Correspondence:
 

Social play: Evolution, cooperation, negotiation, and minds

Jethro (a dog) runs towards Sukie (another dog), stops immediately in front of her, crouches on his forelimbs (bows), wags his tail, barks, and immediately lunges at her, bites her scruff and shakes his head rapidly from side-to-side, works his way around to her backside and mounts her, jumps off, does a rapid bow, lunges at her side and slams her with his hips, leaps up and bites her neck, and runs away. Sukie takes wild pursuit of Jethro and leaps on his back and bites his muzzle and then his scruff, and shakes her head rapidly from side-to-side. They then wrestle with one another and part, only for a few minutes. Jethro walks slowly over to Sukie, extends his paw toward her head, and nips at her ears. Sukie gets up and jumps on Jethro's back and bites him and grasps him around his waist. They then fall to the ground and wrestle with their mouths.

This description of a play encounter between two dogs (it could be other canids, felids, nonhuman primates, or humans) shows that when they engage in social play they perform behavior patterns that are used in other contexts such as aggression, reproduction, and predation. They, and other animals, including humans, also use actions that are important for initiating and maintaining play, in this case "bows." Social play in nonhuman animals (hereafter animals) is usually a cooperative turn-taking venture, and an important question arises, namely, "How do animals negotiate cooperative agreements?" I will consider this and other questions in this essay, for the study of animal play provides access into animals' minds. Available data strongly suggest that play-soliciting actions seem to be used to communicate to others that actions such as biting, biting and shaking of the head from side-to-side, and mounting, are to be taken as play and not as aggression, predation, or reproduction. On this view, bows are performed when the signaler wants to communicate a specific message about her desires or beliefs. While we cannot be sure that two dogs, for example, are engaging in first- or second-order intentional behavior (after Dennett, 1983), some data do suggest this possibility. For example, suppose we wanted to know why Sukie permitted Jethro to nip at her ears. One explanation may be that Sukie believes Jethro is playing. And perhaps Jethro believes that Sukie believes that Jethro is playing. Providing answers to questions such as these is one of the challenges of research in animal cognition.

The category of behavior called social play has challenged students of animal and human behavior for a long period of time (van Hooff, 1973; Bekoff and Byers, 1981; Fagen, 1981, 1993; Burghardt, 1988, 1996a; Rosenberg, 1990; Bekoff and Allen, 1992; Jamieson and Bekoff, 1993; Allen and Bekoff, 1994; Bekoff, 1995a,b; Pellegrini, 1995; Bekoff and Jamieson, 1996). For some, social play was written off as a behavior that was not worth studying, while for others, the difficulty of studying social play in a rigorous manner prompted them to claim that it was worth studying but that a sufficiently "scientific" study of social play could never be done. Others, recognizing that social play behavior was interesting and fun to study embarked on a challenging journey that has yielded valuable data concerned with such questions as why do animals perform behavior patterns that are unique to social play, why do they perform sequences of behavior that incorporate actions from many different contexts, how do animals negotiate cooperation and agreements to participate in social play, how do they develop socially and learn rules of social interaction, how do they ask permission to engage in social play. It also has become at least a little less unorthodox to speculate on the existence of a humorous or comic dimension in the natural lives of animals, and the possible evolutionary implications of such a dimension (R. Fagen, personal communication; Burghardt, 1996b). For example, we can ask how animals' obvious enjoyment of engaging in social play and their persistent efforts in luring others to play influences the development and evolution of this behavioral phenotype.

The tractability of social play as an evolved phenotype also makes it attractive to those interested in learning more about nonhuman and human cognition (Pellegrini, 1995). Understanding activities such as play may also be important for developing new research dealing with comparative approaches to cognition, including building androids (Caudill 1992, p. 5). The broad comparative study of social play also lends itself nicely to Tinbergen's (1951/1989, 1963) suggestions that ethology should be concerned with (at least) the evolution, adaptation, causation, and development of behavior (Jamieson and Bekoff, 1993; Burghardt [1996b] suggests adding a fifth category to Tinbergen's, namely, the pleasurable aspects of behavior), and it is clear that careful studies of social play do inform these areas. Studies of social play also cross disciplinary boundaries -- those interested in social play come from various academic areas (e.g. zoology, psychology, anthropology, computer science, cognitive science, medicine, philosophy, and sociology) and consider questions that deal with topics such as evolution, ecology, development, social communication, neurobiology, and cognition. For example, many scientific researchers provide data essential for understanding social play within the context of the natural history of a species, with respect to evolutionary and development contexts, or with respect to individual differences within a species. These accounts, while useful and stimulating can be pretty dry and unexciting to those interested in larger questions about, say, possible relationships between play and cognition. Philosophers and others not locked into science usually are more willing to talk about animals' beliefs and desires and how these sorts of attributions help us to explain the existence of play in an animal's behavioral repertoire. While I am a fan of dry and laborious analyses of social play, I also am a proponent of letting one's hair down and entertaining challenging ideas that will stimulate further research. Animals obviously enjoy playing and perhaps researchers' enjoyment of studying play will result in our learning more about animal cognition and the evolution of cognition. Lorenz (1996, p. 219) stresses the necessity for "a playful interest in animals," and notes how this attitude was prevalent among early ethologists.

In this paper I consider how analyses of mammalian social play (hereafter play) inform inquiries about the evolution of cognitive mechanisms. Social play has been described in most mammals in which its possible existence has been investigated (Fagen, 1981, chapter 3). Some of the most elaborate and complex forms of play occur in such mammals as the social carnivores (wild and domestic cats, wolves, coyotes, domestic dogs, bears) and nonhuman primates, animals with highly developed cerebral cortices and relatively high encephalization quotients (Bekoff and Byers, 1981; Fagen, 1981; the encephalization quotient is the ratio of actual brain size to brain size expected from the regression equation for the relationship between brain size and body size; see Jerison, 1973). Perhaps it is a combination of these and possibly other factors that have played a significant role in the ability of individuals to engage in play -- to have the cognitive skills to negotiate cooperative agreements, to ask for permission to engage in a specific activity, to encapsulate one's behavior within a pretend context, and perhaps even to make mental attributions to others. With the goal of learning more about how studies of animal cognition can inform studies of human cognition, mammalian social play is a good behavioral phenotype on which to concentrate for when animals play they typically perform behavior patterns that are used in other contexts (e.g. predation, aggression, or reproduction), and yet these behavior patterns take on a different meaning when encapsulated in the context of play. For example, when Jethro bit Sukie and shook his head rapidly from side-to-side, he did not continue to the point of eliciting submissive behavior from her, and she did not squeal and roll over submissively or run away. He also did not attempt to copulate with her when he mounted her from behind, and she did not try to shake him off of her back.

