What We Need to Learn About Missing Dogs
The Missing Animal Response Network is an international network of lost-pet recovery specialists (volunteers and professionals) who are interested in learning as much as we can about the behavioral patterns of lost pets. The foundation for our missing animal response (MAR) training program is based on the existing techniques used in search and rescue or SAR, and we’ve been very successful in recovering missing pets by using SAR techniques like search probability theory, area search dogs, scent trailing dogs, and forensics.
Back in 1997 when I first started to experiment by using my search dog to conduct physical searches for lost dogs and cats, I was a volunteer search manager for a local search-and-rescue team. I made the realization that since studies have been published related to lost person behavior and that data is used to help recover missing persons, perhaps we could do the same for missing pets. It took 20 years, but in 2017 I was able to see my dream come true when we collaborated with the University of Queensland to conduct the Missing Cat Study. The study was published as “Search Methods Used to Locate Missing Cats and Locations Where Missing Cats Are Found” in the journal Animals in 2018; you can read the full text here.
Read more about the study and my work on locating missing cats in this article from the IAABC Journal.
This type of collaboration is invaluable because it pairs research-minded academics with individuals who have boots-on-the-ground experience with the subjects being studied. Many questions for the Missing Cat Study were developed based on the knowledge of those with years of experience searching for and recovering missing cats.
The data from the Missing Cat Study confirmed and legitimized lost cat behavioral patterns that I had observed many years ago, including the fact that displaced (escaped) indoor-only cats do not travel far from their escape point. It touched on some of the differences between searching for and recovering an indoor-only cat versus an outdoor-access cat. It was a great step forward in understanding lost cat behavior, but much more research and further analysis of lost cat behavior is still needed.
We are currently in the early stages of developing a Missing Dog Study in a partnership between the Missing Animal Response Network and another non-academic lost dog recovery organization that has the ability to both fund and widely promote the study during the data collection phase. However, before we launch the study, we are seeking an academic partner to help us craft the study questions. The academic partner is also needed to analyze the data and then help publish the study in the scientific literature.
We hope to learn many things from this study, including determining if there are any specific patterns in the distances that dogs travel based on factors such as the size, breed, or temperament of the dog. We hope to determine the most effective methods that owners used to recover their missing dog. We plan to look at the various ways that dogs become lost, including what percentage of recently adopted dogs end up escaping from their new home. We will look at patterns of distances traveled based on motivational factors such as the difference between a dog who meanders out an open gate to explore versus a dog who bolts to chase after a deer or runs in fear from fireworks. We believe that this study will help lost pet recovery experts to develop better search plans when they are overseeing a lost dog operation. It should also help pet rescue groups and animal shelter staff better coach dog owners about how and where to search for their lost dogs.
My preliminary studies and the need for more detail
There is some data on lost dogs—both published and some that I have collected myself—but it is not really enough to help me create evidence-based tools to help recover lost dogs. I conducted an informal, non-scientific study back in 1999, in which I interviewed 254 dog owners who had lost and then found their dogs. I asked a series of questions and categorized the dogs based on breed, and then looked at a specific distance of a quarter-mile radius and determined what percentage of the dogs were recovered within that radius.
My data suggested that:
- Mixed-breed dogs were allowed to travel an average of 14 miles before they were picked up, whereas purebred dogs were allowed to travel only an average of 2 miles before they were picked up.
- Toy breeds were the least likely to travel more than three-quarters of a mile from where they were last seen.
- Breed and size seemed to have an effect on the distance traveled—pointing breeds were more likely to travel farther than retrieving breeds.
This initial study was helpful to me as I was developing educational materials for pet owners and training missing dog locators, but the numbers are 20 years old now, and the landscape is very different.
Other existing research
This paper from Lord et. al (2007) was a great first step toward collecting data related to lost dogs. It determined the importance of identification tags, licensing, and microchips, as well as the importance of having a “search plan” if a dog becomes lost. But no new information was discovered in the study that could help a dog owner develop an actual search plan.
The data that was reported in the 2007 study was actually collected in 2005. Back in 2005, Facebook was only a year old, and there were very few lost pet recovery services, businesses, apps, or lost pet Facebook pages compared to today. Some questions that we believe need to be answered today are:
- Just how successful are lost-and-found Facebook pages and neighborhood social networks such as Nextdoor in recovering lost dogs? Should they be an integral part of a search plan?
- Are dogs that are perceived as more dangerous by the general public found farther from their homes than other breeds due to recuers’ reluctance to interact with them?
