Are dog breeds sufficient indicators of behaviour?

1. Introduction

Dogs are the oldest domesticated animal (Serpell, 2017) and show considerable variation in morphology, genetics, and behavior due to long periods of artificial selection. The process of artificial selection has caused many dog breeds to sustain significant changes that have subsequently influenced breed-typical behavior (Svartberg, 2005) and resulted in a notable shift away from breeding for function towards breeding that is dominated by appearance. This is an important area of study due to the sheer number of dogs kept as pets around the world (more than 400 million in 2018, according to Statista, 2021). Furthermore, once a breed label is assigned to a dog, it can have sociological consequences for the animal’s owners and sets behavioral expectations (Dickey, 2017). This paper will explore the origins of breed and behavior, and discuss whether breed labels perpetuate these behavioral expectations and ultimately whether they are sufficient indicators of behavior at all.

2. Origins of breed

Mehrkam & Wynne (2014) define dog breeds as groups of individuals within the subspecies canis familiaris that strongly resemble one another, citing general agreement across popular media and scientific literature that breeds of dogs differ behaviorally in “substantial, consistent and predictable” ways. In their study of breed-typical behavior, Svartberg (2005) notes that efforts to establish standardized dog breeds coincided with the advent of organized dog shows. This standardization has resulted in seven breed groups that are today recognized internationally by Fédération Cynologique Internationale (FCI), American Kennel Club (AKC) and The Kennel Club (KC).

The breeds listed in Table 1 are a small representation of each group according to KC (2021) and have been categorized in this way due to similarities based on original breed function. Conversely, the original breed function upon which these categorizations are built have diminished in recent decades (Svartberg, 2005) with greater focus on appearance leading to debates over the most accurate way to define a breed (Mehrkam & Wynne, 2014). One hypothesis is that breeds developed as a result of adaptation (Coppinger & Coppinger, 2002) while others suggest human intervention (by way of artificial selection) played a fundamental role in establishing the 218 breeds now formally recognized by The Kennel Club (2021) (Lindsay, 2000).

3. Origins of behaviour

Dog behavior has been shaped by millennia of contact with humans (Bradshaw, 2012; Shipman, 2011; Berns et al. 2012) and is now widely accepted to be determined by both genetic and environmental factors (Mackenzie et al. 1986; Storengen, 2016). One of the most influential studies of canine behavioral genetics began in 1945 and used data collected over several years (Scott & Fuller 1965, as cited in Mackenzie et al. 1986). The study compared different breeds of dogs in environmentally similar conditions so that behavioral differences could be assigned to genetics rather than environment or a combination of both. Scott & Fuller concluded that their results disputed the claim that all members of the same breed exhibit the same patterns of behavior, cautioning the reader “against accepting the idea of a breed stereotype” (Scott & Fuller 1965, as cited in Mackenzie et al. 1986). More recent studies have drawn similar conclusions (Wilsson, 2015; Wilsson, 2016; Lenkei et al. 2021) that behavior is influenced by both genetics and environment.

The neuroplasticity of dogs’ brains confirms that behavioral traits can change in response to new information, surpassing original beliefs that learning later in life took place only in existing receptors. We now know that while the juvenile period is especially suited to neuroplastic adaptation (as a puppy’s immature brain is organizing itself) there is hard neuroplastic change throughout adulthood too (whenever new information is processed) (Malacarne 1793, as cited in Rosenzweig, 1996; Lillard & Erisir, 2011). Now that we know dogs’ behaviour can change, we might consider Tinbergen’s four ‘major problems of biology’ (1963, as cited in Bateson & Laland, 2013) also known as the four whys of behavior: survival value, ontogeny, evolution, and causation.

Table 2 gives two examples of dog behavior and demonstrates how this behavior has changed both phylogenetically and ontogenetically, encouraged by the need to adapt to closer contact with humans. Another example of anthropocentric impact on dogs’ behavior is the modification of innate behaviors such as prey drive, producing dogs that stalk but do not attack or attack but do not kill (Beuchat, 2016) through artificial selection and training.

To conclude, behavior originates in both genetic and environmental influences and is ever-changing in response to ontogenetical and environmental adaptation. Moreover, dogs’ capacity for behavioral and neural plasticity suggests that breed labels are too rigid to be appropriately indicative of behavior.