How important it is to negotiate play and to agree that play is the name of the game? Very much so. For, when animals play, behavior patterns for the most part are "borrowed" from other contexts, and individuals need to be able to tell one another that they do not want to eat, fight with, or mate with the other individual(s), but rather, they want to play with them. In most species in which play has been observed, specific actions have evolved that are used to initiate ("I want to play with you") or to maintain ("I still want to play with you regardless of what I just did to you or regardless of what I am going to do to you") play. These actions seem to function in negotiations between participants, the result of which is that they foster an agreement to engage in play rather than to partake in aggression or predation, for example. There is no solid evidence that animals invite others to play and then exploit them (Bekoff, 1978). Furthermore, self-handicapping (e.g. Altmann, 1962; Watson and Croft, 1996) and role reversals have also been observed, in which, for example, dominant individuals allow themselves to be dominated only in the context of play.

In considering how play is initiated and maintained, I will discuss in varying depth issues including the sorts of information that are shared during play, what cognitive psychologists who study humans can learn from cognitive ethologists who study other animals, and what play can tell us about the emergence of mind in animals. For example, pretend play, in which an individual creates an imaginary situation or distorts reality, has attracted the attention of many interested in the development of human behavior (see Leslie, 1987 and Perner, 1993, for a detailed discussion of the different sorts of pretend play), and also seems to occur in nonhumans. Dogs and cats often chase their tails and bite them as if the tail was prey or a competitor, and they frequently attempt to kill their food bowl or mate with it. Some dominant animals also engage in self-handicapping or role-reversing during play, in which they seem to pretend that they are not really dominant, perhaps to get others who would otherwise not play with them to play with them (it is rare for the self-handicapping individual to then attempt to dominate their partner; Watson and Croft, 1996). While there are few data that are concerned with the evolutionary precursors to social pretend play in human children (for further discussion see Mitchell, 1990 and Jensvold and Fouts, 1993), it may be the case that when a wolf is playing with another wolf by using behavior patterns that are typically used in sexual or predatory encounters, she is thinking about mating or eating her partner and pretending to do so, although the end result of the interaction -- play and not killing or mating -- do not allow this explanation to be made with any degree of certainty. However, requiring that the individual is thinking about the activity in which s/he is pretending to engage is not a criterion that is applied to human children, and there is no reason to require it in the case of animals (in which its occurrence could not be discovered by direct questioning: Were you thinking about mating with your food bowl or with your play partner?) Perhaps neurobiological analyses in nonhumans and humans would help us along here by showing that similar EEG patterns are produced (multiply realized) in play and in what are deemed to be real instances of mating, preying, or fighting.

Before looking at the data, one points needs to be addressed. Useful analyses of social play requires words that some deride as being anthropomorphic (animals enjoy playing, they gain pleasure from playing, they want to play, they believe that others want to play with them). The point here is that these critics hold that words such as "wants," "believes," or "intends" apply solely to humans. While some categorically dismiss any analyses of animal behavior that use anthropomorphic terms, others believe that careful use of anthropomorphism can be useful for the study of animal behavior and animal minds (for discussion see Burghardt, 1991 and Bekoff and Allen, 1996). I believe that detailed comparative analyses of social play can be useful for informing questions about the evolution of animal minds -- do individuals have beliefs, do they have desires, do they make attributions of intentions to others, do they represent internal states of others, what is the content of their mental states? It seems clear that without a cognitive vocabulary, it would be impossible to discuss social play in any useful way (Mitchell, 1990; Jamieson and Bekoff, 1993; Bekoff, 1995a). For when most animals play they use specific signals to signal their intentions -- their desire to play and their beliefs that if they perform a particular behavior then play will occur -- and also perform behavior patterns that are typically performed in other contexts, but whose meaning is changed in the context of play. The richness and diversity of social play can only be captured by the use of mental terms. A detailed consideration of some selected aspects of mammalian social play can inform philosophers about important issues in philosophy of mind including naturalistic (and other) accounts of intentionality, representation, and communication, and perhaps cause them to refine some of their ideas.

What is social play?

From early serious efforts to come to terms with social play behavior, a number of definitions were offered (Bekoff and Byers, 1981; Fagen, 1981; Martin and Caro, 1985), and these definitions provided the framework for numerous comparative and evolutionary studies. Nowadays, it is difficult to pick up a journal that deals with the comparative study of animal behavior and not find at least one article that concerns itself with the structure or function of social play (predominantly in mammals). While social play is easy to recognize, there does not seem to be any stipulative definition that would apply to all instances of what is called "social play" in the diverse species in which the activity has been observed. However, lacking a rigorous definition need not be an impediment to conducting solid research; requiring a rigorous definition prior to empirical research may unreasonably require possession of knowledge that must first be gained by empirical research (Allen and Bekoff, 1994). To get the ball rolling, I will use a general definition that has weathered the years (Bekoff and Byers, 1981, pp. 300-301): "Play is all motor activity performed postnatally that appears (our emphasis) to be purposeless, in which motor patterns from other contexts may often be used in modified forms and altered temporal sequencing. If the activity is directed toward another living being is called social play." (Martin and Caro, 1985, p. 65, provide a slightly modified version of this definition.) Note that this is a positive definition, in that play is defined in terms of what it is, and not in terms of what it is not. For years, it was trendy to refer to social play as being not aggression, not predation, or not reproduction, but these sorts of definitions proved to be of little or no value to those who were interested in learning more about play. Note also that the definition centers on the structure of play sequences -- what animals do when they play -- and not on possible functions of play. Suffice it to say here, functional accounts of play, while yielding interesting comparative information, would really muddy the definitional waters. For analyses of the possible functions of play have resulted in unconsolidated mixes of speculations (R. Fagen, personal communication), and it is unlikely that there is one single reason for why all animals play. Suffice it to say, there seem to be many possible and not mutually exclusive functions of play (e.g. socialization, physical training, cognitive development, energy regulation) depending on the species being studied and perhaps the ages and sexes of the participants and the location in which they are playing, a discussion of which goes beyond the bounds of this essay (for reviews see Bekoff and Byers, 1981; Fagen, 1981; Martin and Caro, 1985; Burghardt, 1988, 1996a). Furthermore, functional accounts of play that appeal to possible benefits and costs of playing are fraught with difficulties as are most discussions of function in terms of potential benefits and costs (Allen and Bekoff, 1995a). With respect to play, both short-term and long-term benefits and costs need to be considered, and the necessity to account for benefits and costs over long periods of time for individuals of long-lived species in which play is a notable activity, is not an exercise that would be recommended to nontenured professors!

