Dog faces exhibit anatomical differences in comparison to other domestic animals

Abstract

Faces and their movements provide us the ability to understand parts of the evolution of communication, social behavior, cognition, and the brain (Burrows, 2008; Parr & Waller, 2006; Schmidt & Cohn, 2001). Humans and other mammals communicate using numerous signals that include visual displays, vocalizations, tactile signals, and olfactory signals (Brecht & Freiwald, 2012; Burrows, 2008; Liebal, Waller, Burrows, & Slocombe, 2013). Each of these signals is generated or received at least partially using the craniofacial complex. Within the visual realm, the face is the primary communication tool for terrestrial mammals and this is where facial expressions are generated. Many mammals also have facial fur of contrasting patterns and colors that are themselves used as a kind of signaling in social interactions (Caro, Walker, Santana, & Stankowich, 2017; Santana, Alfaro, & Alfaro, 2012; Santana, Alfaro, Noonan, & Alfaro, 2013). Human (and most mammalian) faces make facial expressions using the numerous mimetic muscles found deep to and attached to the skin of the face. These facial expressions help individuals with cohesion of social and kin groups, maintaining relationships, and potentially signal the intent and emotional state of the sender (Burrows, 2008; Burrows & Cohn, 2014; Parr & de Waal, 1999; Parr & Waller, 2006). A large majority of these studies comes from primates but recent work has increased our understanding of how some domestic mammals use facial expression in social interactions with conspecifics and humans (Caeiro, Burrows, & Waller, 2017; Kaminski, Hynds, Morris, & Waller, 2017; Maglieri, Prato-Previde, Tommasi, & Palagi, 2020; Waller, Julle-Daniere, & Micheletta, 2020; Wathan, Burrows, Waller, & McComb, 2015). Because facial expressions and mimetic muscles are associated with social groups, complexity of social interactions, and the cognitive/neural networks that manage these interactions (Barrett, Henzi, & Rendall, 2007; Dunbar, 2016), they are of interest to people studying the evolution of social behavior and the brain within ourselves, primates, and all other mammals. While most of our understanding of facial expression and the evolution of social behavior has been focused, understandably, on primates, more recent work has begun turning to domestic mammals, their faces, mimetic muscles, and facial expressions. Humans are the only living species that have domesticated other species for our own interests. Studying our closest living relatives, the chimpanzees and bonobos, can inform our understanding of how our own social behavior, cognition, and brains evolved, but what a gaping hole we leave in that understanding by missing the very species that we have created! Domestication involves multiple processes but it occurs quickly, relative to the descent of most wild species (Darwin, 1859, 1868; Larson & Fuller, 2014). Archeological and molecular evidence indicates that humans started bringing wild animals “into the fold” around 30,000 years ago with domestication of the dog from gray wolves (Driscoll, Macdonald, & O'Brien, 2009; Druzhkova et al., 2013; Ovodov, Crockford, Kuzmin, Higham, & Hodgins, 2011). The specific population(s) of gray wolves that gave rise to domestic dogs is now extinct but this process that began in the Late Pleistocene seems to have occurred simultaneously in Western Europe, East Asia, and Central Asia and pre-dates the dawn of agriculture in these areas (Clutton-Brock, 1995, 1999; Frantz et al., 2016; Vila et al., 1997). Domestication of other mammals followed much later than dogs and includes goats, horses, and cats, among others, beginning around 10,000 years ago (Driscoll et al., 2009). Domestication of mammals generally results in new species that are distinguished both morphologically and behaviorally from their closely related wild counterparts as well as their ancestors (see Box 1). Domesticated mammals typically display a suite of morphological, physiological, and behavioral traits that are unique relative to their wild ancestors and counterparts and this phenomenon is sometimes referred to as the “domestication syndrome” (Wilkins, Wrangham, & Fitch, 2014). Primary among these is decreased aggression, changes in pelage (coat) color and pattern, decreased size of the midface (snout) and teeth, ears that are floppy (with some notable exceptions), and a generally more gracile craniofacial complex (Darwin, 1859, 1868; Larson & Fuller, 2014; Wilkins et al., 2014). Something about the process of domestication, across the wide phylogenetic range of mammals, results in this generalized suite of characteristics. Several recent studies have pointed to neural crest cells, a group of stem cells peculiar to vertebrate animals, as being the common shared developmental connection in this “domestication syndrome”, from fish to birds to dogs. As outlined in Wilkins et al. (2014), domestication syndrome may involve alterations of neural crest cell numbers, or “neural crest hypofunction”, which produces the changes in morphology that we see in reduced brain size, pelage pigmentation, jaws, teeth, and ears. If there are indeed neural crest cell deficits in domestication syndrome, they could arise via three mechanisms: reduced numbers of original neural crest cells formed at the outset, lower migration of neural crest cells and lower numbers at their final destinations, or decreased proliferation of neural crest cells at their final