top of page
Genetics & Behaviour – Does Breed Matter?


To gain a proper understanding of how genetics influences behaviour traits, we need to start at the beginning – with our understanding of how dogs became dogs. This has been explained beautifully by biologist Raymond Coppinger in "DOGS: A New understanding of Canine Origin, Behaviour and Evolution" – while I have tried to summarise the contents of this book briefly here, I would urge anyone who would like to explore this subject in more detail to read it in full.


According to Coppinger, dogs began to emerge from a wolf-type ancestor around the time that humans switched from a nomadic to an agricultural lifestyle. As people settled into permanent villages, waste products were discarded outside of these villages. The dog’s ancestor was able to take advantage of this ready food supply and began scavenging on these “dumps”. According to natural selection, those individuals that were the fittest for that environment – in this case, those that were able to tolerate human proximity better (shorter flight distance and higher threshold for fight/flight response to the smell or sound of humans) – were most successful and survived to pass on their genes to the next generation. Over many generations, this process of variation, selection and further variation led to tamer and tamer individuals that eventually wandered around and into human settlements to take advantage of the food found there.[i] These early “dogs” were tolerated because they cleaned up waste, kept certain pests and predators away and, as awful as it might sound to us, could be turned into an easily accessible meat supply in lean times. “Dogs” that were safe for humans to be around were successful in human communities, while those that displayed dangerous levels of aggression were swiftly removed from the gene pool. What emerged was an animal that was slow to display defensive aggression towards people and whose predatory inclinations were in no way a threat to us.


This last point is very important – for dogs to be safe around people, dogs cannot see people as food or direct predatory behaviour towards them. This was brought home to me many years ago when a Cheetah outreach programme visited the school I was volunteering at, as part of their public awareness campaign. Shadow, their Cheetah ambassador, was quite “tame”. He walked on a harness and lead and was used to people stroking him. All the children were invited to meet Shadow and he sat quietly and purred, like a giant cat. However, this visit occurred at the end of the school day and the children began to leave for home while Shadow was still in the school hall and the door to the hall was open. Suddenly Shadow’s handlers seemed to panic and yelled for the door to be closed. A child had cycled past the door and Shadow had caught sight of him. The handlers knew that a moving target like this would be interpreted as prey and would turn the very placid Shadow into a hunter in a split second. I could just imagine the phone call to the child’s parents: “I am so sorry, but John was eaten by a Cheetah…”! Anyway, Shadow’s handlers were aware that while he was “tame” he was genetically still a hunter and once his predatory sequence was triggered, it would only end one way – luckily because they knew this and were aware, they acted quickly and disaster was averted.


The alteration to predatory behaviour that has made the dog what it is today, is the main focus of this article. Becoming a scavenger and living among humans drastically altered the predatory behaviour sequence in dogs – unlike their wild ancestors, dogs are no longer hunters, but scavengers. In keeping with this, their skull, jaw and tooth size has been reduced significantly in proportion to their body size and dogs are able to more easily digest a much wider range of food sources. However, remnants of the predatory sequence remained in early domestic dogs and humans took advantage of this to use dogs for various types of work. Dogs that displayed virtually no predatory behaviours were ideal for guarding livestock, those that possessed a propensity to stalk and chase animals without biting them, were used for herding and those that went quickly from spotting smaller animals to grabbing and killing them, were used for hunting or pest control. This is really where the first “types” of dogs emerged. Dogs were selected for the work that they could do, NOT for their looks. However, in accordance with Darwin’s “mysterious law of correlation”  the selection for particular behavioural traits resulted in particular physical traits emerging in that group. While today we may select more for FORM (breed standard looks), in most cases the behavioural traits are closely linked and follow along.


To understand better the various types of dogs that emerged and where and how the breeds we have today fit into these categories, we need to understand “neoteny”. Neoteny is the theory that domestic animals retain juvenile characteristics even into adulthood. In other words, dogs never really grow up - they remain socially, emotionally and behaviourally immature. It is this immaturity which makes them “safe” around people and so this immaturity that was selected into their genetic make-up. What is interesting is that along with permanent behavioural immaturity, have come physiological changes such as floppy ears, shorter muzzles and coat colour variation. So not only are dogs behaviourally immature, but they can also have puppy-like physical characteristics. In a fascinating experiment, Russian scientist Dmitry Balyaev selectively bred foxes farmed for their fur by choosing only the “tamest” or easiest to handle individuals in the programme. These foxes were notoriously aggressive towards people and difficult to deal with (who can blame them), yet by selectively breeding for “tameness” after only a few generations, the foxes showed virtually no fear of their caretakers and even wagged their tails and licked them. What is more, by selecting for this behavioural trait, there were accompanying physiological changes, such as floppy ears, variation in coat colour and curly tails. By selecting for tameness, Balyaev created a neotenised version of the silver fox. His experiment is a miniature version of how we ended up with the domestic dog and the effect of neoteny.


