Features October 2011 Issue

How to React When Your Dog Begins Resource Guarding Against Other Dogs

Improve the behavior – and emotional response – of dogs who guard resources from other dogs.

Resource guarding is a natural, normal canine behavior. In fact it’s a natural behavior for most warm-blooded animals. Even we humans guard our resources – sometimes quite fiercely. Think about it. We lock our doors. Store clerks have loaded .22 rifles under checkout counters, while homeowners keep shotguns and baseball bats leaning in the corner by the back door. Companies hire security guards. Banks keep valuables in vaults. Some of us get insanely jealous if someone pays too much attention to our significant other. I could go on . . .

Peanut guards his food bowl from the other resident dog, Otto – even though Otto couldn’t care less about Peanut’s food. In order to prevent scenario # 3 described above, we’re working to counter-condition Peanut’s response to having Otto near his food.

Dogs guard their resources as well, sometimes quite fiercely. This is most troublesome when they guard from humans, but can also get them in hot water when they guard from other dogs. That said, some dog-dog guarding behavior is quite appropriate and acceptable. The wise dog owner not only knows the difference between appropriate and inappropriate guarding, she also knows when and how to intervene, manage, and modify.

Guarding Scenarios
If dogs didn’t guard their resources from other dogs they’d be in danger of starving – both in the wild and in multi-dog households. It’s this survival instinct that triggers everything from the canine dirty look known as a “hard stare” to the ferocious blood-letting, sometimes fatal battles that can occur when dogs fight over valuable, mutually-coveted resources: food, toys, objects, locations, beds, and human attention.

There are several different scenarios that can occur when one dog chooses to guard something from another dog, ranging from a very healthy, normal interaction, to those that risk the very lives of one or more of the combatants:

1) Ideal: Dog A and Dog B are both appropriate – The ideal resource-guarding scenario probably plays itself out frequently in multi-dog households, dog parks, doggie daycares, and anyplace dogs randomly gather. It looks something like this: Dog A is chewing happily on a (insert any valuable resource here). Curious, Dog B approaches. Dog A gives Dog B “the look.” Dog B quickly defers, saying, “Oh, excuse me!” by calmly turning and walking away. No harm done. Much of the time the dogs’ owner isn’t even aware that this occurred.

2) Second Best: Dog B is inappropriate but Dog A defers – Dog A is chewing on (insert resource). Dog B approaches. Dog A gives Dog B “the look.” Dog B gives Dog A “the look” back. Dog A defers, “Oops, sorry!” by dropping the resource and leaving. Dog B was inappropriate, but Dog A didn’t want to argue about it. The danger here is that somewhere along the line Dog A may get fed up with Dog B’s inappropriate behavior and decide not to defer.

3) Now We’re in Trouble, Part I: Dog A is inappropriate – Dog A is chewing on (insert valuable resource). Dog B approaches. Dog B would defer if warned, but rather than giving “the look,” Dog A leaps into action and attacks without giving Dog B the opportunity to defer.

4) Now We’re in Trouble, Part II: Dog B is socially inept – Dog A is chewing on (insert valuable resource). Dog B approaches. Dog A gives “the look.” Dog B is oblivious, and keeps blundering forward, until Dog A feels compelled to escalate the intensity of his message, to aggression if necessary, to get his point across.

5) Now We’re in Trouble, Part III: Dog B is inappropriate and Dog A doesn’t defer – Dog A is chewing on (insert valuable resource). Dog B approaches. Dog A gives “the look.” Dog B gives Dog A “the look” back. Instead of deferring, Dog A takes offense and escalates his aggressive behavior to maintain possession of his resource. Dog B reciprocates with increased aggression, and a serious fight ensues.

The same five scenarios can apply to other guarding situations – the dog who doesn’t want to share his sticks or toys; who becomes tense if another dog approaches him on his bed; or who claims his owner’s total and undivided attention. So whether it’s food or some other valuable possession your dog is guarding, what do you do about it?