4. Breed is not indicative of behaviour

Artificial selection dominated by physical appearance coincided with the emergence of organized dog shows around 150 years ago (Svartberg, 2005). Kenttämies et al. (2002) and Trut at al. (2004), among others, have illustrated how basic emotional traits can be rapidly altered in just a few generations under intense selection. Thus, while organized dog shows are young in evolutionary terms, it is reasonable to assume that their efforts to standardize breeds have contributed to changes in breed-typical behaviors. While the aim of artificial selection is to isolate and propagate desirable traits only (Encyclopedia Britannica, 2021), this is not always reliable. This is because a good understanding of heritability is needed for successful breeding programs, whereas currently only estimates of heritability are being used in most cases (Hradecká et al. 2015). As defined by Beuchat (2019) heritability is a statistic that indicates the strength of the association between genotype (genetic makeup) and phenotype (expression of behavior). Beuchat (2016) goes on to explain that breeders select candidates based on phenotypic behavior but that this is not necessarily an accurate reflection of good genes. For example, if the heritability of a trait is low, such as dog fear (MacLean et al. 2019), then the expression of that trait will not be a good indicator of genotype; the dog might have low genes for dog fear, but poor socialization will outweigh them. The outcome is that dogs with good genes are removed from the breeding process because of bad experiences, and vice versa. Moreover, while the heritability coefficient for characteristics like eye or coat color are usually high, there is more ambiguity when it comes to behavior, meaning that breed is not necessarily the dominant indicator of behavior.

Several studies that support this hypothesis include Svartberg’s study of 13,097 domesticated Swedish breeds who took part in standardized behavioral tests (2005). The results identified both “between-breed” and “within-breed” differences across several aspects of breed-typical behavior. Similarly, Lenkei et al. (2021) found that breed played no role in attachment patterns in family dogs. One of the most infamous arguments against breed labels is the debate over breed-specific legislation, which prohibits or restricts certain breeds based on legal presumption that such dogs exhibit dangerous behavioral traits (Dangerous Dogs Act, 1991). Studies that refute these claims include Ott et al. (2008) whose comparison of aggression in dogs affected by breed-specific legislation and Golden Retrievers identified “no significant differences”. It is important to note that other breeds not currently affected by breed-specific legislation have endured their time in the dangerous dogs ‘spotlight’, with dog bite-related fatalities being likened to a “perfect storm” of bad human-canine interactions (Gladwell, 2006). This idea of bad human-canine interactions introduces the role of environment on behavior. In their discussion of nature versus nurture, Wilsson (2016) identifies three key developmental stages and comments on the effect of environment on dogs’ behavioral development:

a) Neonatal (1-2 weeks). Dogs handled from birth were more explorative, superior in problem-solving and more social compared to non-handled dogs (Fix & Stelzner 1966, as cited in Wilsson, 2016).

b) Socialization (4-12 weeks). Only 40 minutes of exposure to humans is sufficient to produce a non-fearful, “socialized” dog.

c) Juvenile (3-12 months). Dogs growing up in the company of conspecifics showed less fear and aggression towards unfamiliar dogs compared to those living alone.

Wilsson (2015) contends that with most studies evaluating temperament showing heritability in the range of 10-30%, this leaves 70-90% of phenotypic variation that is non-genetic. It is thus reasonable to argue that the environment’s profound effect on behavior is equal to, if not greater than, breed.

Image 1. Stray dogs in Thailand.