During the selection of the modern domestic dog, 4 main types of dog emerged:




These types are categorised according to a combination of FORM (their appearance) and FUNCTION. Function is determined by how much of the predatory sequence they retain and therefore which type of work they are suitable for. The full predatory sequence for a wild canid is as follows:




It is not normal for a domestic dog to carry out this full sequence. While some may still go all the way to KILL-BITE and some may continue to rip apart an animal that they have killed, it is extremely rare for a dog to actually eat it. Most dogs do not know how to “open up” a dead animal – this is why there are stories of dogs lying next to dead animals apparently guarding them or carrying them in their mouths. It is not because they are looking after the dead animal, but because they don’t know how to dissect it in order to eat it. Even some of the huskies I know that have killed and eaten “prey” have swallowed the animals whole, because somewhere along the line they don’t know how to eat prey properly. So, looking at the 4 types of dogs again:


ADOLESCENTS are the most “immature”. Physically they tend to have floppy ears, broad skulls and short (not flat) muzzles. In other words, as adults they appear simply as “super-sized” versions of what they looked like as puppies. Behaviourally, these dogs show very little predatory behaviours – they may rehearse various segments of the predatory sequence in play, but these behaviours are seldom used in any functional manner. If they do develop parts of the predatory sequence, these behaviours only emerge more seriously after the socialisation period. Adolescents are typically your livestock guarding dogs. They are fit for this task, because their long socialisation periods, with an absence of predatory behaviour, allow them to bond with livestock they are placed with and never rehearse predatory behaviour towards them. As a result, livestock have no cause to run from them.


OBJECT PLAYERS are just one step up in maturity, with floppy ears and generally a very playful attitude. They have retained a few parts of the predatory sequence, with an added modification: CHASE, GRAB-BITE – “CARRY/BRING IT BACK”. Retrievers and gun-dog types that fetch and carry game-birds fit into this category. They typically have “soft” mouths and a tendency to want to pick up and carry things.


HEADERS are more mature. Their ears may be pricked or just turned over at the tips and their muzzles are longer. HEADERS have very strong EYE-STALK-CHASE behaviours, which they tend to rehearse over and over again, stopping short of “grab-bite”. Herding dogs like border collies are typical of this type.


HEELERS are the most mature of the dog types. Some of them resemble their wild ancestors in body and head shape, but the main characteristic is the retention of the end parts of the predatory sequence. Headers may retain everything from ORIENT, EYE, CHASE, GRAB-BITE, SHAKE-BITE to KILL-BITE or they may simply have an emphasis on the last few parts of the predatory sequence “GRAB-BITE, SHAKE-BITE, KILL-BITE”. Everything from hunting dogs such as lurchers, more “primitive” dogs such as northern breeds (huskies) and small dogs used for pest control (terriers), would fit into this group.


So, why does all of this matter? How does it affect our pet dogs?

It matters because rehearsing the predatory sequence or parts of the predatory sequence is HUGELY reinforcing for dogs. When dogs rehearse these genetically programmed behaviours, it triggers the chemical reward cascade in their brains, ending in a rush of dopamine . When a border collie herds and a retriever retrieves, they feel REALLY good. They do not need any external reinforcement for this behaviour – it is hard-wired to make them feel awesome. Understanding this helps us to do the following:


  1. Accept the type of behaviours that the type of dog we have in our lives may find reinforcing and manage the environment to prevent dangerous or inappropriate rehearsal of these behaviours.

  2. Find appropriate, safe ways for our “type” of dog to rehearse these behaviours, to help them to get the “high” they need, without causing harm or endangering other people or animals.

  3. Use these inherently reinforcing behaviours in our training programmes to reinforce in a way that is MOST appropriate for the type of dog.


Rather than looking at recognising genetic predispositions in a negative way, understanding and accepting that a particular type of dog is likely to engage in a particular type of behaviour can help us to keep our dogs safe, keep the public safe around our dogs and find ways of enriching our dogs’ lives as much as possible. So, what does this look like in practise?