Despite its easy to recognize features, social play has been a very difficult behavioral phenotype with which to deal rigorously. While there are very few people who would claim that animals do not engage in this easily recognizable activity, broad comparative analyses of play are still forthcoming, although the data base has greatly increased over the past five years (Burghardt, 1996a). Not only does play between species vary in structure (behavior patterns used and their temporal sequencing), but there also are marked individual differences within species that make generalizations difficult and tenuous. This variability can be the result of numerous factors, including the ages of the participants, their sexes, their relative social ranks, their experience with one another as play partners, their energy levels, their level of physical (aerobic and anaerobic) fitness, food abundance, how permissive are their caretakers, and where they are playing. Indeed, it is the flexibility and versatility of play makes it a good candidate for comparative and evolutionary cognitive studies. As Allen (1996) noted, "Behavioral flexibility is relevant to mentalistic attributions because it is connected to an organism's monitoring of its own performance. An organism that cannot detect when its states misrepresent its environment will be limited to adjusting its behavior only when the proximal causes of those states are removed." Indeed, behavioral flexibility (and communication) and an individual's ability to adjust to variations in environmental situations is the main criterion to which Griffin (1992) appeals in his discussions of consciousness in many nonhuman animals. Here I will not take on issues of animal consciousness, for this sort of endeavor would take more far afield of where I want to go, and because it would probably a journey with little or no direction because of all of the problems associated with dealing with consciousness in animals (including humans). However, I will appeal to the behavioral flexibility that is apparent in the social play of many mammals to discuss different aspects of how animals initiate and maintain social play, and how these patterns of behavior are useful for learning about the cognitive abilities of animals and the evolution of cognition.

Play signals and intentional behavior: Do animals negotiate agreements?

" . . . cooperation merely depends upon the behaviour of one animal serving as a stimulus that elicits a certain response from the other." (Pearce, 1987, p. 261)

"Levels of cooperation in play of juvenile primates may exceed those predicted by simple evolutionary arguments . . . " (Fagen 1993, p. 192)

While some of the issues raised by skeptics deserve serious consideration, a full discussion is beyond the scope of this essay (see Bekoff and Allen, 1996 and Allen and Bekoff 1995b, and references therein). The point here seems to be a simple one. In some (perhaps most) instances of social play, I have argued (Bekoff, 1978, 1995a), based on empirical data, that the participants need to cooperate and to negotiate agreements so that they can cooperate in ongoing social play. And, after they agree to play there seem to be social regulations and social obligations to maintain the play mood. Furthermore, there are data that suggest that animals often seek permission to play with another animal, by performing behavior patterns that indicate that play is desired and not another activity. For example, in the self-handicapping (e.g., inhibited biting or clawing; Steiner, 1971; Watson and Croft, 1996) and role-reversing (being chased rather than chasing) in which dominant animals engage to get more subordinate individuals to play, the dominant individuals are seeking permission to engage in play with animals who might otherwise avoid interacting with them. The end result is that the two individuals agree to cooperate, with self-handicapping and role-reversing being a form of negotiating this agreement. These data are important to consider in discussions of the evolution of cognition for they tell us that certain cognitive abilities must be present even before an individual has the linguistic sophistication of a human. Let's look at some data in more detail.

The dance of play: Initiating, maintaining, and agreeing to play

Because play is composed of actions that are also used in other contexts, an individual needs to be able to communicate to potential play partners that he is not trying to dominate them, eat them, or mate with them. Rather, he is trying to play with them. Behavioral observations of many animals who engage in social play suggest that they desire to do so and believe that their thoughts of the future - how the individuals to whom their intentions are directed would be likely to behave - would be realized if they clearly communicated their desires to play using signals that in some cases seem to have evolved specifically to communicate play intention (for further discussion see Tomassello et al., 1985; Tanner and Byrne, 1993; Bekoff, 1995). On this view, play is seen as a cooperative enterprise, the result of negotiations that allow play to occur and decrease the likelihood that aggression or predation will ensue. While other animals might also cooperate, including those for whom we are hard-pressed to grant the presence of sophisticated cognitive abilities (e.g., cleaning fish), the flexibility of play and the ease with which even two strange animals are able to play after only brief introductions to one another, lead to the conclusion that interactions between players are not merely the result of (somewhat hard-wired and inflexible) coevolved systems but rather the result of ongoing negotiations that require careful assessments throughout an encounter. It seems to be the flexibility of play that is important in allowing animals to negotiate the desired outcome. (Given the state of knowledge concerning animals for whom we are reluctant to grant the presence of sophisticated cognitive abilities, it might be premature to write them off as not having well-developed cognitive skills that are used for different types of negotiations.)

Do animals cooperate in play, and how do they agree or negotiate what is they are going to do? To begin to answer these questions, let's consider in more detail the question of whether or not signals that appear to be used to communicate play-intention (play-soliciting signals) to other individuals could foster the cooperation among participants that is necessary for play to occur. Such play-soliciting signals seem to transmit messages such as "what follows is play," "this is still play despite what just happened or is going to happen," or "let's play again, wasn't it fun" (Bekoff, 1995a). Supporting evidence concerning the importance of play signals for allowing cooperative social play to occur comes from studies in which it is shown that play-soliciting signals show little variability in form or temporal characteristics and that they are used almost solely in the context of play. For example, one action that is commonly observed in the context of social play in canids and some other mammals is the highly stereotyped "bow." In certain canids, the bow seems to function to permit recipients to engage (or to continue to engage) in social play (Bekoff, 1977, 1995a). Also, the first bows that very young canids have been observed to perform are highly stereotyped, and learning seems to be relatively unimportant in their development. These features of bows can be related to the fact that when engaging in social play, canids typically use action patterns that are also used in other contexts such as predatory behavior, agonistic encounters, or mating, where misinterpretation of play intention could be injurious.

