Jim Ha, PhD, CAAB
The modern science of animal behavior, which we call ‘ethology’, has come a long way in the past few decades, from a largely observational, descriptive science to a modern, quantitative science based on solid foundations of evolutionary biology and quantitative methodology. One of the most common situations in which I realize this is when I see, read, and hear old, out-dated animal behavior concepts and ideas and long-ago-rejected hypotheses used by pet animal behaviorists. Many trainers and veterinarians received whatever animal behavior education they might have gotten long ago, and often have not stayed up to date. As a professional and academic ethologist, I of course have the time and professional need to peruse the latest journals, read and review the latest textbooks, and make sure that my university courses are up-to-date. But when I enter the world of companion animal behavior, I am often taken back to a time long, long ago, to terms presented even to me in my long-ago introductory courses as historical concepts, mistakes, or simplifications used only for pedagogical purposes.
Let me give you a few examples: I think that they fall into four categories. There are concepts that we now know simply are not true: Fixed Action Patterns fall into this category. Fixed action patterns were a concept from the early days of ethology, mid-20th-century, in which animals were thought to display some behaviors that were absolutely of genetic origin, the classic “hard-wired” behaviors (another term we are trying to dump!). When gull chicks saw a red dot on the parent’s bill, they “automatically” gaped their mouths open to get food. The behavior was thought to be fixed or unchanging, burned into the animal’s brain pathways by their genes, and never to be changed or modified in any way. But then we discovered that they could get better at it, quicker or more discriminating as they grew older: wait, that’s not genetic, that’s called learning! So the behavior was NOT fixed. Finally, the field concluded that there really was no such thing as a fixed action pattern. We went through the usual struggle of dying concepts, modifying the concept to death, trying to keep it alive, before finally burying it and moving on. But I still hear trainers referring to highly stereotyped, very ‘fixed’ behaviors performed in response to a clear signal as a Fixed Action Pattern, or at the very least, as being ‘hard-wired’, whereas we now know that there is no hard-wiring. It immediately dates the user of the term to a certain generation and suggests a lack of later education in ethology!
There are concepts that are sort-of, basically true but that we now know are far more complicated, and thus the original terms and concepts simply don’t do the job any more: the Nature vs. Nurture dichotomy falls into this category (as do many dichotomies: the world is not a black and white place). This issue is related to the concept of the fixed action pattern, but this one has taken longer to die. Why? It’s the caveat about being kinda, sorta right. But here the point is that it is not a dichotomy. No matter where we look, it’s a continuum, from highly genetically controlled (but never entirely: see point above) to virtually entirely environmental, what we in applied animal behavior would call ‘learned’, but again, never without a genetic component to the behavior. So now we use terms like ‘genetic predispositions’ or ‘strong learning component’ and really talk about these two ends of the continuum as explaining a proportion of the variability in a trait, a concept that is on a continuous, not a dichotomous, scale.
Another example of this continuum issue that I hear all the time is the question of whether dogs have a ‘dominance hierarchy’ (beyond the fact that most companion animal behaviorists are working with VERY old ideas about what a dominance hierarchy is: another blog topic altogether!). So all dogs, individually, and all dog breeds, have to either possess (be influenced, include in their behavioral repertoire) a dominance-based form of social structure, or not. Wolves have a dominance hierarchy-based social system and dogs do… or do not, depending on your point-of-view. OK, first, the existence of dominance hierarchies in wolves (as well as numerous other species) varies on the basis of ecology, their lifestyle, prey, metabolic needs, etc. So it’s not open-and-shut for wolves. Now for dogs: the facts show clearly that the use of dominance, the importance of dominance hierachies in the life of, even the ABILITY to communicate such information, varies widely from breed to breed, due to our artificial selection of this, and more often, other related traits, like coat color, hair type, and temperament. So it’s a continuum: for some breeds, social structure is VERY important; for others, it can influence their behavior; and for yet others, they can’t even recognize these signals. A major disaster that we see regularly in metropolitan dog parks is when the first type meet the third type, without supervision: but again, perhaps a topic for another blog. And the same goes for the common argument that while dogs may have social structure, they don’t include humans in their hierarchies… again, as an across-the-board claim, it’s clearly wrong. Some breeds do, some breeds don’t.
Back to other old animal behavior concepts and terms: there are also concepts which were originally just that, only a concept, and for which we have never discovered a mechanism. A nice idea, but without a mechanism, just not useful anymore. These are often terms and concepts that describe a phenomenon functionally or descriptively but which have no mechanistic, explanatory power behind them: an example is Instinctive Drift. Unfortunately, this term has led to a term more commonly used in applied animal behavior, Predatory Drift, based loosely on the idea of instinctive drift, descriptive of something that we see occur but without any mechanistic, useful purpose. The same behavior can be more clearly and with a stronger empirical basis explained in other ways.