Let’s look at some examples:


The border collie is always one of the easiest examples to use when talking about breed/type specific behaviours, because their predatory sequence is SO obvious. Border collies are HEADERS and they love to EYE, STALK, CHASE stuff. From puppyhood they will start eyeing, stalking and chasing anything that moves. If one does not provide an appropriate outlet for this behaviour, they will find an inappropriate one – children, flies, bicycles, joggers, motor-bikes, cars and even shadows. Not being able to rehearse the behaviour leads to frustration and frustration leads to reactivity. Border collies that do not have sheep to herd, should be given opportunities to do things like chase a ball or frisbee, herd a gym ball or other similar activities. One should not be surprised when they chase moving objects and unless they have been habituated to these things, they may need to be kept on lead around traffic, mountain bikers or runners. In our training classes, I find that border collies love exercises where they can chase a toy and they love tricks where they have to circle objects and their owners.


When it comes to the OBJECT PLAYERS, this is a category of dog where understanding what “makes them tick” can really help to avoid behaviour problems and help them feel content. Retrievers and spaniels are two types of dogs in this category and it is no surprise that they are often the dogs referred to us for “stealing” stuff around the house or resource guarding. These dogs feel really good when they have things in their mouths. They like to carry stuff – in fact many anxious dogs of this type can be helped greatly, by simply giving them something to carry. Playing retrieve games is also really beneficial. Understanding why this behaviour tends to occur, means that we can train these puppies early to happily give up objects and prevent resource guarding.  


What about a less obvious breed:

What category does a Rottweiler fit into? Rottweilers have a mixed work history and can fulfil various tasks. Physically, they fit right into the ADOLESCENT category – their conformation changes very little from puppy into adulthood. They were originally livestock guarders and have many characteristics of these dogs – they are slow to develop, they remain extremely playful as adults and when they do display predatory type behaviours, these are usually quite “dysfunctional” and playful. Hunting for rocks is a common Rottweiler trait – many Rottweilers dig up rocks from the garden or haul them off the bottom of rivers and proceed to bat them around with their paws and pick them up and drop them, squeaking excitedly, as though they have “caught” something. This really is completely “useless” behaviour from a survival point of view and is purely entertainment. However, Rottweilers have also been used for herding cattle and with the right input, their EYE-STALK-CHASE can be developed into something useful for this purpose.


Finally, let’s look more closely at dogs that would be regarded as HEELERS: One example is the husky and other northern breeds. These dogs tend to have the “fullest” predatory motor pattern and many huskies that I know give testament to this with a “kill list” ranging from mice to bats, to seals to cats. Huskies are prone to want to go off and hunt on walks, which can make recall difficult. Most huskies also have zero interest in playing with toys – offer them a tug toy and they’re like: “Do you think I’m a stupid Rottweiler? I know that thing isn’t alive – show me the squirrel!” However, having a strong hunting instinct and a pretty functional predatory sequence, does not mean that a husky can’t be trained. It may however mean that the husky will require a lot MORE training for recall and an owner who finds creative ways to reinforce behaviours other than hunting. I have a good friend whose husky is an agility champion, the only one in South Africa. People thought she would never be able to let him off lead on a course, but he has exceeded all expectations, due solely to the enormous effort my friend put into training his recall. He is however, still very much a husky and when we walk on the beach with all the other dogs crowding around our feet, he is off in the distance looking for moles in the dunes and if a pigeon happens to cross his path a bit too slowly, it is, I am afraid, lunch. EXTENSIVE training has made him into a dog that has achieved great things, but his genetics cannot be completely erased and will emerge in the “right” circumstances.


I mentioned earlier that terriers would also technically fit into the HEELER category, because they display the end parts of the predatory sequence. Terriers are interesting in that their predatory sequence has been shortened, but in a way that leaves the middle parts out and exaggerates the end parts. The predatory sequence for a terrier typically looks like this:




This makes complete sense when it comes to dogs like Jack Russells that were bred to kill mice and rats. I am always fascinated when I see the single-minded determination of the Jack Russell Terrier at the stables where I ride, hunting around the haybales for rats. She will stare at a spot where she has seen movement for what seems like hours (ORIENT) and if the rat moves into sight it is dead in a second (SHAKE BITE, KILL BITE).