Beware, and Be Aware
First, you have to be aware of the guarding tension. It’s hard to miss in Scenarios 3, 4, and 5, but if you have dogs engaged in one of the first two you may have overlooked it. Time to sit up and take notice! With Scenario 1, where both dogs respond appropriately, all you need to do is keep an eye on things and breathe a sigh of relief. As long as the pattern repeats itself, you needn’t worry. You just need to stay calmly observant and take note if the pattern changes – if, for example, Dog B is slower to defer over time, which may cause an increase in Dog A’s tension and possibly escalate to higher-intensity guarding. Many dogs live happily together their entire lives politely signaling and deferring in relation to valuable resources. That’s how it’s supposed to work – perfectly appropriate and normal.

If you see subtle signs of increasing tension, however, or if you see Scenario 2 behavior, where Dog B is bullying Dog A into giving up the resource, you have potential trouble brewing. It’s possible that Dog B will calmly defer for the rest of the dogs’ lives together. You could continue to observe, and intervene only if things start to escalate. Maybe it never will. Or you could intervene with management and/or modification now, before you have significant relationship damage to repair, and a more difficult behavior modification challenge.

Of course, anything more dramatic than  Scenario 2 behavior requires immediate action in the form of management and, if you choose to do so, modification.

You Can Manage
I’m a huge fan of management. If your dogs’ list of guardable items is relatively short and the dog-dog guarding interactions are reasonably predictable, then management may be a realistic option. Feed meals to your dogs or give them pig ears only when they are safely crated or closed in separate rooms. If you have a toy guarder, do toy-play with the dogs separately, and put coveted toys away when the dogs are together. Case closed.

Modification is in order, however, if battles grow increasing likely to erupt unexpectedly over an ever-growing list of miniscule triggers, such as a crumb dropped on the floor, a preferred resting spot on the rug, the hallway to the kitchen, equitably delivered treats to both dogs, or a rapidly growing radius around a valued human. Of course you’ll manage in the meantime, but since management always carriers a risk of failure and guarding battles can be fierce, the more generalized the guarding, the more critically important it becomes to convince your dogs to act appropriately with other dogs in the presence of high-value resources.

Modify
Aggression is caused by cumulative stress that pushes a dog over his aggression threshold. We’re all grumpier when we’re stressed. (See “A New Threshold,” WDJ October 2010). Begin your modification program by minimizing as many other stressors as possible in your dogs’ world. That includes creating structure and predictability in their lives; exploring and treating any possible medical conditions that may cause pain or distress; and eliminating the use of any coercive or pain-causing training tools and methods (shock, choke and prong collars, physical or harsh verbal punishment).

At the same time, incorporate calmness-inducing products and procedures such as increased aerobic exercise, the “Through a Dog’s Ear” recordings, Thundershirts or Anxiety Wraps, calming massage, and TTouch.

Trainer Sarah Richardson feeds Otto (a.k.a. “Dog B”) a small, high-value treat. Peanut (Dog A, the guarder) is tethered. He intently watches Otto eat. - Then Sarah feeds Peanut a treat. She repeats this Step 1 exercise a number of times. Soon, Peanut doesn’t look at Otto when Otto is eating a treat . . . - Instead, the little guarder brightens and looks at Sarah every time Sarah feeds Otto, anticipating correctly that he (Peanut) will receive a treat next.

There are a few different options for modifying resource-guarding behavior between dogs. You can classically condition Dog A (the guarder) to love having another dog around him even in the presence of valuable resources; you can operantly condition Dog A to perform a different response when he’s in possession of a valuable resource and another dog approaches; and you can operantly condition Dog B to avoid the guarder when he has a valuable resource. Here’s how each of these work.

Counter-Condition Dog A
The point of counter-conditioning is to change Dog A’s emotional response to the proximity of Dog B in the presence of a guardable resource. This procedure will require dogs with very solid sit-stays and down-stays. Alternatively, you can use tethers. It’s critically important that Dog A not be triggered to guard during these training sessions; awareness of threshold distance and the dogs’ proximity to each other is paramount.