In addition, the study published in 2007 included dogs that were never recovered. The study determined that of the 187 dogs that were lost, 132 (71%) of them were recovered. But when I read the study, my first question was this: What happened to the other 29% of lost dogs, and why were they never found? They weren’t abducted by aliens—so where did they go? Could we, through more knowledge about lost dog behaviors and the most effective lost dog recovery techniques, have increased the recovery rate of lost dogs?
I believe that once we publish a Missing Dog Study, we can educate dog owners and animal welfare groups within the pet industry, train more community-based lost pet recovery teams, and improve the effectiveness of lost-and-found Facebook pages and groups so that the recovery rate will be much higher than 71%.
The research questions we propose to look at in the Missing Dog Study
Why do dogs get lost?
We want to determine the primary reasons dogs become lost. For example, what percentage of dogs become lost because they:
- Wandered out an open gate or door
- Escaped by some means, such as digging under or climbing over a fence
- Fled in a blind panic from home or out on a walk
This information would be used to help with lost dog prevention. For example, if we discover that 85% of 1,200 dogs in the study became lost because they were newly adopted dogs that escaped from their yard, then shelters and rescues can better educate new adopters in how to escape-proof their yard.
In many cases, the motivational factors may not be known, but we hope to at least look at a subset of data where the owner knows why the dog got lost. For example, did the dog run away after prey or bolt in fear due to fireworks? The reality is that most dog owners discover their dog missing from their yard and might never know why their dog escaped. However, dogs that have shown a history of prior escape attempts from their yard or have been known to bolt and chase after rabbits or deer would likely be reported based on anecdotal history. Whether or not the motivational factor can actually be studied is something the team working to develop the study will need to determine before we include it.
We plan to look at what percentage of dogs recently adopted end up escaping while in transport (i.e., by a rescue group, foster family, transport group, etc.) because we have seen a dramatic increase in the number of escapes that come from these groups. Many undersocialized and skittish dogs that previously would not have been adopted from shelters are now being transported to be rehomed elsewhere in the country, and we suspect that some of these dogs are escaping due to various factors including carelessness and unpreparedness. If the study supports this hypothesis, resources can be put into developing educational materials for these groups on how to prevent their dogs from escaping.
Where do they go?
We want to explore whether certain characteristics can be used to determine how far a lost dog might travel. For example, do certain breeds travel typical distances? If we discover that 80% of toy breeds are found within a few blocks of their escape point (which is what our 1999 study showed) then we can better coach owners of toy-sized dogs to focus their initial search for their missing dog within their own neighborhood. We suspect that certain breeds/types (pit bulls, Rottweilers, Dobermans) that are perceived by the general public as “dangerous” are likely avoided by many would-be rescuers and thus allowed to travel much farther than dogs perceived to be non-threatening (i.e., Yorkshire terriers, Chihuahuas, etc.), even though many of the toy breeds may be just as likely to bite.
We have not yet explored the various questions that we will ask, but we will certainly be looking to examine what types of areas that dogs, especially skittish, hard-to-catch dogs, choose as their new “residence.” We’ve seen that many dogs with fearful temperaments that escape near isolated areas (wooded areas, golf courses, cemeteries, large parks, etc.) prefer to remain in those quiet areas, but we do not know what percentage. More than likely our questions about geography will look at terrain features that influenced the dogs travel patterns. For example, a canyon, a brick wall along a freeway, and a raging river are all geographical barriers that would inhibit and influence a dog’s direction of flight. Another geographic factor would be the impact of weather on travel distances. What kind of effect would the weather in a desert in July have on the distance a lost dog can travel (and survive) compared with the weather in a lush agricultural area in early March?
A critical factor related to geographical features is the population density where a dog is lost. A dog that slips their collar while being walked in midday in Manhattan (where there might be thousands of people within a one-block radius) is going to travel much less of a distance compared to a dog that slips their collar at a rest stop in an isolated area in the middle of Nevada. For the dog in Manhattan, it would be likely someone had picked up the dog quickly, so I’d post the lost dog using the app “Find Shadow” (they have a very large volunteer base in New York) and focus my search in that immediate area. For the dog at the rest stop I would create giant, neon posters for all major exits from that highway within a 30-mile radius.
What factors affect what happens to them?
We also want to determine if the temperament of the dog and the circumstances of their disappearance influence the distance traveled when lost. We suspect that wiggly, friendly dogs are typically recovered close to home, but skittish, fearful, undersocialized dogs might travel much farther. Likewise, a dog that wanders out an open gate might not travel as far as a dog that runs in fear from fireworks. This data will help lost pet recovery specialists and animal shelter staff to better coach dog owners about how and where to search for their lost dogs.