Another example of this phenomenon that I ran into recently is the “Hydraulic Model of Drives,” created in the mid-20th-century by the father of modern ethology, Konrad Lorenz. It was actually called the Psycho-Hydraulic Model, and many believe that it was actually a little ethology joke: the description of his model was based on a European toilet system, a water closet (geeks are just SO funny!!). But for a decade or two, it was tested as a model, but by the end of his career, in the ‘70’s, even Lorenz had abandoned it for more powerful models, models which fit the data that we were collecting. So this model has taken its historic place, and certainly acted in a heuristic fashion to generate considerable discussion and much experimental work before being rejected in the end. But it lives on in the literature and discussions of some dog trainers!
Finally, there are concepts that are simply made-up, never really true, often derived from wherever those old wives produce their tales: a classic example of this is the ‘alpha roll’ or ‘scruffing’ of dominant wolves towards their subordinates, and thus assumed by some trainers and veterinarians to be a useful concept in the domestic dog. Domestic dogs are evolved directly from gray wolves: the evidence is quite clear on that now, but there is no such thing as alpha rolling, in wolves or domestic dogs… so why do some trainers keep using it? It’s punishment, pure and simple.
So let’s kill some of these useless concepts and terms: Fixed Action Patterns, Nature vs Nurture and hard-wired behaviors, Predatory Drift, and the alpha roll. Let’s get modern: pick up a good textbook in animal behavior… I have a great reading list available. Don’t date yourself… we ethologists have come a long way in a relatively short time, and companion animal behaviorists have much to contribute to the modern science of ethology, but to do so, we have to stay up front with the science. Contact me if you are interested in more ways to do this: if there is enough interest, I’ll write a future blog about it.
Wow. Great post.
I see the “alpha roll” all the time with dogs that show too much excitability in obedience class for the handlers’ taste, and the handlers (and class instructor!) never seem to catch on that the more appropriate and effective procedure would be to positively reinforce an alternative behavior. When the dog doesn’t get better, they increase the intensity of the roll, thinking that the reason the dog isn’t getting better is because they aren’t rolling the dog frequently or strongly enough. It’s sad.
Thank you very much for the informative post!
The thing that interested me most is the relations btw the use of dominance hierachies and the ability to communicate such information and breed, coat color, hair type, and temperament.
Where could I possibly read about that?
Christine Hibbard says
Dr. Jim Ha has written two other articles where he suggested additional reading. A Reading List in Animal Behavior Part One: https://companionanimalsolutions.com/blogs/a-reading-list-in-animal-behavior-part-one/
A Reading List In Animal Behavior Part Two: https://companionanimalsolutions.com/blogs/a-reading-list-in-animal-behavior-part-two/
Thank you for reading Behind the Behavior. Happy reading!
Simply want to say your article is awesome. The clarity in your post is simply impressive and i can take for granted you are an expert on this subject. Well with your permission allow me to grab your rss feed to keep up to date with forthcoming post. Thanks a million and please keep up the gratifying work.
Leonard Cecil says
Extremely interesting. Especially about FAPs. James O’Heare prefer the term Modal Action Patterns and define it thusly in the AABP glossary: “Modal Action Pattern. A sequence of behaviors, which is relatively invariant and is considered relatively innate, activated by a specific environmental stimulus. Used to be called fixed action patterns implying that the response is fixed and unchangeable.”
Now I’m wondering about humping. This has been a cause of endless discussions as to what humping is and what it isn’t. Having a female humper myself,the experience I’ve had is, that if I can catch her as she’s getting into position, I can call her out of performing the humping. Once she’s started humping, I’ve got a 50% chance at most, practically only if the humpee shows specifically that he/she doesn’t want to be humped.
Now if we now do NOT classify this as a FAP or a MAP, does it hold to the definition as set forth by the AABP? Or is this definition also dated. Your article was great in explaining why former FAPs aren’t anymore, but I missed an explanation as to what they ARE. I don’t really need to classify humping as this that or the other for myself. I work with the onset, since frankly I do not see any kind of antecedent stimulus that my dog does. Why she does it with one dog but not with another. And she does it only very sporadically, the sex of the other dog playing no role whatsoever. Moreover, when helping a client deal with this, what is the best way to accurately describe to them the whats and whys as part of detailing what we can do about it?
thanks in advance!
Greta Kaplan says
The only one of these four terms that I want to talk about (right now, anyway), is “predatory drift.” This is the one term that did not in any way come out of ethology — old, new or otherwise. As I understand it, ethology textbooks do not address it, so I am not sure reading ethology textbooks is going to resolve the questions and concerns I have about this term and what it describes.
Before moving on to my plea, I would like to share the history of the term, which I have researched and to which you briefly alluded. The term “instinctive drift” was coined in the famous 1961 Breland and Breland article, “The Misbehavior of Organisms,” to describe an observed phenomenon. Whether it was an unfortunate word choice does not change the observation that certain behaviors were much harder to train using operant conditioning, and appeared to be “interfered with” or “hijacked by” species-specific action patterns. (Notice I do not use the word “fixed” or “modal.” They are actions in a pattern, though!) The Brelands were behaviorists in the old sense — they had studied with Dr. Skinner and had a successful business using applied behaviorism (i.e. operant and classical conditioning) to train multitudes of animals to do multitudes of interesting and difficult behaviors. I have confirmed with their last surviving colleague, Bob Bailey, that he does not have any recollection of the variant term, “instinctive drift” being used in the Breland era, or by himself thereafter. (Personal communication, 9/10/2008.)
