The Dog Project: Researching Canine Behavioral Genetics

An investigation into the genetics of canine anxiety, phobias and fears
By Karen B. London PhD, November 2008, Updated June 2021

“This is the job that Solo got me,” says Melanie Lee Chang, PhD, a biologist who got her doctorate in evolutionary biology and physical anthropology at the University of Pennsylvania and is currently working in canine molecular genetics. Solo is her eight-year-old Border Collie. The job is as a postdoctoral scholar in the Department of Psychiatry at the University of California, San Francisco (UCSF) conducting research on the Canine Behavioral Genetics Project—commonly referred to as the Dog Project. The project’s primary purpose is to explore the relationship between genes and behavior, both normal and abnormal, in domestic dogs. The secondary purpose is to assess the amount and nature of genetic diversity in domestic dogs, both within and between breeds.

The Dog Project is a collaborative effort. Steve Hamilton, MD, PhD, is a psychiatrist and geneticist who has long been interested in the question of how genetics influences susceptibility to mental disorders in humans. Certified Veterinary Behaviorist Karen Overall, PhD, DACVB, a research associate in the Department of Psychiatry and Center for Neurobiology & Behavior at the University of Pennsylvania, emailed him about a paper he wrote on anxiety in humans. They soon realized they had a number of parallel interests. Dogs and humans have irrational fears and phobias that are similar biologically, in terms of treatment, and in their clinical manifestations. Dogs and children can both suffer from separation anxiety, and dogs exhibit obsessive-compulsive behavior similar to that seen in humans.

Hamilton hopes to be able to apply what is learned in this study to humans, but long before the research has an impact on people, the dog community will feel its effects. Among them, they could include influencing dog breeding and changing how dogs with behavioral issues are evaluated. A further understanding of the genetics of anxiety could even lead to insights about the very domestication of dogs.

A Hot Topic
Bringing up the potential for changes in breeding practices based on behavior and genetics is an exciting way to get a heated discussion going among researchers, behaviorists and breeders. There is no consensus. Certified Applied Animal Behaviorist Steve Zawistowski, who has a PhD in behavior genetics and works at the ASPCA as the head of national programs and science advisor, asks rhetorically, “If you are breeding Springers, wouldn’t it be nice to know if they are prone to Springer rage and breed accordingly?” As he points out, responsible breeders make decisions based on physical attributes; for example, they will only breed dogs with sound hips, passing over those individuals with hip dysplasia. So why should behavioral soundness be immune from these kinds of breeding decisions? His comments are in line with the beliefs of many behaviorists—in favor of breeding dogs with the goal of producing individuals that are least likely to develop the serious behavioral problems that result in suffering for both dogs and people.


Sign up and get the answers to your questions.

Email Address:

In contrast, Hamilton believes that while “using such information may be crucial in cases where behavior disorders appear to cause intense suffering in dogs within a pedigree or breed, such information can be used prophylactically to signal which dogs warrant early intervention and treatment.” And Chang is concerned about the possibility of unwarranted oversimplification of the complexity of the links between genetics and behavior leading to inadvertent selection against behaviors that are desirable in certain breeds.

The Canine Behavioral Genetics Project does not breed dogs with behavioral issues. Rather, these researchers are analyzing the genes of existing dogs, and are recruiting dogs to participate ( Specifically, the researchers are looking for dogs who suffer from panic, fear, anxiety, obsessive-compulsive behaviors or aggression; unaffected close relatives of dogs with these issues; and dogs who do not have any of these behavioral issues or any known relatives who have them, either.