5. Breed is indicative of behaviour

In their study of breed-typical behavior, Svartberg (2005) identifies several scientific and non-scientific papers that suggest behavioral breed differences can be explained by differences in selection during the breeds’ origin. Some examples include breeds selected for ‘ratting’ (such as Terriers) being less fearful than other breeds (Mahut, 1959) and Gundogs (such as Spaniels) scoring high in handling and responsiveness tests compared to dogs of other breed groups (Seksel et al. 1999). Another example we might consider is the Border Collie with a great number of studies using the breed (Coren, 2006; Pilley & Reid, 2011; Riemer et al. 2016). When tested for instinctive intelligence (ability to complete original breed function), adaptive intelligence (ability to learn for themselves), and working intelligence (ability to learn for humans), the Border Collie scored highest in all three (Coren, 2006). When trained intensively for three years, Chaser displayed the ability to learn and retain the proper noun-names of 1,022 objects (Pilley & Reid, 2011) while dogs in Kaminski et al.’s experiments (2009) proved their ability to understand the representational nature of objects by fetching both the items themselves and photos of said items. These findings demonstrate why the artificial selection of Border Collies is highly profitable (BBC News, 2021a) and why they have earned their place in popular culture as “the smartest dog in the world” (American Kennel Club, 2019). However, while these findings do reinforce the breed stereotype, it could be contested that they focus on intelligence and cognition as opposed to behavior.

In their study of highly heritability breed differences in dog behavior, MacLean et al. (2019) found that trainability was among the highest scoring traits for among-breed heritability. Given that trainability refers to an individual’s capacity to gain proficiency in a particular skill (APA Dictionary of Psychology, 2021) this may address the breed-typical intelligence of Border Collies discussed above.

Figure 1

Heritability estimates for 14 behavioral traits as found in MacLean et al. (2019). The grey points are measurements of heritability of dogs of the same breed. The green and yellow points refer to the heritability across breeds, taken from the original datasets of Hayward et al. (2016, as cited in MacLean et al. 2019) and Parker et al. (2017, as cited in MacLean et al. 2019).

This study combined a large set of behavioral assessments with DNA genotyping data from two separate studies (Hayward et al. 2016 & Parker et al. 2017, as cited in MacLean et al. 2019) and identified 14 behavioral traits of interest. Figure 1 demonstrates that heritability in dogs of the same breed is generally low (h2 = 0.3 or less) whereas heritability across breeds ranges from 0.4 to 0.7, which means 40-70% of variations in behavior can be attributed to variations in genotype. The traits with the highest heritability were trainability (h2 = 0.73), stranger-directed aggression (h2 = 0.68), chasing (h2 = 0.62) and attachment and attention-seeking (h2 = 0.56), which MacLean et al. (2019) contend is consistent with the hypothesis that behaviors correspond to original breed function and developed during the formation of modern breeds. Thus, breed remains indicative of behavior.

6. Conclusion

Levi-Strauss’ theory of binary opposition observed the human need to categorize and define the world (as cited in Dickey, 2017). Such is the need to define the ambiguous that many behaviorists divide a single behavioral category, such as aggression, into as many as eight subcategories (Beaver, 1983; Lockwood, 2021) which leads some to argue that our expectations of dogs have simply become unrealistic (Dickey, 2017). It is thus reasonable to question whether the assigning of breed labels perpetuates these expectations, and consider what role these labels play in socioeconomic problems such as increasing abandonment rates (of particular interest following the Covid-19 pandemic) (BBC News, 2021b).

As concluded by Mackenzie et al. (1986), Storengen (2016), and Lenkei et al. (2021), it is now widely accepted that behavior is a mix of genetics and environment with certain traits – such as trainability (MacLean et al. 2019) – having higher heritability and thus greater influence over dogs’ expression of the associated trait. Conversely, studies at each stage of development (Wilsson, 2016) were consistent with the theory that the impact of environmental influences on behavior increases with age; a hypothesis supported by dogs’ continued neuroplasticity.

Without one definitive answer to the dominant origin of behavior, it is reasonable to suggest that breed labels should not be relied upon. Instead, a new kind of categorization based on the seven internationally recognized groups (see table 1) might be a more appropriate way of setting realistic behavioral expectations, encouraging conversations about breed traits more generally, and highlighting the importance of a dog’s individuality.

Unless otherwise stated, all images are the author’s own.

7. References

American Kennel Club (2019). Remembering Chaser, the “Smartest Dog in the World”. Retrieved from: <https://www.akc.org/expert-advice/news/remembering-chaser-the-smartest-dog-in-the-world/> [Accessed 27 November 2021].

APA Dictionary of Psychology (2021). Trainability. Retrieved from: <https://dictionary.apa.org/trainability> [Accessed 27 November 2021].