Years later, apparently Dr. Ian Dunbar threw out the term “predatory drift” quite casually to describe an apparently related, observed phenomenon — dogs who are engaged in social behavior and switch abruptly into predatory behavior (often with deadly results) toward a member of a species with which the dog had been socialized — dogs, sometimes people, and quite often cats. Jean Donaldson (who is careful to note that “predatory drift” is not a sanctioned term and that the underlying phenomenon has not been studied and reported on in a scientific fashion — see her video “Predation in Family Dogs”) heard this and was the first to start discussing it in public, in more detail. So there are really no ethologists in the history, although Dr. Dunbar did a substantial amount of dog ethology long ago when in graduate school.
I suspect I can identify one reason this phenomenon (whatever you call it) has not been formally studied, analyzed, and labeled by ethologists. It is a behavior that is exhibited by domestic dogs, and domestic dogs have received extremely little attention from ethologists, in the past. (This is changing!) I believe this phenomenon is, in fact, almost by definition going to be limited to domesticated predators. This is because any non-domesticated (large) predator who kills a small dog, cat, or child will be said to have predated. Not being domesticated means that “socialization with humans” is a much more limited effect. Only dogs have been persistently and intensively bred exactly for the ability to socialize with humans in a way no other species can. So, a behavioral phenomenon that seems to be basically limited to dogs has had very little chance to be studied.
Regardless of whether “predatory” and/or “drift” are unfortunate choices of terminology, there clearly IS a a phenomenon at work here. We see behavior that *looks like* dog social behavior (e.g. lots of social signalling such as turning sideways, play bows, “sexy ears,” rolling over, barking, growling, very inhibited playful bites to the head and neck, or pokes to the flank in some breeds, and so on) be rapidly replaced by behavior that *looks like* predatory behavior (silence, very contained stalking, staring, straight and rapid rushing toward the target, very hard and damaging bites to the loin or underside or a very hard grab and shake resulting in a broken neck). In fact, it appears to me that pit fighting in game bred pit bulls resembles this predatory behavior and not social fighting behavior.
If this isn’t some sort of abrupt “switch” into a predatory motor pattern, what is it? There has to be some kind of genetic component here. (I realize there could be a genetic component that doesn’t involve predation, etc., but still, what is it?) Experienced trainers know the breeds most likely to engage in this type of behavioral shift — Siberian Huskies are one good example, but we see it in some terriers such as the Jack Russell, pit bulls, and Fox terriers. I also see it in Dobermans, Rottweilers, German Shepherds, and many of the Belgian Shepherds, though probably far more in protection-bred Malinois than in the other Belgians. This is my observation — I’d be thrilled for a more systematic study on breed frequency. If there are breed tendencies on this, as appears to be the case, then there is a genetic component. Interestingly, I think this range of breeds is quite varied in how close to wolf-type dominance behaviors they tend to show, and it is hard to rely on, say, “offended dominance notions” to explain the phenomenon. So if it’s not some sudden shift to actual predatory behavior, what is it?
Since this is a low-frequency but extremely high-stakes behavior, it needs study for practical reasons. Trainers need to be able to know roughly what it is, what it looks like, and how likely it is to occur with a given dog. Can we make guesses based on breed? Prior history? How reliable are our typical guesses about what will trigger it? The “cat is a friend inside, but dead from a neck shake when it runs outside” scenario is very familiar to experienced dog trainers. And so is the apparent trend for dogs killed in predatory-like attacks to be small and often white, often fluffy, and squealing and/or moving rapidly. Sometimes it seems to be a child who triggers it. This is a very tough one to get across to clients, because apparently, it simply does not matter how much the dog “likes” or “is friends with” the other dog or child at other times. It does not appear to correlate *at all* to fearfulness. Getting people to understand that “he’s doing better with my Chihuahua now” may not affect the chances of him killing the same Chihuahua later is very tough. Having an explanation for the *why* of this behavior would be useful in these cases.
In terms of scientific interest, well, I think it’s pretty interesting. How does a brain handle the tension between later-evolved social behaviors and early-evolved food-acquisition behaviors? Other than kittens provoking parental nurturing responses, cats don’t have to deal with other cats that are one-tenth their size, so the question of “prey buttons being pushed by an adult conspecific” is just not likely to come up. Dogs are unique in this multi-species social group affiliation and this behavior pattern seems to me to reflect a tension there. Or I could be wrong. But I’d really like to know.
So on to my plea. Dr. Ha, you *are* an ethologist! You, more than any of us, are in the position to possibly study this, or get it studied. It’s important. It’s interesting. I vote for further study. Thanks!