Participation consists of sending DNA acquired from a cheek swab of the dog and filling out a 25-page questionnaire about the dog’s behavior and any available pedigree information. Rather than ask, for example, if dogs bite people, the questionnaire is highly detailed, asking whether the dogs bite, whether the bites have broken the skin, how many bites there have been, and whether the bites are from the back, the front, or occur from any direction. The checklist about the dogs’ behavior after biting includes whether or not the dogs’ pupils are dilated, the dogs appear disoriented, dazed, tremulous, tense, or uneasy, whether they hoard objects, whether they are quick to recover, are light sensitive, their eyes appear glazed, whether they stare or are frozen in one position, whether they always hide in the same location, appear unaware of their own action, whether they react to benign motions such as raising an arm while reading the newspaper, and if there are any other post-bite reactions. The questionnaire is similarly detailed for separation anxiety, noise phobias and a host of aggressive contexts.

Canine Genes Rich in Information
When considering organisms destined for genetic study, fruit flies and peas initially spring to mind, but the domestic dog belongs at the top of the list. The dog has accidentally become the greatest genetic experiment in human history, and is thus the perfect system in which to address many genetic questions. Dogs display enormous variation in both morphology and behavior, both of which have been selected for intensively over many generations, yielding a one-of-a-kind opportunity to study morphological and behavioral variance. Researchers may be more likely to learn about the influence of genetics on components of behavior by studying dogs than by studying humans.

There are two general approaches to finding genes correlating to specific behavioral tendencies, and the Canine Behavioral Genetics Project uses both. The first approach entails looking at purebred dogs, which gives the researchers access to populations of dogs with limited genetic variation because they are all descended from a limited number of individuals. This low genetic variation makes it likely that at least most dogs within a breed with a particular trait are more likely than dogs from different breeds to share the same genetic variation. This fact significantly simplifies the search for genes related to the trait of interest, and reveals why the domestic dog is such a valuable research tool for geneticists. Families of purebred dogs are even more genetically similar. Therefore, if the researchers can find close relatives in which some dogs are affected by a behavioral disorder and other individuals are not affected, it is even easier to pinpoint where the dogs’ genetics differ.

The second approach is to study both behaviorally affected and unaffected dogs that are unrelated either within breeds or across breeds. Because unrelated dogs are not as similar genetically, the areas in which their DNA is similar is particularly interesting if correlations can be found with abnormal behavior. If a group of unrelated dogs with the same behavioral problem all have a particular variation in their genes, then that variation is worth investigating as a possible cause of the disorder.

Solo’s Lucky Day
Solo’s behavioral issues made him available, but Chang got Solo by luck. Early on, she thought it was bad luck. He was a hodgepodge of behavioral issues: fear, nervousness around strangers, noise phobia, separation anxiety, emitting distress vocalizations if Chang fell asleep, and on and on. For weeks, she could only stop him from pacing and howling at night by singing. Every time she fell asleep, he would start crying again and wake her up. Her feelings have changed so much that Chang now speaks about Solo the way couples in love talk about one another, but they certainly had a rocky start.

She had just lost a magical dog—a Pomeranian named Harley whom Chang could take anywhere because she was always good. Chang took her to classes she attended, to classes she taught, to her office, and with her when she went out. She sums up her grief by saying, “When you lose a dog, there’s a huge empty space, and in my case, it followed me around since I took her everywhere.”

After she lost Harley, Chang got connected to an independent rescue group, but something did not seem right to her at the arranged meeting with Solo. The man who brought Solo didn’t ask her much about herself, he didn’t check any references, and he seemed in a huge hurry to hand Solo off. Even though the whole situation seemed wrong, and she felt sure this was not how rescue was supposed to work, she obeyed the voice in her head saying, “Just let this happen.”

Chang realized immediately that Solo was nothing like Harley. She brought Solo to school, partly because it was apparent after just one night with him that he could not tolerate being alone. She left Solo in her office in the company of two female classmates and went to class. Five minutes later, her advisor, Art Dunham, came to get her because Solo was screaming his head off. Not only could her advisor recognize the seriousness of the problem, but he was fortuitously married to veterinary behaviorist Karen Overall. Dunham told her, “Karen can see you Saturday.”