How are agreements reached in the context of social play? How is turn-taking by individuals accomplished? In most species in which play has been described, play-soliciting signals appear to foster some sort of cooperation between players so that each responds to the other in a way consistent with play and different from the responses the same actions would elicit in other contexts (Bekoff, 1975, 1978); play-soliciting signals provide aid to the interpretation of other signals by the receiver (Hailman, 1977, p. 266). For example, in coyotes, the response to a threat gesture after a play signal had immediately preceded the threat or a play signal had been performed in the beginning of an interaction, is different from the response to threat in the absence of any preceding play signal (Bekoff, 1975). The play signal somehow altered the meaning of a threat signal by establishing (or maintaining) a "play atmosphere" Unfortunately, there have been no other similar quantitative analyses, but observations of play in diverse species support the idea that play signals can, and do, serve to establish a social context that allows play to occur and alters the significance of behavior patterns that are borrowed from other contexts and used in social play (see below)

Some other characteristics of bows and also some of the properties of social play support a cognitive explanation of play, and can be used to inform questions about their use for initiating and maintaining social play. For example, bows themselves occur throughout play sequences, but usually at the beginning or towards the middle of playful encounters. Intentional explanations of the way in which bows are used in ongoing play are helpful for understanding how animals negotiate agreements, especially because of the apparent flexibility of how bows are used. For example, in a detailed analysis of the form and duration of bows (Bekoff, 1977), it was shown that bows are highly stereotyped and that bows were always less variable when performed at the beginning, rather than in the middle of, ongoing play sequences. These data were used to argue that bows were important in the initiation of play. Other observations also support this idea. For example, one dominant female coyote pup was successful in initiating chase play with her subordinate brother on only 1 of 40 (2.5%) occasions, her lone success occurring on the only occasion in which she had signaled previously with a bow (Bekoff, 1975). There may be more variability for bows performed during play bouts when compared with bows performed at the beginning of play sequences because of (1) fatigue, (2) the fact that animals are performing them from a wide variety of preceding postures, or because (3) there is less of a need to communicate that "this is still play" than there is when trying to initiate a new interaction. Data for distinguishing among these three possible explanations do not currently exist.

Play signals are also used to maintain social play in situations where the performance of a specific behavior during a play bout could be misinterpreted. This is of interest because it suggests that an individual knows what he is going to do next and possibly the effect his behavior will have on the recipient. In a recent study (Bekoff, 1995a) I hypothesized that if bites accompanied by rapid side-to-side shaking of the head or other behavior patterns typically used in other contexts such as aggression or predation could be or were misread by the recipient and could result in a fight, for example, then the animal who performed the actions that could be misinterpreted might have to communicate to its partner that this action was performed in the context of play and was not meant to be taken as an aggressive or predatory move. On this view, bows would not occur randomly in play sequences; the play atmosphere would be reinforced and maintained by performing bows immediately before or after actions that could be misinterpreted. These ideas had not previously been analyzed empirically. I found that the distribution of bows during play that involves these actions is not random; instead they tend to occur immediately before or after a potentially misinterpretable action (for descriptions see Bekoff, 1974 and 1995a; Hill and Bekoff, 1977). The bows seem to serve as a form of punctuation that clarifies the meaning of other actions that follow or precede them. In addition to sending the message "I want to play" when they are performed at the beginning of play, bows performed in a different context, namely during social play, might also carry the message "I want to play despite what I am going to do or just did -- I still want to play" when there might be a problem in the sharing of this information between the interacting animals.

Species differences were also found that can be interpreted by what is known about variations in the early social development of these canids (Bekoff, 1974; see also Feddersen-Petersen, 1991). The interspecific differences also are related to the questions about how individuals monitor their own behavior -- what they might know about themselves and others. For example, infant coyotes are much more aggressive and engage in significantly more rank-related dominance fights than either the infant (or adult) dogs or the infant wolves who were studied. During the course of my study, no consistent dominance relations were established in either the dogs or the wolves, and there were no large individual differences among the play patterns that were analyzed in this study. Social play in coyotes typically is observed only after dominance relationships have been established in paired interactions. Coyotes appear to need to make a greater attempt to maintain a play atmosphere, and indeed, they seem also to need to communicate their intentions to play before play begins more clearly than do either dogs or wolves who have been studied (Bekoff, 1975, 1977). Indeed, as mentioned above, one dominant female coyote pup was successful in initiating chase play with her subordinate brother on only 1 of 40 occasions, her lone success occurring on the only occasion in which she had signaled previously with a play bow (Bekoff, 1975). Furthermore, subordinate coyote infants are more solicitous and perform more play signals later in play bouts. These data suggest that bows are not non-randomly repeated merely when individuals want to increase their range of movement or stretch their muscles. However, because the head of the bowing individual is usually below that of the recipient, bowing may place the individual in a non-threatening (self-handicapping, see below) posture. Self-handicapping may be explained by an individual's assessing a situation and deciding what she has to do to enable play to continue, and suggests that one is aware of how others will interpret their behavior (e.g. Watson and Croft, 1996). Tanner and Byrne (1993) observed a captive female lowland gorilla concealing her play face, and concluded that the gorilla was aware of her spontaneous facial expression and the consequence it entailed, namely that play would follow within a few seconds.

In summary, detailed data and careful observations of a wide variety of animals leads to the conclusion that play-soliciting signals appear to foster cooperation between players so that each responds to the other in a way consistent with play and different from the responses the same actions would elicit in other contexts (see Thompson, 1996 for a comprehensive list of play solicitation behaviors in primates, rodents, insectivores, carnivores and ungulates). It is important to remember that cues other than those provided by visible actions may also be important in the solicitation of play (Fagen, 1981). In his analysis of play captive common marmosets (Callithrix jacchus), Chalmers (1980) identified what he called "play markers" (e.g., the play-face), behavior patterns that always appear playful -- their existence is considered sufficient but not necessary for regarding behavior as playful. He also found that animals were more persistent in bouts with markers and that bouts with markers lasted longer than bouts without markers. Play markers have also been observed in pigs (Newberry, Wood-Gush, and Hall, 1988). The information that is shared by use of play signals or play markers concerns the readiness of an individual or individuals to engage in social play. Play can be a risky affair (Fagen, 1981; Caro, 1995), but also is an important activity for young and old animals alike, and by using play signals individuals can engage in and enjoy the activity. Perhaps it is the case that playing has benefits that are important for developing organisms as well as adults that makes it unlikely that animals will use play signals to deceive others (Bekoff, 1978).