Leonard Cecil says
Thanks Great! I’ve read the Breland article. As you might be aware, this article is floating all over the internet, posted by compulsion trainers as proof, that so-called positive methods involving OC/CC don’t work.
Be that as it may, it is an interesting insight into the state of research at that moment and it’s development up to Dr. Ha’s blog. I also found interesting your breed list of dogs prone to “predatory-like attacks” like attacks. I believe it’s Trish King in her excellent DVD set Different Breeds, Different Needs who mentioned in her area having seen a spate of Golden Retriever who’d killed smaller dogs, usually Chihuahuas.
When you appeal to Dr. Ha to start some research, I’d like to also place a request:
I’ve looked high and low for some peer reviewed research on the so called “Calming Signals” which of course includes all the nomenclature variants and extension so many people have developed (appeasement, distance gaining, etc.). I found one dubious Italian study that has pretty much been discounted due to methodology problems. As far as I can tell, we’ve got mostly Turid Rugaas and her personal work and very little else of a substantial, recognized scientific base, yet we all speak of these “signs” or “signals” without really having any formal study having been done as to cause-effect. There are for example groups who say, that these “signals” are in no way intentionally used by dogs to signal anything, they are simply reactions to certain stimuli and are nothing more than an expression of inner feelings/conflicts. Others are completely convinced that dogs intentionally use these in hierarchical patterns from mild through full blown aggression to achieve their goals of distance (increasing or decreasing). Are these people basically correct (who? some? all?) in their anecdotal observations? Why can’t these signs/signals be systematically be studied?
Leonard Cecil says
P.S. – Fixed Action Pattern.
The AABP Glossary writes about Modal Action Pattern (After writing about FAP “see Modal Action Pattern”):
“Modal Action Pattern. A sequence of behaviors, which is relatively invariant and is considered relatively innate, activated by a specific
environmental stimulus. Used to be called fixed action patterns implying that the response is fixed and unchangeable. ”
In an assignment for James O’Heare during my CASI course of study he wrote, that a better description of such behaviors (as humping) or retrieving (in a retriever breed) would be a “General Behavior Trait”, which is defined in the AABP Glossary as
“”General Behavior Trait. Tendencies (as opposed to specific behaviors) strongly influenced by genes. Includes activity level, aggression, introversion and anxiety. More flexible than fixed action patterns (Chance, 2003, p. 17).””
Is this the current stand or do I get myself in trouble using that?
Marilyn Wolf says
What’s the link to the Part 2 reading list?
Marilyn Wolf says
Never mind. I saW IT ABOVE.
James Ha says
Well, what a wonderful response to this blog!! It’s so nice to know someone is reading these!!
Let me respond to several of these comments in one response here:
FAP’s or MAP’s: the terms are being misused in either case, in the example provided by LC. In either case, as pointed out O’Heare, these terms refer to a sequence of behaviors, performed in stereotypical fashion, triggered by a specific and characteristic stimulus called a sign stimulus, and having a strong genetic basis. The term Modal Action Pattern was suggested to replace FAP in 1986 by George Barlow, to imply a less rigid requirement for stereotyped patterns of behaviors, given that we could not identify any behavior sequences that fit the definition of a FAP. Here is the definition from Paul Chance’s Learning and Behavior Sixth Ed.: “A Modal Action Pattern is a series of interrelated acts found in all or nearly all members of a species. Resemble reflexes, and have a strong genetic basis.”
The more recent term which has replaced a FAP or a MAP is simply a stereotyped motor pattern or an even more up-to-date term, a motor pattern with a high heritability. But in the case of humping, this behavior does not fit the definition for several reasons: it does not constitute a sequence of behaviors, there is no triggering sign stimulus, and it is unlikely that there is any genetic basis to the behavior. Just because the behavior is stereotyped does not make it a sequence of stereotyped behaviors or genetic in basis. In fact, it is not likely to be genetic, or most animals of that species would perform the behavior… which they don’t… thank God!
Humping is a pathological behavior, resulting generally from anxiety due to lack of proper early socialization and is used to self-soothe in stressful situations. It’s NOT a FAP or a MAP, but a stereotyped behavior (as most behaviors related to reproduction are) triggered by internal anxiety levels… quite different.
The term “instinctive drift” was coined many decades ago by comparative psychologists, and refers to what we call today phenotypic plasticity. That is, animals are constrained by their genotype and early experience, their biology, in what they can learn and the degree to which they can learn it. Thus, an animal may not be able to learn certain associations, or may not be able to retain some associations as long as others: there are biological predispositions, as we say today, for certain behaviors, including learning. So again, a once-useful term has been updated.
But the real discussion is about “predatory drift.” There is no such concept in ethology, and the term doesn’t even fit with the idea of instinctive drift in my mind. Are the users of this term suggesting that the default behavior for domestic dogs is predation, that when play or social behavior is performed for a while that the tendency for dogs is to revert to predation (as animals are supposed to revert to innate behaviors in instinctive drift)? This whole concept fails for me.