We will likely look at the differences between purebred and mixed-breed missing dogs, but I am not certain what the study will show. Probably the only way to study that issue would be to interview people who’ve found a dog and asked questions about what they were thinking and what they did with the dog (i.e., kept it, took it to a shelter, posted it on Craigslist, posted it on a lost-and-found Facebook page, etc.). We are also considering asking people what led them to the conclusion that a rescued dog was lost and being searched for, rather than dumped or abandoned by owners who no longer wanted them.
What recovery methods are most effective?
We hope to determine the most effective methods owners used to recover their missing dog. If we find a much greater return on investment for signs than we see for robocall phone alert services, we can guide the efforts of dog owners accordingly.
Cash rewards are another aspect that we plan to examine within the Missing Dog Study. Years ago, I used to suggest large rewards to grab the attention of the public in order to help facilitate a quick recovery. However, with the dramatic increase in public interest of lost dog recoveries through social media posts (facilitated by the growth of lost-and-found Facebook groups), an alarming problem has developed. Many well-meaning but untrained rescuers are inadvertently chasing skittish, loose dogs into traffic, with tragic consequences. Many volunteers with lost dog Facebook groups report that, in cases where a dog owner has offered a substantial reward, the lost dog was chased when sighted, supposedly because the chasers wanted the reward money. We hope to develop questions for the Missing Dog Study that will reveal whether this is true or not.
We’ve also learned new information related to the contrast between offering monetary incentives versus altruism that have us hesitating to recommend monetary rewards. According to the book Sway: The Irresistible Pull of Irrational Behavior, the financial compensation part of our brain simply cannot operate at the same time as the altruistic part of the brain—only one or the other kicks in when we are thinking. The nucleus accumbens is a primitive part of the brain that is known as the “pleasure center,” and it is where we respond to financial compensation. On the other hand, the posterior superior temporal sulcus is the area that responds to social interactions and is known as the “altruism center.” The authors mention studies that indicate that “we can approach a task either altruistically or from a self-interested perspective” and that these two different areas of the brain “run on different fuels and need different amounts of those fuels” to work. Since most communities have social networks with lost and found pages and the majority of people who monitor or join these pages do so because they love dogs and want to help reunite these dogs with their families, it makes more sense to appeal to the altruism center and advise dog owners to use the words “PLEASE HELP FIND” on their posters rather than “REWARD OFFERED.” I have more information about this on a blog post titled “To Reward or Not to Reward: That Is The Question” on my Missing Animal Response website.
The practical benefits our study could bring
Having data from a Missing Dog Study would enable pet detectives, lost-and-found Facebook administrators, veterinarians, shelter staff, volunteer rescuers, and others in the pet industry to quickly develop a search plan that goes beyond “post flyers in the area, put an ad on Craigslist, and check the shelters.” Most dog owners are already doing those things, but they want better, more sound advice. This study would provide that!
For example, if we knew the likely distances that different dogs would travel, this would help determine the area to focus the search. And if we knew how likely it was that the dog was picked up versus still running loose, it would help determine what type of advertising and search strategies would work best.
If lost pet recovery specialists, shelter staff, and dog owners knew that their German shepherd/Labrador retriever mix that escaped might be ignored or go unnoticed by many passersby, they might know to put more emphasis on marketing techniques that would make their dog’s disappearance more noticeable and memorable to the public. For example, using giant neon LOST DOG posters or “tagging” their cars with neon window markers. Remember, the preliminary (non-scientific) study in 1999 showed that mixed-breed dogs traveled seven times farther than purebred dogs traveled. I suspected this could be because many people, especially non-dog lovers, would only stop to pick up a loose dog if they perceived that the dog was valuable. This, of course, means that a study needs to be conducted about the thought processes of rescuers (both dog lovers and non-dog lovers) who find loose dogs. What those rescuers think (and believe) influences how they behave when they find a dog, and ultimately determines the probability that the dog will be returned to their guardian.
Being involved in academic research as part of the Missing Cat Study was amazing and I learned so much. I loved being part of a group of professionals with different backgrounds and talents as we collaborated on an important study that was long overdue. The professional researchers needed the input and knowledge that the pet recovery specialists brought to the group just as much as the pet recovery specialists needed the professional researchers who helped build and refine the entire study. It was also exciting to see that the study was not only published, but that it was featured on the cover of the MDPI’s academic journal Animals. We are quite hopeful that the Missing Dog Study will be just as, if not more, influential.
Albrecht, K. (2016) The science of finding lost pets. The IAABC Journal.
Huang, L. et. al (2018) Study search methods used to locate missing cats and locations where missing cats are found. Animals 8(5)
Lord, L.K. et al. (2007) Search and identification methods that owners use to find a lost dog. Journal of the American Veterinary Medical Association 230(2), pp. 211-216.