Overall diagnosed Solo with generalized anxiety as well as separation anxiety. He was alternately agitated/panicky, pacing, scanning, panting, drooling or catatonic. When he was catatonic, he would lie on the floor with his eyes open, drooling, and he didn’t blink even if a hand was waved in front of his face. When Chang left him alone (even to take out the garbage) he would fly into a blind panic and cry at full volume. Solo immediately began to take anti-anxiety medication, and Chang began behavior modification that incorporated deference, relaxation, and desensitization to arrivals and departures. A combination of medication (Solo is taking Elavil and Prozac and used to take Alprazolam, drugs also used to treat human mood and anxiety disorders) and lots of practice has resulted in an extraordinary change in Solo’s behavior and quality of life.

Solo lives a full and basically normal life—he’s taken part in rally obedience, agility, stock work, and has appeared in instructional videos. While he does not want to be petted by strangers on the street and hates July Fourth, in most circumstances his behavior is completely unremarkable, which Chang considers his greatest accomplishment. Chang gratefully says, “I wouldn’t still have my dog if it weren’t for Karen Overall.” Years after this appointment, when Overall and Hamilton were looking for a biologist who was also knowledgeable about dogs to work on the Canine Behavioral Genetics Project, Chang was a natural choice.

Chang has gotten used to the idea that Solo will be on medication for the rest of his life. “If giving him drugs makes his world look safer and more normal to him, then I think it would be cruel not to give them to him.” Solo’s environment is also hugely influential on his behavior. Like many people whose dogs have behavioral issues, Chang is hyperaware of her environment and protects Solo as much as possible from things that upset him. Sheepherding has been a godsend for both Solo and Chang, who says, “I started working him on sheep because he was weird, and it was the only place where I got to pretend I had a normal dog. The more he sees sheep, the better he is.”

Community-Based Study
Solo is one of many participants in the Dog Project. The dog community has really stepped up to take part. People’s eagerness to participate no doubt stems in part from the study’s lack of harm to any animals. As Certified Applied Animal Behaviorist Patricia McConnell, PhD, comments, “I’m gratified to hear that the researchers are avoiding breeding for conditions that could cause suffering in the dogs that have them and are using existing dogs to help us understand both dogs and humans who suffer from genetically-mediated behavior problems.” Chang adds that, “One of the major advantages of a community-based study is that the dogs are living in their normal environmental context, the one that they must be able to cope with, so we are looking at in-context behavior rather than behavior in an artificial lab environment.”

The helpfulness of the dog community has led to DNA samples from more than 3,000 dogs, and matching questionnaires for most of those. Almost 300 dogs are in the study of noise phobia involving Border Collies. Noise phobia is an example of a behavioral issue that occurs more commonly in certain breeds, including the Border Collie. McConnell notes, “It’s not always a problem in that surely, sound sensitivity is why a quiet whistle can get a Border Collie to lie down when he’s 500 yards away. However, the flip side of that can result in a dog who is thunder phobic or terrified on July Fourth.” McConnell’s experience with Border Collies and noise phobia goes beyond her professional life into her personal life. Several of her Border Collies have been uncomfortable during storms, and one, named Pip, was truly thunderstorm phobic. Pip was successfully treated/managed with a combination of counterconditioning, supplements, and body wraps. There’s little question in McConnell’s mind that she came with a genetic predisposition to be especially responsive to loud noises. Several other dogs in her lineage seemed to be particularly sensitive to loud noises.

The scientists working on the Dog Project are also interested in anxiety issues, panic, compulsive behaviors and aggression. They are especially interested in extreme forms of aggression in which dogs respond in a particularly reactive way, and PhD student Jennifer Yokohama is building a project focusing on these behavioral problems. According to Hamilton, “We’re looking for a disproportionate, out-of-the-blue aggressive response to a stimulus to which normal dogs even within the same breed do not react.” They are also looking for more typical, milder forms of aggression that may be a result of some of the same factors causing other, anxiety-based problems.