Regardless, there are numerous reasons why a dog might “appear” to be playing or behaving socially and suddenly attack a smaller dog, reasons that may or may not include predation. A major reason is that many (most?) small dogs do not communicate in the same was as more primitive breeds of dogs, a point driven home in Erik Zimen’s excellent work with wolf and poodle comparative behavior (and more recent studies) and reinforced by our modern knowledge of genetic relationships among breeds. When you are interacting with another dog, you throw a lot of body language at them, to check them out, to determine their emotional state, to anticipate their intentions, and when you receive nothing coherent back, or perhaps even (inadvertent) signals of anxiety, aggression, or weakness, a dog of certain temperament might react with defensive aggression… this is just one common example that I have seen in my cases.
I really don’t see how the Bolles work on instinctive drift, or the Breland article popularizing that work, or for that matter, the more modern understanding of phenotypic plasticity (or constraints, if you are a “glass is half empty” type) can support or weaken the compulsion (or confrontational, as we call them in applied animal behavior research) trainers. The concepts would apply equally to any kind of learning…
Now for “calming signals”: another term not part of the animal behavior literature, or certainly not in any form similar to how they are used by trainers and poorly trained behaviorists. I have heard of all sorts of behaviors in this category: displacement behavior, redirected behaviors, reconciliation (the new form of appeasement) behaviors, vacuum behaviors, behaviors intended to reduce anxiety in the performer, behaviors intended to reduce anxiety in the recipient, just an incredible mish-mash of stuff. There is no established use of such a term, and in the uses of the term “calming signals” that I have heard, I have not even detected a clear or consistent understanding of what calming signals are, how they are defined or exhibited. So the term is useless: why not use the terms in standard usage? In some cases, calming signals are displacement behaviors, in some cases they are redirected behaviors, in some cases they are self-soothing behaviors, etc. These are established and well-studied concepts in behavior. “Calming signals” is another good addition to my collection of bad ethology!
So I hope that this reveals a bit more of modern ethology… there is a real science out there, as accessible, or inaccessible, as any other science like genetics, organic chemistry, and ecology. Some evolution of terminology is inevitable but usually reflects new knowledge and new discoveries, and good-quality education, and then keeping up-to-date with ongoing new discoveries, is the key to tapping into that new knowledge to benefit yourself, benefit your clients, and especially to benefit the companion animals on which we are focused.
Leonard Cecil says
Thanks for the exceedingly interesting reply – will print it out later to digest it further. On the matter of humping though I did read and re-read and in several cases – specifically in the case of ALL my dog’s prothers and sisters (sorry, English just failed me), I find this rather unsatisfying.
The breeder is one of the best here in Switzerland and that is not something to be taken lightly. Terrific quarters for the pups in her house with dozens of different type of flooring for the pups to get used to, 2 other dogs and three cats in the household, play time inside and outside. change of environment every week or so, meaning tearing down and building up new quarters inside and play areas outside. social walks as a group in town, car rides, people (future owners, family members neighbors school children (she^s a biology and geology teacher). After getting the pup at 10 weeks, despite all of this, the first job was to desensitize the pup to the car, sine she SCREAMED the who way home 45 minutes, even though sitting on lap. now she’s ok with the car. Humping started at approx 9 months. intermittent, depending upon the other dog. Back and forth one after the other with her bros and sisters, None whatsoever with other dogs. Starts within 30 seconds with others. fortunately, if the other dog shows that this behavior is not on, she stops.
1) Interesting, that ALL her bros and sisters show this behavior.
2) extremely sensitive example of a Flat-coated Retriever. Was people and dog reactive (not aggressive, but rather completely and utterly over-the-moon friendly. Look-at-me has gotten this pretty much under control, no longer a problem with people.
3) puppy classes and then 6 months of adolescent dog classes. Well socialized with other dogs – but still typical Flattie with their typical manner of greeting.
4) the fact that, as in all her family, the reaction to the other dog does seem to make a difference as to whether the behavior starts or not and then whether it is continued cannot be ignored. The other dog must be supply some kind of antecedent stimulus, but I am not seeing it. I don’t have your experience or eye for this.
5) I can redirect her about 9095% of the time if she’s not completely into the act. If she is, too late.
6) I really wonder about humping not being genetic, after all, it is, whether meant at that moment or not, an act of procreation, which under “normal” circumstances is necessary to pass on the genes. If it were a learned behavior, not a genetic one, Although I’m not a behaviorist as you are, I do understand the basic mechanics of ABC. There must be some kind of trigger in play that, combined with the successful completion of the act as positive consequence is keeping this alive. But the behavior itself wasn’t learned in the sense that she didn’t know how it goes until someone or another dog taught her.
7) As a humpee she is extremely sensitive to what dog tries it with her. She’s been “raped” a couple times in dummy classes, such that she stopped going after the dummy, because the “rape” happened just as she left the line to get a dummy.