Information sharing and theories of mind: The usefulness of comparative evidence

Studying social play in nonhuman animals (perhaps even in ants; Darwin, 1871, p. 448) is useful for informing ideas about the evolution of cognition. With respect to the evolution of cognition and ideas about evolutionary continuity (and perhaps questions of homology with respect to structure and function; see Atz, 1970), the question "How might information between senders and recipients be shared?" lends itself nicely to comparative evolutionary studies (despite Ingold's [1988] claim that animals have no thoughts whatsoever and are mindless communicators). For example, it is possible that the recipient shares the intentions (beliefs, desires) of the sender based on the recipient's own prior experiences of situations in which she performed bows. Do available data suggest that some animals who engage in social play have a theory of mind -- can they and do they make attributions of mental states to others? While I realize that I am now stepping out on to thin ice, I want to pursue this discussion in some more depth, for I believe that there are data -- some of which have been discussed here -- that allow a tentative answer of "yes" to these questions.

In a paper on human behavior that has yet to find its way into comparative ethological circles, Gopnik (1993, p. 274) has argued that " . . . certain kinds of information that comes, literally, from inside ourselves is coded in the same way as information that comes observing the behavior of others. There is a fundamental cross-modal representational system that connects self and other." Gopnik (see also Meltzoff and Gopnik, 1993) claims that others' body movements are mapped onto one's own kinesthetic sensations, based on prior experience of the observer, and she supports her claims with discussions of imitation in human newborns that can occur spontaneously at birth. For example, Gopnik wants to know if there is an equivalence between the acts that infants see others do and the acts they perform themselves, and imagines "that there is a very primitive and foundational 'body scheme' that allows the infant to unify the seen acts of others and their own felt acts into one framework" (Gopnik, 1993, p. 276). If by "primitive and foundational" Gopnik means phylogenetically old, then there should be some examples, or at least precursors, of this ability in other animals. Gopnik and her colleague Andrew Meltzoff also consider the possibility that there is "an innate mapping from certain kinds of perceptions of our own internal states . . . In particular, we innately map the body movements of others onto our own kinesthetic sensations. This initial bridge between the inside and the outside, the self and other, underlies our later conviction that all mental states are things both we and others share" (Gopnik, 1993, p. 275). Flanagan (1992, pp. 102ff) also is interested in ways in which mental states can be shared, and introduces the notion of a "mental detector" that is used to detect others' invisible mental states.

How these ideas might apply to nonhuman animals awaits further study, but even if Gopnik's case is based entirely on imitation in human infants, there is ample evidence that nonhuman animals engage in imitation (Byrne, 1995a; see also Galef, 1990/1996, 1996 for critical reviews of the literature). And, there seems to be little reason why, for example, in social play, one dog might not be able to know that another dog wants to play by knowing what she feels like when she performs a play bow. Among the questions that need to be studied in detail is "Does a dog have to have performed a bow (or other action) to know what a bow means to be able to make attributions of mental states to other individuals -- to know about others' beliefs, desires, expectations, feelings, thoughts, and plans? The following two hypotheses would have to be distinguished: (i) viewing a play bow induces a play mood in the recipient because of kinesthetic mapping and (ii) viewing a play bow induces knowledge in the recipient of how the actor feels (Colin Allen, personal communication). With respect to bows, at least, there are data that suggest that there is a genetic component to them; the first bows that are observed to be performed by young canids are highly stereotyped and occur in the correct social context (Bekoff, 1977). Could these data support Gopnik's ideas "that there is a very primitive and foundational 'body scheme' that allows the infant to unify the seen acts of others and their own felt acts into one framework" (Gopnik, 1993, p. 276)? And, if so, how is learning incorporated into the development of social communication skills? Regardless of how nature and nurture mix, sparse evidence at hand supports the view that studies of animal cognition can inform the study of human cognition, and that much more comparative research is needed.

There are other suggestions that Gopnik's ideas might enjoy some support from comparative research on animal cognition. For example, Savage-Rumbaugh (1990, p. 59) noted that "Likewise, if Sherman screams when he is upset or hurt, Sherman may deduce that Austin is experiencing similar feelings when he hears Austin screams. This view is supported by the observation that Sherman, upon hearing Austin scream, does not just react, but searches for the cause of Austin's distress." This cause-effect relationship is generated after sufficient experience -- if an animal screams when he is upset or hurt he may deduce that another is experiencing similar feelings when he hears a scream. Tomasello, Gust, and Frost (1989) also note that some gestures in chimpanzees may be learned by "second-person imitation" -- "an individual copying a behavior directed to it by another individual" (p. 35). They conclude (p. 45) that chimpanzees " . . . rely on the sophisticated powers of social cognition they employ in determining what is perceived by a conspecific and how that conspecific is likely to react to various types of information .. "

In an interesting study of predator detection in mixed flocks of emberizid sparrows, Lima (1995) classified birds into different categories based on whether or not an individual detects (detector) or does not detect (non-detector) an attack on the flock in which it is a member. According to Lima, non-detectors "infer the possibility of an approaching threat based on the occurrence of departures from the flock . . . " (p. 1097). They seem to know that departures by others mean that a threat is impending. One wonders if the ability to make this inference is based on an individual's own knowledge of why it departs in certain contexts. Studies of naive birds, individuals who have no prior experience with threatening situations, might shed light on how this ability is acquired.

Others also have been interested in ways in which an individual might come to attribute states of mind to others. For example, in his discussion of research on the use of mirrors to study self-recognition, Byrne (1995a, p. 117) asks "How do mirror-self-recognizing organisms make the intuitive leap to treating self and other as equivalent in mind?" His discussion bears on some of what seems to be going on during social play and also Gopnik's ideas about how mental states may be shared between two individuals. Byrne considers two possibilities: "A concept of self might be acquired by learning first that mental states are useful concepts in predicting the behaviour of others; then, by taking the others' point of view, the self, too, is viewed as having similar mental attributes. Learning that mental states are useful would seem to require having the concept of mental states already and that might require more than we want to give at the beginning (Colin Allen, personal communication). "Or, an intuitive understanding of self may come first, and later the individual finds it useful to attribute similar mental states to others -- the better to understand their mind and behaviour." Byrne notes that Humphrey (1983) defends this view, by holding that "we treat others as different versions of our own self, and therefore endow them with similar properties -- knowledge, intentions and so on." The situation can become more complex when one engages in social referencing -- "one person using another person's appraisal of a situation in order to form an understanding or interpretation of that situation"

(Feinman, 1982 as cited by Itakura, 1995).