destinations. No single gene is thought to be responsible for this neural crest cell hypofunction but it appears instead to be polygenic. While hypothesized genetic changes that impact neural crest cell proliferation and/or migration show promise in our understanding of “domestication syndrome”, it is also possible that general changes to the thyroid gland morphology and physiology could have played a role (Crockford, 2000). Increasing our understanding of the anatomical and behavioral differences between domestic mammals and their wild counterparts/ancestors can also inform our understanding of what humans thought was valuable in the domestication process as well as how humans and domestic mammals communicate with one another via facial expression. It may also help us understand our own “self-domestication”, the progressive set of domestication syndrome traits increasingly present throughout human history, including reducing body mass, shortening of the midface and tooth size, reduction in aggressive behavior, and reduction of sexual dimorphism (Hare, 2017; Wilkins, 2017; Wrangham, 2019). Horses, cats, and dogs all use facial expressions in their interactions with humans and conspecifics but to varying degrees. While equivocal, accumulating evidence is demonstrating that humans and dogs are uniquely attuned to one another's facial expressions to the exclusion of other domestic mammals (Albuquerque et al., 2016; Schirmer, Seow, & Penney, 2013; Waller et al., 2013). Not only are humans good at reading facial expressions of dogs, but dogs are good at reading facial expressions of humans (Correira-Caeiro, Guo, & Mills, 2020). Does this mutual ability to read one another's facial expressions indicate that dogs and humans co-evolved and influenced one another's brain evolution? Cats typically do not recognize the face of their human handlers by way of visual cues, as opposed to the situation in dogs, but they do recognize faces of other cats (Lomber & Cornwell, 2005). Other studies have suggested that cats are not particularly responsive to their owners' faces and that people are not as good at reading cat facial expressions as dogs (Bennett, Gourkow, & Mills, 2017; Galvan & Vonk, 2015). Explanations for these phenomena focus on several areas, including the evolution of dogs from an already highly social canid ancestor while cats evolved from a less social felid ancestor, as well as the longer time frame that humans and dogs have lived together. Domestic cats are only marginally different from the African wild cat in terms of morphology and behavior (Driscoll et al., 2007, 2009). In Figure 1, both the domestic and wild cat appear relatively similar in terms of craniofacial morphology and are both relatively solitary by nature. The domestic horse is closely related to the wild Przewalski's horse but important facial differences are obvious here. The domestic horse generally has a more gracile head, greater variation in pelage pattern, and a change to a flat mane from the erect mane in Przewalski's horse. While studies of Przewalski's horse in the wild are incomplete, their behavior seems to parallel that of feral domestic horses, with both species being social animals (Boyd, 1988; Feh, 2005; Orlando et al., 2013). The domestic dog (Canis familaris) differs overtly from its close wild relative, the gray wolf (C. lupus). Unlike the cat, dogs have an explicitly different suite of anatomical and behavioral characteristics relative to the gray wolf which include smaller vibrissae (or whiskers), external ears that are typically floppy in most breeds, more gracile craniofacial complex, and overtly affiliative behaviors toward humans. Holding eye gaze has become one of the hallmark features of the relationships between humans and domestic dogs (Miklosi et al., 2003, 2013). Emerging evidence is showing that domestic horses are able to read some aspects of both other horses' and human facial expressions but evidence for humans being able to accurately read horse facial expression is not yet as compelling (Proops, Grounds, Smith, & McComb, 2018; Proops & McComb, 2012; Proops, McComb, & Reby, 2009; Smith, Proops, Grounds, Wathan, & McComb, 2016; Wathan, Proops, Grounds, & McComb, 2016). Horses themselves are social animals, both in domesticated and feral settings and, due to their social natures, may attend to facial expressions (Proops et al., 2018; Smith et al., 2016). Horses, be they domestic horses, feral horses, or Przewalski's horses, are diurnal and rarely solitary by nature (Cameron, Setsaas, & Linklater, 2009; Feh, 1988). E. przewalskii is endangered in the wild but several robust field studies of their social behavior have been conducted, showing that this species is highly social and occurs in herds that number from 9 individuals-23. However, almost certainly these numbers are limited by the few numbers of these horses in existence (Boyd, 1988; Wolter, 2018). Social groups include both harems and bachelor male groups. Aggression in E. przewalskii is suggested to be high (Feh, 1988), but social groups form long-lasting bonds that remain in close proximity to one another and members of harem herds spend considerable time grooming one another. Przewalski's horse is smaller than most breeds of domestic horse, which may reflect the traits human chose during domestication that focused on using horses for transportation. Feral groups of domestic horses have similar social behavior (Cameron et al., 2009; Klingel, 1982). Social communication includes many of the same