I know you used the term “generally” and that may be applicable. OTOH, if it is a learned behavior, then how does the male dog learn to hump to reproduce? Especially if he’s never been in contact with another humping male? I suppose many are about as successful or unsuccessful as other males of other species. (grin). But then, if it is a learned behavior, it should be relatively easy to “unlearn”, like umping up, either by redirection or substituting an incompatible behavior. Of course, being that it’s a self-rewarding behavior, that makes it more difficult – there’s probably not much I can offer the dog that will exceed the reinforcers it procures for him- or herself by humping. But the fact that it doesn’t seem redirectable once it’s fully started causes me to question this – as if at that moment the brain checks out and a ??? sequence of actions takes over.
Now either I’ve got it all wrong, or there could be some basis to a genetic action being carried out, one anchored in reproduction, but abducted, yes very likely as a result of over anxiety or over arousal.
Or … ?
James Ha says
You need to be very careful when you take the following data: “all of her brothers and sisters do it” and draw the conclusion that it is somehow genetic: I would immediately use the scientific process of Occam’s Razor and suggest that all of them were treated in the same way at a young and formative age, hence there is a strong relationship in how they behavior now, and this may be what we are seeing here.
Now, let me be clear as well: “humping” (copulatory behavior) is, obviously, quite genetic for clear evolutionary success reasons… get it wrong and those genes don’t pass on! But what we are discussing here is inappropriate RELEASE of the behavior. And let me also be clear: (inappropriate) humping is a pathological behavior used as for self-soothing of anxiety, and ANXIETY may (in fact, sounds likely) is genetic in this family line. You speak of extreme anxiety at a young age, in the car, for instance, very unusual. Also, the humping starts at about 9mo, you say: exactly as I would expect if it’s a response to anxiety and fear… 7-9mo is the neophobic critical period at which many fear responses appear in dogs.
So my (admittedly long distance) diagnosis of the humping behavior in this case would be inappropriate manifestation of profound anxiety, probably of genetic origin. Not genetics for inappropriate humping but genetics for anxiety, which is manifesting itself in inappropriate humping. Why humping? One major factor is reproductive status, which I don’t think you mention: unneutered dogs and dogs neutered at too young an age are more likely to manifest anxiety in this way.
Hope this helps,
Jim Ha, PhD, CAAB
Jen Robinson says
Great article. Great discussion. Thanks.
On humping . . . I appreciate the need to debunk misconceptions, but I don’t see why the alternative hypothesis is ‘profound anxiety’. At times I’ve had four brood bitches (Labrador Retrievers) living in my house. They humped one another, occasionally, most often an older bitch mounting her daughter or granddaughter, but the unrelated girl did get involved, and the pups sometimes mounted their elders. It never resulted in fights, and wasn’t disruptive or very common, so I let the behavior go. If one or more came on season it became a much more vehement and frequent activity, and at times they end out in conga lines of three or four. I’ve heard similar stories from other breeders with other breeds. Why not an alternative hypothesis that dogs have sexual energy and act on it . . . not always in a procreational way? When hormone levels are higher, the activity increases. You could call it sex play, at the risk of being accused of anthropomorphism. Humans try to inhibit this behavior cause it’s not good human manners, but in the dogs I’ve owned it seems natural as natural and relaxed as other ‘inappropriate’ dog behaviors, like sniffing bottoms. I don’t see why there will be ONE answer. Sexual energy certainly gets entwined with anxieties in humans, and the equivalent could happen in dogs.
To make sense of humping I think you need to pay attention to age, reproductive status, persistence, degree of familiarity between dogs, owner’s reaction, history of behavior . . . perhaps breed . . . and perhaps other things as well.
Lesley Thorpe says
Thank you so much for such an informative article – just fascinating. Have run a large multi-generational pack (up to 20) Great Danes for many, many years, all with free access to the house or paddock as desired. Stud males usually kept separate in other rooms and rotated through each day, although up to 4 related males from certain lines could be kept out together. Any puppy born here just slotted in, but it has never been possible to introduce an outside puppy with the same ease or lifelong trust. I noticed that when introducing a French Bulldog that neither breed could “read” each other very well – the signals and body language are completely different. Although now the two breeds seem to have learnt to read each other very well. Tail vs. no tail would be an interesting study in itself. So pleased to have something to follow up.
James Ha says
To respond to Jen…
Sorry, I should have been clearer: 90% of the dogs out there are spayed and neutered (thank goodness) and so my comments referred to them. In your case, you are referring to reproductively intact dogs, and the story is entirely different. Your comments and conclusions about sexual drives, the role of high levels of hormones, play behavior in practice for real procreation are all valid interpretations of misdirected copulatory behavior in intact animals, especially groups of intact animals. My comments referred to the majority of cases we see which are neutered animals.
Leonard Cecil says
JAmes, somehow I miss your reply on May 27th.
Let me fill in some blanks. We picker her up at 10 weeks of age from the breeder. At the time, the breeder told us that she had a problem with cars that surfaced when she took her mom and grandma together with all the pups to the Swiss Retriever “pope” who is also a Vet for examinations. Ours was the only one who had a problem in the car, screaming the whole way and throwing up in the car. She’s one of a litter of 8.