Unfortunately, though, human beings did not only select this shortened and deadly predatory sequence into small dogs like Jack Russells, they also selected for this behaviour in the larger bull breeds, which still also carry the name terrier e.g. Staffordshire Bull Terrier, Pit Bull Terrier, Bull Terrier etc. These dogs have the same predatory sequence, but they have been selected to target, not mice and rats, but other dogs or even larger animals, like bulls and bears. As unpalatable or offensive as it may sound, these dogs have been genetically selected to rehearse their predatory sequence on their own kind and other large animals. While it may be very hard to accept, a Pit Bull may derive the same pleasure from grabbing, shaking and killing another dog as a border collie does from herding sheep – it is the same area of the brain that is involved and the same internal reinforcement. I say “may” because these dogs may not always be in this frame of mind when involved in conflict i.e. they can of course also experience negative, defensive feelings like any other dog. The problem is though, that the predatory sequence is easily triggered in the conflict situation and once it starts being rehearsed, the defensive emotional state easily switches to one of pleasure-seeking and the dog can start to enjoy the behaviour and seek out opportunities to rehearse it in the future.  Now before I am lynched on the spot – please understand that I would no more blame the Pit Bull for this than I would blame the Border Collie for wanting to chase joggers or the Spaniel for “stealing” socks. All these dogs are what WE have selected them to be. Human beings intentionally and deliberately set about “creating” a dog for this purpose. We are ultimately responsible. However, denying this reality and pretending that it does not exist, continues to lead to tragedy, which we see in South Africa on a regular basis. In this country dog fighting is sadly a thriving business amongst gangs involved in other illegal activities. Dogs used for this cruel “sport” are selected for their “gameness” i.e. their inclination to fight and to become hooked on this behaviour. Worryingly, many of these dogs and their offspring are rescued by pro-life organisations and placed in family homes with ultimately tragic results. The constant pleas by the Pit Bull Federation regarding these dogs and the potential danger they pose to other animals, seems to fall on deaf ears:    We have to stop sacrificing common sense and hard truths for the sake of political correctness.


The question is always asked: “Surely training and socialisation are more important than breed tendencies?”. The answer is that while socialisation and training is important and can help to mitigate and channel genetic tendencies, it will never erase them. A few practical examples come to mind:


My friend with the Husky also has a Border Collie. On our walks together, both my Rottweilers and the collie will chase and fetch a ball. However, 90% of the time, my Rotties always get the ball first – the reason being that the collie cannot help but run in a herding arc around the other dogs on his way to the ball, wasting precious time while they run straight to the ball. It is clear that in this scenario, herding the other dogs is more fun than fetching the ball, because it is more in keeping with his natural behaviour pattern. The apparent lack of reinforcement for not getting the ball pales in comparison to the internal reinforcement of carrying out the herding behaviour pattern.


Another example is two dogs I know who were brought up with the family cats, eating with them, sleeping with them and grooming them. One day, one of the cats suddenly ran past the dogs and they grabbed, shook and killed the cat in a second. There was no animosity towards the cat and I am sure they had no idea what they had done. It was pure instinct. The cat in that moment behaved like prey and the social bonds were forgotten as the stronger instinct in that situation took over. While we may do everything in our power to socialise and train our dogs, we have to be realistic and accept that at the end of the day they are still dogs and if the “wrong” circumstances arise, instinct might just kick in and override learning.


While most of this article has focused on the predatory sequence, something I also want to mention briefly is the impact of genetics on behavioural responses to a threat. While defensive aggression is a normal survival behaviour that any healthy animal should be able to use when circumstances require it, genetics does have an impact on how quickly an animal responds to a potential threat and the strategy they employ to deal with it. We know for certain that fearfulness is a heritable trait: We also know that some types of dogs have a shorter or longer socialisation period, which tends to have an impact on how easy it is to socialise them well, so that they become non-reactive as adults. While the average dog’s socialisation period ends at around 14-16 weeks of age, dogs like Labradors and Beagles seem to stay within this period until around 20 weeks, while German Shepherds have a very short socialisation period, ending around 10 weeks of age:  This absolutely will have an impact on the ease with which the average owner will be able to end up with an easy-going, non-reactive dog.