Beuchat, C. (2016). Understanding the heritability of behavior in dogs. The Institute of Canine Biology. Retrieved from: <https://www.instituteofcaninebiology.org/blog/understanding-the-heritability-of-behavior-in-dogs> [Accessed 21 November 2021].

Beuchat, C. (2019). The genetics of canine behavior goes molecular. The Institute of Canine Biology. Retrieved from: <https://www.instituteofcaninebiology.org/blog/the-genetics-of-canine-behavior-goes-molecular> [Accessed 21 November 2021].

Beaver, B. (1983). Clinical classification of canine aggression. Applied Animal Ethology, 10(1-2), 35-43. https://doi.org/10.1016/0304-3762(83)90110-4.

Bateson, P., & Laland, K. N. (2013). Tinbergen’s four questions: An appreciation and an update. Trends in Ecology & Evolution, 28(12), 712-718. https://doi.org/10.1016/j.tree.2013.09.013.

BBC News (2021a). Welsh sheepdog Kim sells for world record £27,100. Retrieved from: <https://www.bbc.co.uk/news/uk-wales-55935911> [Accessed 27 November 2021].

BBC News (2021b). Dogs bought in lockdown being abandoned. Retrieved from: <https://www.bbc.co.uk/news/uk-wales-58996017> [Accessed 21 November 2021].

Berns, G., Brooks, A., & Spivak, M. (2012). Functional MRI in awake unrestrained dogs. Plos ONE, 7(5), e38027. https://doi.org/10.1371/journal.pone.0038027.

Bradshaw, J. (2012). In Defense of Dogs: Why dogs need our understanding. Penguin.

Coppinger, R., & Coppinger, L. (2002). Dogs: A New Understanding of Canine Origin, Behavior and Evolution (2nd ed.). University of Chicago Press.

Coren, S. (2006). The intelligence of dogs. Pocket Books.

Dickey, B. (2017). Pit Bull: The battle over an American icon. Vintage Books.

Encyclopedia Britannica (2021). Selection. Retrieved from: <https://www.britannica.com/science/selection> [Accessed 21 November 2021].

Gladwell, M. (2006). Troublemakers. The New Yorker. Retrieved from: <https://www.newyorker.com/magazine/2006/02/06/troublemakers-malcolm-gladwell> [Accessed 21 November 2021].

Hradecká, L., Bartoš, L., Svobodová, I., & Sales, J. (2015). Heritability of behavioral traits in domestic dogs: A meta-analysis. Applied Animal Behavior Science, 170, 1-13. https://doi.org/10.1016/j.applanim.2015.06.006.

Kaminski, J., Tempelmann, S., Call, J., & Tomasello, M. (2009). Domestic dogs comprehend human communication with iconic signs. Developmental Science, 12(6), 831-837. https://doi.org/10.1111/j.1467-7687.2009.00815.x.

Kenttämies, H., Nordrum, N., Brenøe, U., Smeds, K., Johannessen, K., & Bakken, M. (2002). Selection for more confident foxes in Finland and Norway: Heritability and selection response for confident behaviour in blue foxes (Alopex lagopus). Applied Animal Behaviour Science, 78(1), 67-82. https://doi.org/10.1016/s0168-1591(02)00078-3.

Legislation.gov.uk. 1991. Dangerous Dogs Act 1991. Retrieved from: https://www.legislation.gov.uk/ukpga/1991/65/introduction [Accessed 21 November 2021].

Lenkei, R., Carreiro, C., Gácsi, M., & Pongrácz, P. (2021). The relationship between functional breed selection and attachment pattern in family dogs (canis familiaris). Applied Animal Behaviour Science, 235, 105231. https://doi.org/10.1016/j.applanim.2021.10523.

Lillard, A., & Erisir, A. (2011). Old dogs learning new tricks: Neuroplasticity beyond the juvenile period. Developmental Review, 31(4), 207-239. https://doi.org/10.1016/j.dr.2011.07.008.

Lindsay, S. (2000). Handbook of Applied Dog Behavior and Training, Volume One, Adaptation and Learning. Iowa State Press.

Lockwood, R. (2021). Ethology, ecology and epidemiology of canine aggression. In J. Serpell, The Domestic Dog (2nd ed., pp. 160-181). Cambridge University Press.