She is unneutered and will remain so, inasmuch as she has a beautiful coat and in brown flatties, many loose their coat quality when neutered. Her last heat went off without a htich and almost no false-pregnancy symptoms. We would re-evaluate this decision iif we determine that she has real problems with her heats, but we have no plans to breed her. We’ve kept in touch with the breeder and with some of the guardians. All guardians have reported to either the breeder or us directly of their dogs being, even for Flatties extremely sensitive and varying in descriptions from nervous to extremely over-friendly.
Your explanation about anxiety/fear would make sense. Usually an enforced but positively introduced time-out usually will help keep the episodes to a minimum. I’ve also noted a marked decrease in the last 6-8 months. She’s now 3.25 years old and can deal with stress situations much better, of course i’ve been working with her using techniques like “Look at me”. Now, when we come across situations of high stress, we seldom need this, I watch her body language and will start this if she’s tense or fixated on something – and she lightens up right away. I suppose such general calming also occur with progressing maturity. OTOH, the reports I’ve received from breeder and guardians NOW is, that we’re miles ahead of her siblings. One gentleman who hires me for computer consulting has her sister and he’s had to stop bringing her to work because she can’t settle and goes bonkers when people enter the office. Mine will respond to attention but will also stay laying at my feet if no one pays her attention. I’m sure, a combination of maturing, but also relaxing protocols and reinforcing calm behavior.
Thanks for your help on this!
Buzz Cecil and Vela
(proud papa alert)
James Ha says
It sounds like you are entirely on the right track in dealing with this: sounds like anxiety and sounds like you are doing a great job with it!
Catrina Ross says
Very interesting article. I have a question about ” predatory drift ” referring to behaviors not being read correctly depending on breed. Is there a breed matching or non matching protocoll then that would lower the risk of this behavior to occur vs just going by size (50%) and hope for the best?
If we see it often in ex herding breeds such as German Shepherds, belgian hurding breeds, knowing that these breeds may be of a more recent descent, what should we take from that? Wouldn’t I assume that these newer breeds have less social communication skills and this “drift” should appear less often than in more ancient breeds such as Basenjis, Shiba Inuh,and mastiff breeds? Where am I going wrong in my reasoning? Matching two newer breeds missing or have less of the “wolf like social hierarcy skills” should have lesser chance of displaying this behavior and so should two ancient breeds playing together? But, we have seen “predatory drift” in two Basenjis playing together, living in the same household. Was that something else? I would greatly appreciate more information on this tough topic.
James Ha says
Well, I am rejecting the idea of predatory drift, so I don’t think that you have seen predatory drift in two Basenjis playing together… I do think that you may be seeing a situation in which they may be doing what appears to be playing together, which turns suddenly to aggression (which may also be play, or social status establishment, etc and not predation). As I mention, there are a number of explanations for this type of behavior expression (play turning to aggression or predation), lack of proper communication skills (due to genetics or early experience) being just one that I discussed.
You are correct: communication should be less of an issue between more closely related breeds (ancient or modern European or any branch in between), but communication may not be the issue. In herding breeds, this sort of behavior is often due to the disruption of inhibition in the herding behavior chain… we have “used” the predatory hunting behavior sequence to develop herding behavior, but the dogs are supposed to inhibit the last phase of the sequence: the kill, and if something disrupts that inhibition network in the brain (and there are several mechanisms for this), then the behavior of hunting (=herding) can suddenly go forward to its final steps of attack and kill.
Thanks for your feedback and questions: keep them coming!
Jim Ha, PhD, CAAB
ren clark says
Seems to me that predatory behaviors are not affective spontaneous fear or defensive or rage
everyents…a predator is not mad per se or afraid…this would support the position that predatory drift is a incomplete description.
Erik Grendahl says
First off, great article. I really enjoyed it. However, I would have loved to see your sources. Not that I think you are wrong, simply that I am left to take your word for it and I would rather see a study.
Simon Gadbois says
Great comments Jim. I would just add that instead of dismissing completely the concept of “fixed” behaviours (FAP’s) and not-so-fixed behaviours (MAP’s) and even less not-so-fixed behaviours (Natural Action Sequences [Fentress, Berridge, Golani, McLeod, Moran, Dawkins, etc.]), we should think about it in terms of a continuum between behaviours being quite stereotyped, and hard to change/modify, to behaviours being very much “controlled” with many degrees of freedom. After all, the concept of FAP is still valid in neuroethology when looking at the behaviour of “lower” vertebrates or invertebrates (e.g., the work of Ewer on prey capture in toads). So I am not ready to dismiss the concept. I am ready though to say that it is not all about FAP’s and nothing else, or FAP’s and learned behaviours. The concept of “action sequences” (more “modern” than MAP even; see Fentress & Gadbois, 2001) is an attempt at recognizing the whole array of “degrees of freedom” in the expression of behaviours, especially long sequences that may vary in levels of stereotypy. But I COMPLETELY agree with you that herding sequences in border collies are NOT FAP’s or MAP’s, but action sequences with some stereotypy, and a healthy dose of degrees-of-freedom.