Not a Quick Fix
Researchers tend to be sensitive to the potential misuses of their research. Chang is uneasy that an unwanted effect might be people being more likely to want medication for their dogs as a quick fix. She emphasizes that medication is not a quick fix. It’s just a way to get some dogs receptive to the behavior modification that is needed. With research into the genetic influences of behavioral disorders, Zawistowski says, “There’s the possibility that someone will use the data to say that this gene is associated with aggression and dogs with this allele [gene form] should be euthanized because they pose a risk. It’s similar to breed-specific legislation because Pit Bulls pose a greater risk than Shih-Tzus. There’s always a risk for that type of issue.”

Chang hopes that when the results are presented, they will not be subject to that kind of misinterpretation. As far as she knows, there is no behavioral problem that is breed specific, and all breeds can exhibit behavioral problems. Hamilton comments that “The most extreme [misuse of the data] would be to overinterpret influence to ‘cause’ and use that information to make breeding decisions based on that or to make sweeping legislation about breeds or groups of breeds.” The researchers all adamantly oppose breed-specific legislation, and Karen Overall in particular has publicly voiced her opposition to it for years.

Despite people’s innate fascination with what drives their dogs’ behavior, until recently the answers to the questions posed by this project were not within reach. Zawistowski says, “I wanted to do behavior genetics with dogs for my PhD, but the systems were not in place for good, solid research. The dog genome work has changed that. At the same time, we’ve seen so much more interest in the behavior of domestic animals—long ignored by animal behaviorists. So this new work is really the culmination of two developments: better genetic methods, and more interest in dogs.”

There remains considerable resistance to the idea that behavioral problems may be genetically mediated. Zawistowski comments that people are uncomfortable with the idea of genetic predispositions to behavior because of the erroneous idea that genes are destiny. Hamilton points out that “Knowledge of the genetics helps shape the destiny.” A major hope that Chang has for the project is that it will have a positive impact on the way people feel about canine behavioral problems. “There’s often a refusal to believe that this is not the owner’s fault. Owners feel so much guilt. I’d like to change that.” Decades ago, psychiatry struggled with these same issues in the “If it’s not one thing, it’s your mother” view of mental illness, but now affected people and their parents are not generally blamed for anxiety or depression. The field of canine behavior is catching up to this more enlightened viewpoint as evidence mounts in its favor.

A Happy Ending
In the two months before Chang got him, Solo lived in five homes. After a few weeks with his littermates, he spent most of his first year in near-total isolation, or as Chang describes it, he lived like veal for a year. For a dog with any sort of predisposition to social anxiety or fear, this environment was far from ideal. Luckily for Solo, Chang listened to the voice in her head that told her to take him. While she may have originally regretted this decision, there came a day when her heart fully opened and she began her fierce commitment to keep Solo no matter what.

Now, in addition to describing Solo as her soul mate, Chang has many other things to say about Solo, including, “He’s the canine John Nash … This is a dog who thinks too much … He got me involved in a whole bunch of things. Solo forced me to learn a lot. If you’re willing to put in the effort and work with a dog with behavioral problems, you’ll learn more from him than from a hundred normal dogs … Having Solo has made me more understanding about other dogs … He’s got a very complicated personality. If he could speak, it would be in complete sentences and he’d use big words … Everyone had let this dog down, and I did not want to be the last person to give up on him.”

Perhaps the most insightful comment Chang has about Solo is one that can be applied to so many relationships where love takes over and everything else is just a detail. In a voice rich with emotion, she says, “Solo doesn’t have the greatest temperament, but he has the best personality!”

Article first appeared in The Bark, Issue 45: Nov/Dec 2007

Image: iStock

Karen B. London, Ph.D. is a Certified Applied Animal Behaviorist and Certified Professional Dog Trainer who specializes in working with dogs with serious behavioral issues, including aggression. Karen writes the animal column for the Arizona Daily Sun and is an Adjunct Professor in the Department of Biological Sciences at Northern Arizona University. She is the author of six books about canine training and behavior, including her most recent, Treat Everyone Like a Dog: How a Dog Trainer’s World View Can Improve Your Life

We Recommend