variation in learning object-names in dogs. 34 naïve family dogs and 6 knowledgeable individuals that knew multiple toy names, found in 2 years of search around the Globe, were exposed to 3 months of training to learn two novel toy-names and were tested in two-way choice tests. Only 1 naïve and all 6 knowledgeable dogs passed the tests. Additionally, only these dogs learned at least 10 new toy names over the 3 months, showing qualitative variation in this capacity. Although previous object-name knowledge could provide an explanation for the superior performance of the knowledgeable dogs, their rarity and the absence of previous training of this skill point to exceptional giftedness in these individuals, providing the basis to establish dogs as a model-species for studying talent.

Rapid learning of object names in dogs.

Learning object names after few exposures, is thought to be a typically human capacity. Previous accounts of similar skills in dogs did not include control testing procedures, leaving unanswered the question whether this ability is uniquely human. To investigate the presence of the capacity to rapidly learn words in dogs, we tested object-name learning after four exposures in two dogs with knowledge of multiple toy-names. The dogs were exposed to new object-names either while playing with the objects with the owner who named those in a social context or during an exclusion-based task similar to those used in previous studies. The dogs were then tested on the learning outcome of the new object-names. Both dogs succeeded after exposure in the social context but not after exposure to the exclusion-based task. Their memory of the object-names lasted for at least two minutes and tended to decay after retention intervals of 10 min and 1 h. This reveals that rapid object-name learning is possible for a non-human species (dogs), although memory consolidation may require more exposures. We suggest that rapid learning presupposes learning in a social context. To investigate whether rapid learning of object names in a social context is restricted to dogs that have already shown the ability to learn multiple object-names, we used the same procedure with 20 typical family dogs. These dogs did not demonstrate any evidence of learning the object names. This suggests that only a few subjects show this ability. Future studies should investigate whether this outstanding capacity stems from the exceptional talent of some individuals or whether it emerges from previous experience with object name learning.

Did we find a copycat? Do as I Do in a domestic cat (Felis catus).

This study shows evidence of a domestic cat (Felis catus) being able to successfully learn to reproduce human-demonstrated actions based on the Do as I Do paradigm. The subject was trained to reproduce a small set of familiar actions on command "Do it!" before the study began. To test feature-contingent behavioural similarity and control for stimulus enhancement, our test consisted of a modified version of the two-action procedure, combined with the Do as I Do paradigm. Instead of showing two different actions on an object to different subjects, we applied a within-subject design and showed the two actions to the same subject in separate trials. We show evidence that a well-socialized companion cat was able to reproduce actions demonstrated by a human model by reproducing two different actions that were demonstrated on the same object. Our experiment provides the first evidence that the Do as I Do paradigm can be applied to cats, suggesting that the ability to recognize behavioural similarity may fall within the range of the socio-cognitive skills of this species. The ability of reproducing the actions of a heterospecific human model in well-socialized cats may pave the way for future studies addressing cats' imitative skills.

Mental representation and episodic-like memory of own actions in dogs.

We investigated whether dogs remember their spontaneous past actions relying on episodic-like memory. Dogs were trained to repeat a small set of actions upon request. Then we tested them on their ability to repeat other actions produced by themselves, including actions performed spontaneously in everyday situations. Dogs repeated their own actions after delays ranging from a few seconds to 1 hour, with their performance showing a decay typical of episodic memory. The combined evidence of representing own actions and using episodic-like memory to recall them suggests a far more complex representation of a key feature of the self than previously attributed to dogs. Our method is applicable to various species, paving the way for comparative investigations on the evolution and complexity of self-representation.

Depths and limits of spontaneous categorization in a family dog.

Categorization has been tested in non-human animals after extensive training procedures under laboratory conditions and it is assumed that in non-primate species categorization relies on perceptual similarity. We report evidence of the ability to categorize objects in absence of specific training in a family dog with vocabulary knowledge of multiple toys, including exemplars of 4 categories. Our experimental design was devised to test categorization in absence of specific training and based on the spontaneously learned vocal labels of the categories, a condition that mirrors human studies more than previous experiments on non-human animals. We also observed that the dog's categorization skills were more accurate when, prior to the categorization test, she was given the opportunity to play with the novel exemplars, suggesting that category representations arise not only from physical resemblance, but also from objects' affordances (function).