Against narrow primatocentrism: Towards a comparative analysis of mind

"After all, from an evolutionary point of view, there ought to be a high premium on the veridicality of cognitive processes. The perceiving, thinking organism ought, as far as possible, to get things right. Yet pretense flies in the face of this fundamental principle. In pretense we deliberately distort reality. How odd then that this ability is not the sober culmination of intellectual development but instead makes its appearance playfully and precociously at the very beginning of childhood." (Leslie, 1987, p. 412)

" . . . great apes are certainly 'special' in some way to do with mentally representing the minds of others. It seems that the great apes, especially the common chimpanzee can attribute mental states to other individuals; but no other group of animals can do so -- apart from ourselves, and perhaps cetaceans." (Byrne, 1995a, p. 147; my emphases)

Attempts to place humans apart from, and above nonhumans, have in sense backfired, for comparative research in animal cognition has demonstrated evolutionary continuity in many cognitive abilities and has shown how connected humans are to other animals. Also, claims such as Byrnes' in the above quotation simply are premature and may well prove wrong when necessary comparative research is completed. (It is important to note that Byrne actually equivocates in this statement and in others in his otherwise excellent book, and he also realizes that little actually is known about the intellectual skills of, for example, carnivores.) It is a fact that very few nonprimate species have been studied concerning the possibility that they have theories of mind, and species-fair tests need to be developed and applied widely before primatocentric claims that can be used to argue against evolutionary continuity can be rigorously assessed. It is unlikely that methods used to study theories of mind in nonhuman primates can be directly applied to nonprimates -- dogs and wolves may have theories of mind but it is highly unlikely that we will learn about them using methods that are used on primates -- and even among primates it is clear that species-fair tests need to be developed to account for species differences in sensory and motor abilities, social organization, and habitat (Byrne, 1995a).

Suffice it to say, there already is ample evidence that comparative studies of animal cognition can be useful for learning about the evolution of cognition in humans and nonhuman primates. As Watanabe, Lea, and Dittrich (1993, p. 372) note concerning their research on pigeons: "The question is not whether pigeons have been proved beyond a reasonable doubt to possess and use concepts, but whether it has proved fruitful to ask whether they do." Surprises also might be forthcoming. Dharmaretnam and Andrew (1994), in their developmental study of vision in domestic chickens (Gallus gallus domesticus), recently discovered that one cerebral hemisphere seemed to be used by the other. Their important result " . . . shows how phenomena which might have been considered as peculiarly human, and integral to the highest level of cognition, are in fact accessible to study in other vertebrates . . . " (Dharmaretnam and Andrew, 1994, p. 1405).

There is little doubt that a broad comparative and evolutionary cognitive perspective will be very useful in future analyses of social play (Bekoff, 1995b) and other behavioral phenotypes, and that these sorts of studies will be useful for those generally interested in the evolution of cognition. Evolutionary accounts of mental content (Allen, 1992a,b) also are needed. There simply are no substitutes for careful observational and descriptive studies, and also for well informed and well executed experiments. They may be difficult to perform, but difficult should not be read as meaning impossible. Information about continuity of cognitive abilities will help to fill in the huge holes that exist in our current data base. Individual variability also needs to be taken into account, for it would be misleading to conclude, on the basis of the behavior of a few members of a given species, that all members of that species are unable to do something or do not perform specific behavior patterns that indicate well developed cognitive abilities. Much research to date on animal cognition has been done on only a very limited number of species and on only a few members of those species in (often impoverished) laboratory environments (Kamil, 1987/1994), and if research on animal cognition it to inform the study of human cognition, a broader data base and broader evolutionary perspective is needed.

Broad comparative field studies are especially needed, for then we will have a good idea of what animals do in the wild. Byrne (1995b) stresses the importance of using naturally posed problems to learn about cognitive differences among species (see also Bekoff, 1995c, 1996). This knowledge will prevent some from making premature assumptions about what animals can and cannot do (for discussion see Allen and Bekoff, 1996). Furthermore, while there certainly are good reasons to be concerned about the lack of control in field studies, there are also good reasons to worry about there being too much control in captive situations. Overcontrol, resulting in impoverished social environments and also simplified stimulus situations (for example, controlling for possible interactions between visual and auditory stimuli might not allow individuals to have access to a more complex composite stimulus that is needed for eliciting certain types of behavior) might make it impossible to study the problems of interest in the social context in which individuals live (Bekoff, Townsend, and Jamieson, 1994).

In the essay I have used social play as a candidate for further research in comparative animal cognition. I have asked "what is it like to be a playing animal?" For in some instances animals seem to know what actions are likely to violate the terms of play, and they attempt to negotiate with their partner(s) so that play will continue -- they seem to have the capacity to represent the mental states of others and make adjustments in their own behavior. This ability suggests an understanding of obligation or permission structures that may be crucial for understanding human performance on tasks that require more abstract reasoning. When discussing behaviors such as social play, it may be more economical or parsimonious to assume that not everything that an individual needs to be able to do in all situations in which he finds himself is preprogrammed; cognitive explanations can be simpler than cumbersome stimulus-response explanations (de Waal, 1991 and Bekoff and Allen, 1996). Indeed, it is the flexibility and versatility of play (and other behavior patterns; Bekoff, 1995c) that makes it a good candidate for comparative and evolutionary cognitive studies. While general rules of thumb may be laid down genetically during evolution or learned, specific rules of conduct that account for all possible contingencies may be too numerous to be hard-wired. Furthermore, while behavioristic learning schemes appealing to notions such as conditioning, generalizing, and substituting can account to a limited extent for behavioral flexibility, behavioral integration, and the use of internal states and images of absent objects in some organisms (e.g., Holland, 1990), learning at high degrees of abstraction from sensory stimulation seems less amenable to behavioristic analysis (Bekoff and Allen, 1992; see also Kamil 1987/1994). Cognitive models of learning provide explanatory schemes for such cases. It might actually be more parsimonious to appeal to cognitive explanations in terms of accounting for complex patterns of behavior with fewer explanations. And this practice, motivated by research on cognition in nonhuman animals, might find its way into studies of human cognition which often are unnecessarily embellished because of the lofty position in which many researchers place humans.

Where to from here? Open-mindedness is the key. Single failures do not doom a field, not do single successes warrant celebrations. Double-standards that demand that cognitive ethology be more rigorous than other behavioral sciences are unlikely to be productive, especially when these sorts of biases lead to the demise of studies of nonhuman cognition. Double-standards within the field of cognitive ethology also need to dispensed with. Much of what we know about cognition in nonhuman primates is based on anecdotal reports and "scattered revelations" (Byrne, 1995a, p. 182), but for some these sorts of stories carry more weight than they do for, say, domestic dogs or other companion animals. Brown (1996) has posed what he calls the "consistency axiom," in which evaluations of all explanations of behavior are subjected to the same degree of rigor. This seems to be a fair and obvious way to evaluate research. It is unlikely that all of humans' sophisticated cognitive abilities arose de novo. Only time will tell how useful are different approaches. Interaction among proponents of different views are essential if we are to make headway into learning more about animal cognition and how studies of animal cognition are able to inform and to motivate the study of human cognition.