Canid Behaviour Research Team, Dalhousie University, Canada
I’d be very interested in more ways to stay up to date on the science.
James Ha says
Thanks for your wonderful comments! I agree: there is a shortage of scientists willing to translate their work for the public that funds them (in many cases) and that can use their findings in a practical way. It has been my goal to try to fill a little of that gap with my blogs, but time always gets in the way. I have also been successful in developing a number of nice presentations which I have presented a few times, most notably as a day-long mini-course in Brazil.
I am glad you find my writing helpful, and I hope to produce much more of it in the near future!
Elaine Hurford says
Wow – amazing article, questions and answers. I’m so glad I stumbled across this site. I was looking for information on the effects in social interaction on dogs who have stayed with their mothers and siblings too long. I know now from reading above that the term “predatory drift” is not the correct one, but in that kind of context, I have a problem which perhaps you can help me with?
I have a year-old mixed-breed dog adopted from a shelter at 5,5 months. She was with her mom and five siblings until 3,5 months, and then with the remaining two other siblings until 5,5 months. Grace was the last dog to be adopted. If she’s any breed at all, she is a canis Africanis (see Johan Gallant’s in depth work on this breed which has evolved from Xhosa hunting dogs in the Eastern Cape, in South Africa. I imagined that her “stable” background (with mom and siblings) and the positive experience during all this time of an overnight foster home with lots of other dogs and animals would have an extremely positive effect. She”s also been very well socialised since I got her – I have made a point of introducing her to a wide variety of dog playmates – and she immediately established a special friendship with a 6,5-year-old female German Shepherd. The friendship involved ecstatic daily greetings, during which the GS would tumble Grace several times in the greeting, which I regarded as being all in the line of play. After many months of this, things changed suddenly and the GS started biting Grace on the muzzle and neck and I could see Grace was not coping with this so I ended the encounter and took her home. At about the same time a small feisty rescue dog Lucky entered the picture and they both started “packing” on Grace. Alone with Lucky, Grace is fine. But with just the GS, or with both the GS and Lucky, there is a big problem. A friend who is a canine behaviourist has fortunately just been here to give a talk and consultations, and we had plenty of time – as she was staying with me – to observe her two Border Collies with Grace, as well as the GS and Grace together. The older male border collie right at the outset showed dominance behaviour with Grace but within a day or so, with the right checks and assistance from my friend Claire, Grace had learnt to lift her head and walk away instead of doing her usual shimmying, wiggling, submissive thing. Good. But back to the GS…..after watching the GS with Grace, Claire concluded that the GS’s behaviour could easily tip over into real aggression and said that she needed to be taken back to the basics of obedience training (she had been a star pupil in her day according to her owner) and that she and Grace could continue to see each other but only on leashes and with a respectful distance between each other. Yesterday, the GS arrived at the play field on leash – Grace was already off-leash playing with her buddies – and the owner kept her on leash until Grace had had a fair chance, and then we reversed it so that the GS could go off and play with her friends too. Instead of which, she turned round after having ignored Grace totally, all the time (they were just a metre from each other, sitting quietly) and went for her from a cold start, biting her muzzle and her neck. Grace never seems to lose her affection for the GS and goes through an obsequious greeting each and every time which involves much licking of the GS lips and muzzle. When the GS “bites” Grace of course flattens herself and ducks and runs if she must.
I must add here that Grace has not been to obedience classes, although she is very responsive to the bit of training I have done with her (recalling, no jumping etc) and mostly regards every other dog as her best friend – she rushes over all joy and exuberance to every dog she meets and I do realise that this is not a good thing – as Claire (my dog behaviourist friend) has pointed out, Grace has not learnt the normal and appropriate social signals, including the careful butt-sniffing ritual – in other words, she doesn’t speak Dog – and that this is entirely due to her over-long stay in the mother-and-sibling environment. All this makes sense to me, and I’m noticing now that there are other dogs – she was exposed to a bull-terrier yesterday – who just want to pile in and go for her. Both the bull terrier and Grace were on leashes but the signals from him were unmistakable and it was obvious that Grace was terrified. (This was a playdate gone wrong because the bull terrier was not supposed to have been there at the time so when I saw him I leashed Grace, which was just as well. We were on the playing fields where all the dogs are usually unleashed and where for Grace, the Play! Have fun! associations are very strong).
What to do? I don’t want her going through life having dogs pick up her social defencelessness (for lack of a better term) and beating up on her. It means, in practice, no off-leash beach runs where she may meet strange dogs, and constant vigilance on my part.
How can she learn what she needs to learn?
I’m sorry this has been such a long story !
I’m also interested to read that you say dogs are directly descended from grey wolves; I thought there’d been a genetic branch off many centuries ago, into canis familiaris. Is this a moot point with ethologists perhaps.
Anyway I loved this blog and will be following you closely – thank you!