Social behavior in dog puppies: Breed differences and the effect of rearing conditions.

INTRODUCTION: Most of the studies investigating the effect of early rearing environment in dogs used laboratory dogs and reported that early experiences markedly affect the puppies' behavior. However, the subjects of these experiments cannot be considered as representatives of family dogs. METHODS: In this study, we investigated whether different raising conditions shape social behavior toward humans in 8-week-old family dog puppies of two breeds, Labrador and Czechoslovakian wolf dog. The puppies were tested in a series of tests that represented typical situations of family dogs. RESULTS: We found that Czechoslovakian wolf dog puppies were more active than Labrador puppies in general, as they were more likely to explore the environment and the objects and spent more time doing so. Tendency to gaze at humans also varied between breeds, but in a context-specific way. Additionally, puppies housed separately from their mother interacted more with toys, puppies housed in a kennel tended to stay closer to the experimenter than puppies raised in the house, and puppies housed in a kennel tended to stay in the proximity of the experimenter more than puppies raised in the house. CONCLUSIONS: Our results provide evidence for early keeping conditions influencing social behavior and also highlight breed differences in puppies' behavior. Whether these differences are due to different developmental patterns and/or behavioral predispositions remains to be explored.

Social learning of goal-directed actions in dogs (Canis familiaris): Imitation or emulation?

A goal-directed action is composed of two main elements on which the observer may focus its attention: the movement performed (i.e., the action) and the outcome (i.e., the goal). In a social learning situation, consequently, the observer may imitate the action of the model or emulate the result of its action. In humans and primates, the tendency to selectively engage in any of these two processes is considered to be dependent upon the availability and saliency of information about the goal, implying the capacity to recognize the goals of others' actions. Dogs are skillful in learning socially from humans, and, when trained with the Do as I Do method, they imitate human actions. Here, we tested trained dogs for engaging in imitation or emulation based on information about the goal. We found that dogs observing the demonstration of an object-related action in the absence of a clear goal tended to solve the task by matching the body movement of the human demonstrator. In contrast, when they could observe the exact same movement, but the goal was apparent, they attempted to solve the task by their own means, that is, by emulation, instead of imitating the demonstrated action. These results provide experimental evidence for dogs being able to recognize the goals of others and adjust their behavior accordingly, relying flexibly on imitation or emulation. (PsycINFO Database Record (c) 2019 APA, all rights reserved).

Social learning from conspecifics and humans in dog puppies.

Social learning is especially advantageous for young individuals because it reduces the risks of trial-and-error learning, while providing an efficient way of acquiring information. Whereas adult dogs are known to excel in social learning skills, the ontogeny of this process has been mainly overlooked. The focus of our study was to investigate whether the capacity of social learning is already developed in dogs at an early age. We tested 8-week-old dog puppies on their ability to learn socially to open a puzzle box baited with food and on their capacity to retain the acquired information in their memory. Puppies were tested with conspecific and human demonstrators. We further investigated on whether demonstrations performed by their mother or by an unfamiliar conspecific model affected puppies' learning trend differently. We found that social learning skills are present in 8 weeks old puppies and they remembered this experience for 1 hour. Puppies learned to solve the task from both conspecific and human demonstrators, thereby endorsing dogs' flexibility in learning from different social partners. Unexpectedly, puppies were more likely to learn from unfamiliar conspecifics than from their mother, probably as a result of greater attention payed to the demonstration performed by the unfamiliar model.

Sex differences in dogs' social learning of spatial information.