Acknowledgments

I thank Colin Allen, Denise Cummins, Gordon Burghardt and Alison Gopnik for comments on an earlier draft of this paper, and Dale Jamieson, Michael Pereira, Jack Hailman, Susan Townsend, Alex Rosenberg, and some students in my animal behavior classes for discussing some of the ideas contained herein. Others who have helped me along are mentioned in some of my other papers that deal with similar issues.

References

Allen, C. (1992a.) Mental content. British Journal of the Philosophy of Science, 43, 537-553.

Allen, C. 1992b. Mental content and evolutionary explanation. Biology and Philosophy 7:1-12.

Allen 1996, Animal cognition and animal minds. In P. Machamer & M. Carrier (eds.) Philosophy and the Sciences of the Mind: Pittsburgh-Konstanz Series in the Philosophy and History of Science vol. 4. Pittsburgh University Press and the Universitätsverlag Konstanz. (Forthcoming)

Allen, C. , and Bekoff, M. 1994. Intentionality, social play, and definition. Bio. & Phil. 9, p. 63-74.

-- -- & BEKOFF, M. (1995a). Function, natural design, and animal behavior: philosophical and ethological considerations. -- Persp. Ethol. 11: 1-46.

Allen, C., and M. Bekoff. 1995b. Cognitive ethology and the intentionality of animal behaviour. Mind and Language 10: 313-328.

ALTMANN, S. A. (1962). Social behavior of anthropoid primates: Analysis of recent concepts. -- In: Roots of Behavior (E. L. BLISS, ed). Harper, New York.

Atz, J. W. 1970. The application of the idea of homology to behavior. In L. R. Aronson, E. Tobach, D. S. Lehrman, and J. S. Rosenblatt (eds.). Development and evolution of behavior: Essays in honor of T. C. Schneirla. W. H. Freeman and Co., San Francisco. pp. 53-74.

BEKOFF, M. (1974). Social play and play-soliciting by infant canids. -- Amer. Zool. 14, p. 323-340.

Bekoff, M. 1975. The communication of play intention: Are play signals functional? Semiotica 15:231-239.

Bekoff, M. 1977. Social communication in canids: Evidence for the evolution of a stereotyped mammalian display. Science 197:1097-1099.

Bekoff, M. 1978. Social play: Structure, function, and the evolution of a cooperative social behavior. In G Burghardt and M. Bekoff, eds. The Development of Behavior: Comparative and Evolutionary Aspects. New York:Garland.

Bekoff, M. 1995a Play signals as punctuation: The structure of social play in canids. Behaviour 132: 419-429.

Bekoff, M. (1995). Cognitive ethology and the explanation of nonhuman animal behavior. In J.-A. Meyer & H. Roitblat (Eds.), Comparative Approaches to Cognitive Science. Cambridge: MIT Press. pages 119-150.

Bekoff, M. 1995c. Vigilance, flock size, and flock geometry: Information gathering by western evening grosbeaks (Aves, fringillidae). Ethology 99: 150-161.

Bekoff, M. 1996. Cognitive ethology, vigilance, information gathering, and representation: Who might know what and why? Behavioural Processes. 35: 225-237.

Bekoff, M., and Allen, C. 1992. Intentional icons: towards an evolutionary cognitive ethology. Ethology 91:1-16.

Bekoff, M., and Allen, C. 1996. Cognitive ethology: Slayers, skeptics, and proponents. In R. W. Mitchell, N. Thompson, and L. Miles, eds. Anthropomorphism, Anecdote, and Animals: The Emperor's New Clothes? Albany, New York: SUNY Press.

Bekoff, M., and Byers, J. A. 1981. A critical reanalysis of the ontogeny of mammalian social and locomotor play: An ethological hornet's nest. In K. Immelmann, G. W.

Barlow, L. Petrinovich, and M. Main, eds. Behavioral Development: The Bielefeld Interdisciplinary Project. New York:Cambridge University. Press.

Bekoff, M., and Jamieson, D. (eds.) 1996. Readings in Animal Cognition. MIT Press, Cambridge, Massachusetts.

Bekoff, M., Townsend, S. E., & Jamieson, D. (1994). Beyond monkey minds: towards a richer cognitive ethology. Behavioral and Brain Sciences, 17, 571-572.

Brown, A. 1996. The minds of animals: Theoretical foundations of comparative psychology. Ph.D. Dissertation, University of Colorado, Boulder, Colorado.

BURGHARDT, G. M. (1988). Precocity, play, and the ectotherm-endotherm transition: profound reorganization or superficial adaptation. In: Handbook of Behavioral Neurobiology, Vol. 9 (E. M. BLASS, ed). Plenum Publishing Company, New York, pp 107-148.

Burghardt, G. M. 1996a. Play. In G. Greenberg and M. Haraway (eds.). Encyclopedia of comparative psychology. New York: Garland. forthcoming.

Burghardt, G. M. 1996b. Amending Tinbergen: A fifth aim for ethology. In R. W. Mitchell, N. Thompson, and L. Miles, eds. Anthropomorphism, Anecdote, and Animals: The Emperor's New Clothes? Albany, New York: SUNY Press.

Byrne, R. 1995a. The thinking ape: Evolutionary origins of intelligence. Oxford University Press, New York.

Byrne, R. 1995b. Primate cognition: Comparing problems and skills. American Journal of Primatology 37: 127-141.

Caro, T. M. 1995. Short-term costs and correlates of play in cheetahs. Animal Behaviour 49: 333-345.

Caudill, M. 1992. In Our Own Image: Building An Artificial Person. New York:Oxford University Press.

Chalmers, N. R. 1980. The ontogeny of play in feral olive baboons. Animal Behaviour 28: 570-585.

de Waal, F. B. (1991). Complementary methods and convergent evidence in the study of primate social cognition. Behaviour, 118, 297-320.

Dennett, D. C. 1983. Intentional systems in cognitive ethology: The "Panglossian paradigm" defended. Behavioral and Brain Sciences 6:343-345.

Fagen, R. 1981. Animal Play Behavior. New York:Oxford University Press.

Fagen, R. 1993. Primate juveniles and primate play. In M. E. Pereira and L. A. Fairbanks (eds.). Juvenile primates: Life history, development, and behavior. Oxford University Press, New York. pp. 183-196.