Aside from this, there is also the factor of how a dog reacts to a threat. Animals in general have 5 options in response to fear: FIGHT/FLIGHT/APPEASEMENT/FAINT/FREEZE. “Faint” is only used by a few species, such as possums, while “Freeze” is generally an initial response which allows the animal time to process the situation and decide on further action. Horses and cats are generally “Flight” animals, while dogs may employ FIGHT/FLIGHT/APPEASEMENT depending on the situation, past learning (what has or hasn’t worked before) and, I believe, genetics. Think about your typical Golden Retriever for a minute – what picture comes to mind? For me, it is lots of wriggling, waggy tail, open mouthed grinning and rolling over. Most people view the Golden Retriever as the classic, good-natured family dog and one of the reasons that I believe this is so, is because their default response to a threat is appeasement. Unfortunately for Golden Retrievers, this often results in people failing to recognise that rather than just being friendly, the dog may also be insecure and anxious. In contrast to this, think about your sight hounds like whippets and their general response to a threat – avoidance or flight is a very successful strategy for these fast animals and seems to be more their default response to unpleasant encounters. On the other hand, what is the response of the average terrier when confronted by an intimidating large dog – in a majority of cases, it is “fight”! Jack Russells are notorious for what is often fondly termed “cheeky” behaviour. I really don’t think it can be denied that certain types of dogs are more predisposed towards using one of these options rather than the others. Regardless of how well-socialised and friendly my Rottweilers are in their day-to-day interactions with people, try to mug us on the mountain or break into our house at night, you are going to get a very different reaction from them than you would get from the average Labrador. Rottweilers have specifically been selected to “take on” a threat rather than run away. Running away would be seen as a fault in the breed and their willingness to advance towards a threat is actually tested for in the Working Dog Aptitude Test as well as the Rottweiler Breed Assessment.


Does this mean that responses to fear cannot be modulated through socialisation and training? Absolutely not. In fact, I strongly believe that fear responses are far more flexible than the predatory sequence and much more subject to alteration through training and socialisation. The socialisation period hands us much of the answer to addressing innate fear responses on a silver platter. Regardless of which strategy a particular dog or type of dog may be naturally inclined to employ in response to a threat, extensive socialisation can eliminate the dog’s need to respond at all in normal situations, because the dog will not view things in everyday life as a threat. In addition, because all dogs have the option of employing other strategies than the one that they may be predisposed to, we can actually reinforce more socially acceptable strategies such as appeasement and avoidance (mild version of flight) by ensuring that we notice when our dogs are uncomfortable and responding accordingly. Savvy owners can encourage dogs to choose avoidance, by ensuring that it is successful (allowing the dog to walk away from the scary thing and making sure that the scary thing does not pursue the dog further) and heavily reinforcing it with praise and food/play. For example, you notice your puppy is looking uncomfortable because another dog is staring at them, so you respond by calling your puppy away with you calmly and praising like mad and jackpotting with food when he comes. The catch, is that owners of dogs who are inclined to go with the “fight” option, need to be skilled and educated in order to pull this off – which means that breed or type of dog is relevant when choosing a dog. I don’t think there is any trainer, who would deny that certain types of dogs are not suitable for everyone.


In conclusion, while “how you raise them” is incredibly important, a dog’s behaviour is not solely determined by this. Many of us choose specific breeds or types because we enjoy their behavioural traits. Denying that breed or type has any impact on behaviour not only denies the negative stereotypes that breed advocates try to avoid, but it also denies all the positive attributes that so many of us deliberately seek out as well as the opportunity to use our knowledge of a breed or type to help that dog to reach his or her full potential and lead a fulfilled life.  I am not claiming that every dog of every breed is the same or that a dog will always behave according to type. Of course, there is variation within any population group and there are always exceptions that may not fit the mould. But denying that there are breed tendencies or that genetics does not matter is just silly. Part of responsible dog ownership is recognising breed traits, so that one can make a sensible choice when bringing a dog into your family and set yourself and your dog up for success.




[i] Coppinger & Coppinger, DOGS: A New understanding of Canine Origin, Behaviour and Evolution, Chicago Press, 2001, p. 57-62

[ii] April 1982, Smithsonian Magazine, Developmental Stages of the Dog


The subject of genetics and the effect of genetics on behaviour comes up a lot. Many people will happily accept that a terrier may be prone to digging or a border collie is more likely to herd things, because that is what they were bred to do. However, when it comes to things like aggression, people become very touchy at any inference that behaviour may be linked to genetic predisposition. This article is not about assigning “good” or “bad” traits to any breed, but is rather an honest discussion of how genetics influences certain types of behaviour and how understanding this can be helpful in training, managing and fulfilling the needs of the dogs that we live and work with.

This video shows how two different types of dogs may appear to be engaging in the same behaviour, yet their intent can be quite different, due to their breed predisposition. Here Rosie and Cruz, the Rottweilers, and Zeus and Solar, the huskies are all digging. However, Rosie and Cruz are digging around their balls for fun, while Zeus and Solar are digging in a mole hill to try and catch a mole! This shows the difference in "maturity" of these two types of dogs. One type is just having fun while focusing predatory behaviour on a ball in play, while the other type is rehearsing more serious and functional predatory behaviour. Looks the same on the surface, but it's quite different!

bottom of page