Step 1: Start with the two dogs sitting a few feet from each other – farther, if necessary to avoid guarding behavior. Have a bowl of pea-sized, high-value treats. Give a treat to Dog B (the non-guarder), and then give one to Dog A, accompanied by happy-voice praise. If the dogs are so far apart you have to walk some distance to get to Dog A, start praising as you walk. Repeat until you see Dog A brighten noticeably when Dog B gets his treat; this tells you he’s made the association between Dog B getting a treat and the next delicious treat coming to him. This is a “conditioned emotional response” (CER) – the physical manifestation of the emotional change that happens because of the pairing between the presence of the other dog and the arrival of a high-value treat.

If you start with the dogs far apart, when you have established a consistent CER with Dog A, gradually move them closer together, continuing with the counter-conditioning and achieving CERs at each new distance until the dogs are happily taking treats a few feet apart. Depending on your dogs, this could take one session or many.

Step 2: Have a bowl of high-value treats. Hang out with Dog A in a good-sized quiet room with the door closed – watch TV, read a book, work on the computer – but don’t feed him any treats. In fact, ignore him completely. After 20 to 30 minutes, bring Dog B into the room on leash and have him sit. Feed him a treat, then spend 20-30 seconds giving generous treats and praise to Dog A. Then remove Dog B from the room.

At varying intervals, bring Dog B back into the room and repeat the procedure – always bringing Dog B into the room before you make any move to reach for the treats in the bowl. Repeat until Dog A looks consistently happy – the CER – as you move to bring in Dog B.

Step 3: Repeat Step 2, but this time Dog A is in possession of a guardable resource – a bone, a toy, a favorite bed. If there are things he guards less intensely than others, start with a lower-value resource.

Tether Dog A with his resource as far from the door as possible, and hang out for another 20 to 30 minutes before bringing in Dog B. Step in the door with Dog B, have him sit, feed him a treat, and then do your 20- to 30-second praise-and-treat routine with Dog A. Repeat until you have a consistent CER – Dog A brightens happily when you enter the room with Dog B.

As you do these repetitions, occasionally encourage brief eye contact between the dogs several times at each distance, so that trigger gets built into the procedure. If eye contact triggers significantly greater intensity from Dog A, stay at the door until that intensity goes away and you are getting consistent CERs even with eye contact.

Gradually move closer with Dog B, obtaining consistent CERs from Dog A at each new distance before moving closer again. Remember to look for and reward  some eye contact between the dogs.

When you have closed the distance by half, mark that spot and start over again at the full distance, but this time with Dog A untethered. This should not trigger any tension from Dog A, and you should be able to move closer to him with Dog B more quickly than before.

Step 4: Repeat Step 3 with the same value resource, in all the rooms in the house, until Dog A displays consistent CERs everywhere.

If you have multiple dogs who Dog A guards from, repeat Steps 1 to 4 with each of the dogs. Then do the same thing with multiple dogs in the presence of Dog A and a low-value guardable resource.

Step 5: Go back to Step 3, again with Dog A tethered, but now in possession of a somewhat more valuable resource. Repeat all steps with all dogs, individually at first, then in larger groups, until Dog A offers consistent CERs regardless of what dog or what room. Continue up the resource-value ladder until Dog A shows no sign of tension

Step 6: Do occasional “cold trials” without the set-up and repetition – a deliberate “management failure” in which counter-conditioning meets real life. Do at least one cold trial per day, and if you see tension reappearing, go back and do repetitions of the procedure at whatever step is necessary for your dog to regain his equilibrium. Then move through the steps to completion – probably much more quickly than the first time through.

Operantly Condition Dog A
Resource-guarding behavior lends itself beautifully to the “Constructional Aggression Treatment” (CAT) operant conditioning protocol developed by Dr. Jesus Rosales Ruiz and Kellie Snider, MA, a few years ago at the University of North Texas. (See “Build Better Behavior,” WDJ May 2008). Our Dog A gets tense and eventually aggressive because he is concerned that the approaching dog is a threat to his valuable resource. These aggressive distance-increasing signals are often successful in making the other dog go away, hence they are reinforced, and behaviors that are reinforced are repeated.