FEDDERSEN-PETERSEN, D. (1991). The ontogeny of social play and agonistic behaviour in selected canid species. -- Bonn. Zool. Beitr. 42, p. 97-114.

Flanagan, O. 1992. Consciousness reconsidered. MIT Press, Cambridge, Massachusetts.

Galef, B. G., Jr. 1990/1996. Tradition in animals: Field observations and laboratory analyses. In M. Bekoff and D. Jamieson (ed.) Readings in animal cognition. MIT Press, Cambridge, Massachusetts. Pages 91-105.

Galef, B. G., Jr. 1996. Tradition and imitation in animals. In G. Greenberg and M. Haraway (eds.). Encyclopedia of comparative psychology. New York: Garland. forthcoming.

GOPNIK, A. (1993). Psychopsychology. Consciousness and Cognition 2, 264-280.

Griffin, D. R. 1992. Animal Minds. Chicago:University of Chicago Press.

HAILMAN, J. P. (1977). Optical Signals: Animal Communication and Light. Indiana University Press. Bloomington, Indiana.

HILL, H. L. & BEKOFF, M. (1977). The variability of some motor components of social play and agonistic behaviour in infant eastern coyotes Canis latrans var. -- Anim. Behav. 25, p. 907-909.

Holland, P. C. 1990. Event representation in Pavlovian conditioning: Image and action. Cognition 37: 105-131.

Humphrey, N. K. 1983. Consciousness regained. Oxford University Press, New York.

Ingold, T. 1988. The animal in the study of humanity. In T. Ingold (ed.) What is an animal? Unwin Hyman, London. Pages 84-99.

Itakura, S. 1995. An exploratory study of social referencing in chimpanzees. Folia Primatologica 64: 44-48.

Jamieson, D., and Bekoff, M. 1993. On aims and methods of cognitive ethology. Phil. Sci. Assoc. 2: 110-124.

Jensvold, M. L. A., and Fouts, R. S. 1993. Imaginary play in chimpanzees (Pan troglodyes). Human Evolution 8: 217-227.

Jerison, H. J. 1973. Evolution of the brain and intelligence. Academic Press, New York.

Kamil, A. C. (1987/1994). A synthetic approach to the study of animal intelligence. In L. Real (ed.). Behavioral mechanisms in evolutionary ecology. Chicago, University of Chicago Press, Chicago. Pages 11-45.

Leslie, A. M. 1987. Pretense and representation. The origins of "theory of mind." Psychological Review 94: 412-426.

Lima, S. L. 1995. Collective detection of predatory attack by social foragers: Fraught with ambiguity. Animal Behaviour 50: 1097-1108.

Lorenz, K. Z. 1996. The natural science of the human species. An introduction to comparative behavioral research - The "Russian Manuscript" (1944-1948). MIT Press, Cambridge, Massachusetts.

MARTIN, P. & CARO, T. M. (1985). On the functions of play and its role in behavioural development. -- Adv. Study Behav. 15, p. 59-103.

Meltzoff, A., and Gopnik, A. 1993. The role of imitation in understanding persons and developing a theory of mind. In S. Baron-Cohen, H. Tager-Flusberg, and D. Cohen (eds.). Understanding other minds. Oxford University Press, New York. pp. 335-366.

Mitchell, R. W. 1990: A theory of play. In M. Bekoff and D. Jamieson, eds. Interpretation and Explanation in the Study of Animal Behavior: Vol. I, Interpretation, Intentionality, and Communication. Boulder, Colorado: Westview Press.

Newberry, R. C., Wood-Gush, D. G. M., and Hall, J. W. 1988. Playful behaviour of piglets. Behavioural Processes 17: 205-216.

Pearce. J. M. 1987. Introduction to animal cognition. Lawrence Erlbaum Associates, Publishers. Hillsdale, New Jersey.

Pellegrini, A. D. (ed.). 1995. The future of play theory: A multidisciplinary inquiry into the contributions of Brian Sutton-Smith. SUNY Press, Albany, New York.

Perner, M. 1993. Understanding the representational mind. MIT Press, Cambridge, Massachusetts.

Ristau, C.. ed. 1991a. Cognitive ethology: The Minds of Other Animals Hillsdale, New Jersey:Erlbaum.

Ristau, C. 1991b. Aspects of the cognitive ethology of an injury-feigning bird, the piping plover. In C. Ristau, ed. Cognitive Ethology: The Minds of Other Animals. Hillsdale, New Jersey:Erlbaum.

Rosenberg, A. 1990. Is there an evolutionary biology of play? In M. Bekoff and D.

Jamieson, eds. Interpretation and Explanation in the Study of Animal Behavior. Vol. I, Interpretation, Intentionality, and Communication. Boulder, Colorado: Westview Press.

Savage-Rumbaugh, E. S. 1990. Language as a cause-effect communication system. \ Philosophical Psychology 3: 55-76.

Steiner, A. L. 1971. Play activity of Columbian ground squirrels. Zeitschrift für Tierpsychologie 28: 247-261.

Tanner, J. and R. Byrne. 1993. Concealing facial evidence of mood: Perspective-taking in a captive gorilla. Primates 34: 451-457.

Thompson, K. V. 1996. Behavioral development and play. In D. G. Kleiman, M. E. Allen, K. V. Thompson, S. Lumpkin, and H. Harris (eds.). Wild mammals in captivity: Principles and techniques. University of Chicago Press, Chicago.

Tinbergen, N. 1951/1989. The study of instinct. Oxford University Press, New York.

Tinbergen, N. 1963. On aims and methods of ethology. Zeitschrift für Tierpsychologie 20: 410-433.

Tomasello, M., B. L. George, A. C. Kruger, M. Jeffry, and F. A. Evans. 1985. The development of gestural communication in young chimpanzees. Journal of Human Evolution. 14: 175-186.

Tomasello, M., D. Gust, and G. T. Frost. 1989. A longitudinal investigation of gestural communication in young chimpanzees. Primates 30: 35-50.

Watanabe, S., Lea, S. E. G. and Dittrich, W. H. 1993. What can we learn from experiments on pigeon concept discrimination? In: H. P. Zeigler and H.-J. Bischof (Editors), Vision, Brain, and Behavior in Birds, MIT Press, Cambridge, MA, pp. 351-376.

Watson, D. M., and Croft, D. B. 1996. Age-related differences in playfighting strategies of captive male red-necked wallabie (Macropus rufogriseus banksianus). Ethology 102: 33-346.