signals used by domestic horses, including olfactory communication, short-distance vocalizations, and facial expressions including movements of the ears and eye region. Domestication of the horse seems to have begun around 4,000 BCE in Mesopotamia and China in association with their use as a transport animal. There are now more than 400 domestic breeds. Contemporary horses generally bear a more neotenous set of characteristics than seen in Przewalski's horse and other wild equids. While fur on domestic horse faces comes in a variety of colors and patterns, the general form of the skull does not differ markedly from that of wild equids. Recently developed Facial Action Coding System for horses (EquiFACS) has shown that domestic horses have a complex and high number of facial expressions that they display in interactions with one another and with humans (Wathan et al., 2015). Given that both feral and Przewalski's horses have complex social behavior (i.e., close-proximity interactions with a large number of individuals that consist of both kin and non-kin) and live in life-long socially bonded groups, it is not surprising that they have a complex facial expression repertoire. Domestic cats represent a vastly different scenario. The cat (F. catus) appears to have first been domesticated somewhere in the Fertile Crescent around 4,000 years ago during the Near East Neolithic period (Ottoni et al., 2017). The image on the right of the African wild cat (F. silvestris lybica) is the ancestor of all domestic cats. Domestication of the cat has resulted in only minimal anatomic and behavioral differences relative to the African wild cat to the extent that the domestic cat is sometimes considered to be a subspecies of Felis silvestris: F. s. catus (Clutton-Brock, 1999; Driscoll et al., 2009; Ottoni et al., 2017; Randi & Ragni, 1991). One of the only ways to grossly differentiate the domestic cat from the African wild cat is pelage color, with the wild cat typically varying little in the range of pelage coloration and patterning. Domestic cats, like the domestic horse, display a wide variety of coat colors. Associations between humans and wild cats probably were commensal, with cats feeding on rodents that infested grain stores of the earliest farmers (Ottoni et al., 2017). Domestic cats can be remarkably affiliative with humans but they do not have the range of facial expressions seen in domestic horses or in domestic dogs. CatFACS is a Facial Action Coding System developed to quantify and describe cat facial expression and this system has demonstrated a lower number of facial displays relative to horses and dogs (Caeiro et al., 2017). Many cat facial expressions are focused on movement of the ears and changes in the size and shape of the pupils (Bennett et al., 2017; Caeiro et al., 2017) and, unlike in dogs, cat facial expression is not correlated to how quickly humans adopt cats from cat shelters (Caeiro et al., 2017). Cat facial expressions do not appear to have undergone the selection that dogs underwent during the domestication process. All felids are relatively solitary, predatory mammals, with the exception of the African lion (Panthera leo), and domestic cats do not diverge from this general characteristic. Most wild felids are either primarily nocturnal or crepuscular (active at dawn and dusk), with the exception of the cheetah (Acinonyx jubatus). Unlike the wild equids and wild gray wolves that gave rise to domestic horses and domestic dogs, the African wild cat (Felis silvestris lybica) is asocial and, therefore, probably does not have a complex or numerous facial display repertoire that could have served as a basis for domestic cats. Domestic dogs (Canis familiaris), unlike cats, bear far less resemblance to their closest living relative, the gray wolf (C. lupus). While gray wolf subspecies grossly resemble one another, it is typically easy to distinguish a domestic dog from a gray wolf. Domestic dogs generally have smaller heads than gray wolves, most have the ability to or have full-time lowered ears, much smaller vibrissae, smaller teeth and reduced “snouts”. Unlike horses and cats, though, dogs have remarkable variation in not only pelage pattern & coloration on the face but they also have enormous variation in size and shape of the skull, resulting in over 300 recognized breeds today (see Boxes 1 and 2). recognize facial expressions both in other dogs and in humans (Albuquerque et al., 2016), and humans are at reading facial expressions & 2019). ability is not in our relationships with domestic horses and domestic cats. Humans in at reading dog facial expressions than we are at reading facial expressions in our closest living relative, the et al., 2019). The between humans and dogs is one of the characteristics of humans. also in mutual gaze with humans to the exclusion of other this mutual is in both which the that many humans toward dogs et al., 2015). It has been hypothesized that during the process of domestication, dogs human that resulted in the that many humans display toward dogs et al., 2019). the domestication humans may have associated with gray wolves that behaviors that these resulting in dog breeds that generally display facial features relative to gray The is dog Unlike domestic cats, domestic dog faces and both behaviorally and from their closest living relative, the gray wolf. The range in variation in domestic dog craniofacial is greater than the variation seen in among all other canid species (Clutton-Brock, of dog and faces from (or as shown in the and the to (or as shown in the and the Unlike gray wolves, most domestic dog breeds have ears that flat the head, with notable as shown in Figure domestic dog breeds display a of pelage patterns and as well as relative of fur on the from the in with little facial fur to the with a high of fur in Figure dog breeds and breeds, as the in Figure have the of the eye is also the where the is in many dog breeds Waller, & Burrows, a that does not to be present in gray The of this in some dog breeds and breeds, with the color contrasting in the of the eye the of eye gaze in communication between dogs and humans. Domestic cat breeds can from breeds as to relatively breeds as but these differences do not seen in domestic dogs & de 2013; & Most cat breeds have ears that are erect (with some notable exceptions), like the African wild One of the characteristics that the is that at the domesticated animals have more or facial to some of the features humans in themselves 1). is particularly at the dog the wolf (see Figure 2). The shape of the dog is sometimes at almost in a the wolf is In human facial of the are considered more As people and facial become the of their shape is by as or of the in the of this is as an by as being or & other humans to of their to the to the face to a humans dogs to have this shape they it as and in a is the shape of the or in The gray wolf facial shape from face we the in to is relatively flat 1). The dog midface is more where the to to the similar to a human is in all of the domesticated dog breeds shown in Figure in facial shape is seen for the domesticated horse in Figure 1 of the to the wild horse, has a from to the domesticated horse has a to the to the more like a human humans have domesticated animals not to have a smaller but to have one that is more humans or domesticated animals to have traits they in domesticated animals more has demonstrated that domestic horses make facial expressions (Wathan et al., and that domestic cats make facial expressions (Caeiro et al., but humans do not to be as good at these as they are in dogs. Domestic dogs have over facial expressions. a chimpanzees have facial expressions Waller, & and humans have facial expressions. While cats have more facial expressions than we many of these involve movement of the ears while horse involve a of movements of the ears, and the same as horses but the range of facial expressions in dogs is All three of these domestic species have grossly similar of facial (or with some notable differences (Caeiro et al., 2017; & de 2013; et al., Each species has muscles that to the ears, the eye the around the and to the differences in the only for these faces living in relatively large social groups is associated with a of facial but what are the anatomical of this at the of mimetic in these species may provide us with some about the dog face and horse face are more than the Figure of the common to all three using and is attached to the and to the around the of the and this to the of the In it a by the of the and and in other it the same muscles were at and using and to distinguish areas of and were and during followed by with the of a and were from and using The horse was from the of of the cat was from of and the dog was from the of of of these animals was for the of the present Box that domestic cats have in this where we see the in humans and some other primates while both the domestic dog and gray wolf have a robust of in the It may be this that for the greater that we see in the dog face relative to both the cat and horse. Given that the gray wolf in Figure also has a robust is it possible that a was already present in the gray wolf that gave rise to the first domestic the of a to with we how many facial expressions gray wolves have but these are we may between facial expressions in gray wolves and domestic dogs. that the high of the human face is due to more than the mimetic (Burrows, & 2016). The or is a many of the muscles of facial expression and an is shown in Figure to the way a facial muscles both to other and to to facial movement et al., & 2008; & humans to have an enormous of facial expressions of varying and is found in humans (see Figure over the and of the but not in the or and this system is considered to be the for the of the face et al., studies have shown that chimpanzees have a reduced but present while species that are more from not have this to the that the is of the system responsible for high of the human and face. that the to and facial that the facial expressions of humans and, to some The domestic cat of a mimetic the from the A large is also in this which the robust and numerous vibrissae that domestic cats have throughout the of the The of a in the domestic cat, with the of and large vibrissae may to reduced of the face in this region. The domestic horse has greater between the of the face and the mimetic and this includes robust but the is partially with numerous and of the horse face is greater than the cat face (see and and of this greater may be due to the of a in the horse. Figure of a human for The gray wolf and domestic dog show relatively between the mimetic and the this includes relatively few and While we do not have on of this these show that a is in both and appears to be the same it possible that this is partially responsible for the greater number of facial expressions in domestic dogs (see The are to Smith, and for their kind to to this are to the for their that this also to for to use dog

Topics & Concepts