Mackenzie, S., Oltenacu, E., & Houpt, K. (1986). Canine behavioral genetics — A review. Applied Animal Behaviour Science, 15, 365-393. https://doi.org/10.1016/0168-1591(86)90128-0.

MacLean, E., Snyder-Mackler, N., vonHoldt, B., & Serpell, J. (2019). Highly heritable and functionally relevant breed differences in dog behaviour. Proceedings Of The Royal Society B: Biological Sciences, 286(1912), 20190716. https://doi.org/10.1098/rspb.2019.0716.

Mahut, H. (1958). Breed differences in the dog's emotional behaviour. Canadian Journal Of Psychology/Revue Canadienne De Psychologie, 12(1), 35-44. https://doi.org/10.1037/h0083722.

Mehrkam, L., & Wynne, C. (2014). Behavioral differences among breeds of domestic dogs (Canis lupus familiaris): Current status of the science. Applied Animal Behaviour Science, 155, 12-27. https://doi.org/10.1016/j.applanim.2014.03.005.

Ott, S., Schalke, E., von Gaertner, A., & Hackbarth, H. (2008). Is there a difference? Comparison of golden retrievers and dogs affected by breed-specific legislation regarding aggressive behavior. Journal Of Veterinary Behavior, 3(3), 134-140. https://doi.org/10.1016/j.jveb.2007.09.009.

Pilley, J., & Reid, A. (2011). Border collie comprehends object names as verbal referents. Behavioral Processes, 86(2), 184-195. https://doi.org/10.1016/j.beproc.2010.11.007.

Riemer, S., Müller, C., Virányi, Z., Huber, L., & Range, F. (2016). Individual and group level trajectories of behavioural development in Border collies. Applied Animal Behaviour Science, 180, 78-86. https://doi.org/10.1016/j.applanim.2016.04.021.

Rosenzweig, M. (1996). Aspects of the Search for Neural Mechanisms of Memory. Annual Review Of Psychology, 47(1), 1-32. https://doi.org/10.1146/annurev.psych.47.1.1.

Seksel, K., Mazurski, E., & Taylor, A. (1999). Puppy socialisation programs: short and long term behavioural effects. Applied Animal Behaviour Science, 62(4), 335-349. https://doi.org/10.1016/s0168-1591(98)00232-9.

Serpell, J. (2017). The domestic dog (2nd ed., pp. 1-21). Cambridge University Press.

Shipman, P. (2011). The Animal Connection: A new perspective on what makes us human. W. W. Norton & Company. <https://books.google.co.uk/books?hl=en&lr=&id=hN8DlaA-oMwC&oi=fnd&pg=PA9&ots=n_IxMj1ANS&sig=HYJ1dPeza3QX8KEDZhKM9WMpz-U&redir_esc=y#v=onepage&q&f=false> [Accessed 21 November 2021].

Statista (2021). Dog and cat pet population worldwide 2018. Retrieved from: <https://www.statista.com/statistics/1044386/dog-and-cat-pet-population-worldwide/> [Accessed 21 November 2021].

Storengen, L. M. (2016). Genetic studies of canina anxieties (Published PhD thesis).

Svartberg, K. (2005). Breed-typical behaviour in dogs—Historical remnants or recent constructs?. Applied Animal Behaviour Science, 96(3-4), 293-313. https://doi.org/10.1016/j.applanim.2005.06.014.

The Kennel Club (2021). Breed standards. Retrieved from: <https://www.thekennelclub.org.uk/breed-standards/> [Accessed 20 November 2021].

Trut, L., Plyusnina, I., & Oskina, I. (2004). An Experiment on Fox Domestication and Debatable Issues of Evolution of the Dog. Russian Journal Of Genetics, 40(6), 644-655. https://doi.org/10.1023/b:ruge.0000033312.92773.c1

Wilsson, E. (2015). Nature and nurture – how different environmental conditions interact with the behaviour of the maturing dog. Journal Of Veterinary Behavior, 10(5), 441. https://doi.org/10.1016/j.jveb.2015.07.013

Wilsson, E. (2016). Nature and nurture—How different conditions affect the behavior of dogs. Journal Of Veterinary Behavior, 16, 45-52. https://doi.org/10.1016/j.jveb.2016.10.002