We used a modified version of the Do as I Do paradigm to investigate dogs' preference and flexibility in the acquisition of different types of spatial information in social learning situations. When required to match the location of the demonstration, dogs (N = 16) preferentially relied on allocentric information, i.e., the relationship between the location of the demonstration and the various objects surrounding it. However, when allocentric cues were inadequate to solve the task, dogs learned to rely on egocentric information, i.e., the direction-left/right-taken by the human demonstrator. The ease of resorting to the non-preferred egocentric strategy was sex-dependent with males acquiring the egocentric strategy in fewer trials than females. This study shows that dogs rely preferentially on allocentric cues when recalling socially acquired spatial information. However, they are impressively flexible in switching to egocentric strategies according to the task requirements. Whether preference for the allocentric strategy in processing spatial information is embedded in the nature of social learning or restricted to our paradigm is an open question. This study also supports the idea that sex differences in cognitive domains are widespread among mammals and calls for further investigations aimed at shedding light on the evolution, function and mechanisms of such differences.

Recall of Others' Actions after Incidental Encoding Reveals Episodic-like Memory in Dogs.

The existence of episodic memory in non-human animals is a debated topic that has been investigated using different methodologies that reflect diverse theoretical approaches to its definition. A fundamental feature of episodic memory is recalling after incidental encoding, which can be assessed if the recall test is unexpected [1]. We used a modified version of the "Do as I Do" method [2], relying on dogs' ability to imitate human actions, to test whether dogs can rely on episodic memory when recalling others' actions from the past. Dogs were first trained to imitate human actions on command. Next, they were trained to perform a simple training exercise (lying down), irrespective of the previously demonstrated action. This way, we substituted their expectation to be required to imitate with the expectation to be required to lie down. We then tested whether dogs recalled the demonstrated actions by unexpectedly giving them the command to imitate, instead of lying down. Dogs were tested with a short (1 min) and a long (1 hr) retention interval. They were able to recall the demonstrated actions after both intervals; however, their performance declined more with time compared to conditions in which imitation was expected. These findings show that dogs recall past events as complex as human actions even if they do not expect the memory test, providing evidence for episodic-like memory. Dogs offer an ideal model to study episodic memory in non-human species, and this methodological approach allows investigating memory of complex, context-rich events. VIDEO ABSTRACT.

Spatial generalization of imitation in dogs (Canis familiaris).

Dogs, like human infants, are able to imitate human actions after a delay (deferred imitation). This study demonstrates that in deferred imitation tasks, dogs can generalize imitation across context modification to a certain extent. Specifically, they can imitate an object-related action if the object used by the demonstrator is displaced to a different location. However, if the object is interchanged with a different one, their imitative performance drops while they show a spatial bias toward the location of demonstration. We used a combination of the 2-action procedure and the "Do as I do" paradigm and displaced the target objects manipulated by the demonstrator, so that, at the time of recall, dogs could only match either the original location of demonstration or the displaced object, but not both. In conditions with a single object present and displaced after the demonstration, dogs matched the action and the object on which it was shown. In conditions with the location of 2 objects interchanged, dogs more likely matched the location. However, when humans provided ostensive cues and pointing gestures to draw the subjects' attention toward the displaced target object, dogs' predisposition to follow human communication outweighed their spatial bias and, as a consequence, their object matching and imitative performance increased. We conclude that object's physical features function as retrieval cues that facilitate recalling the action. In addition to figurative information, dogs rely strongly on spatial information in Do as I do tasks. The results are discussed concerning dogs' representational system of imitative actions. (PsycINFO Database Record

Do as I Did! Long-term memory of imitative actions in dogs (Canis familiaris).

This study demonstrates long-term declarative memory of imitative actions in a non-human animal species. We tested 12 pet dogs for their ability to imitate human actions after retention intervals ranging from 1 to 24 h. For comparison, another 12 dogs were tested for the same actions without delay between demonstration and recall. Our test consisted of a modified version of the Do as I Do paradigm, combined with the two-action procedure to control for non-imitative processes. Imitative performance of dogs remained consistently high independent of increasing retention intervals, supporting the idea that dogs are able to retain mental representations of human actions for an extended period of time. The ability to imitate after such delays supports the use of long-term declarative memory.