Sarah demonstrates Step 3, where the guarder (Peanut) has a very guardable resource – in this case, a Kong toy stuffed with yummy treats. - She feeds Otto a treat, and then spends 20 to 30 seconds feeding one treat after another to Peanut, praising him warmly the whole time. - Within just a few repetitions, Peanut’s emotional response to the sight of Otto near his Kong has changed from guarded to a happy and anticipatory.

To use the CAT procedure, tether Dog A with a low-to-moderate-value guardable resource, and approach from a distance with Dog B. If you know that Dog A begins showing signs of guarding at a distance of 15 feet, start at 25 feet. As you approach, watch Dog A for any small sign of tension. The instant you see it, stop with Dog B, mark the spot, and wait. As soon as you see any decrease in Dog A’s tension, any sign of relaxation, quickly turn and walk away with Dog B, back to the 25-foot distance.

Pause there for at least 15 seconds (longer, if you think Dog A needs more recovery time), and then repeat, returning to the marker at the spot where you stopped before. Continue these repetitions until you see no sign of tension from Dog A when you arrive at the marker with Dog B.

On your next approach move four to six inches closer and mark that spot. You will likely see Dog A display signs of tension again at this distance. Repeat approaches and departures at this distance until the tension is gone, then decrease distance slightly again.

What you are doing with this procedure is teaching Dog A that a new behavior – acting calm and relaxed – makes the threat to his resource go away. As he continues to deliberately act calm and relaxed, he actually becomes calm and relaxed, and eventually no longer feels threatened by the approach of Dog B. Ideally you will see “crossover” behavior, where he acts very friendly and affiliative as Dog B approaches, offering distance-decreasing signals instead of his previous repertoire of distance-increasing signals. When you’ve worked through the procedure with low-to-moderate-value resources, repeat with high-value resources.

With counter-conditioning, you change your dog’s emotional response, and as a result his behavior changes. With operant conditioning (CAT), you change your dog’s behavioral response, and as a result his emotional response changes.

Operantly Condition Dog B
You can also operantly teach Dog B a new behavior in the presence of Dog A and a valuable resource. This is a useful second line of defense, in combination with modifying the behavior of the guarder. You can teach Dog B to withdraw on your cue; you can also teach Dog B to withdraw in response to any noticeable warnings from Dog A, such as a hard stare or a lip curl. The advantage of a cue from Dog A is that it happens, and Dog B responds by leaving, even if you’re not there to give your cue.

Eventually you may find that the mere presence of Dog A-with-resource becomes the cue for Dog B to leave, which is just fine and dandy. If you see Dog B leaving the room before the cue, go ahead and reinforce that – it’s a good thing!

When the cue is given (yours, or Dog A’s), guide (lure or prompt) Dog B to a specified target, ideally in another room. Throw a treat-and-praise fest there for Dog B, and hang out with him there for several minutes before returning to Dog A’s room and repeating the process. Dog B should soon be dashing to the other room when the cue is given – either yours, or Dog A’s lip curl.

So what happens if Dog A is in the designated target room with a valuable resource? Great question! It’s a good idea to operantly condition Dog B to a second target location in a different room. When Dog A is in Room X, Dog B learns to target to the spot in Room Y. But if Dog A is in Room Y, Dog B learns to target to a spot in Room X.

It’s Worth the Work
Keep in mind that you are likely to always need some degree of management, even with your successful modification programs. For example, even if you’ve done a great job of modifying the behavior of a dog who tends to guard toys, the high-arousal of a dog-to-dog game of tug carries a high likelihood of retriggering guarding aggression. Reserve his tug playing for games with you, and limit his play time with his canine pals to romping and running games. Be smart. Manage as needed, keep your eyes open for signs of returning tension, and be prepared to do a little remedial modification as needed.

So there you have it. Select the method(s) that appeal to you and get started. It will do your heart good to see the decreasing tension between your canine family members. It will also be gratifying to see your guarder gain new associations and learn new behaviors without fear of losing his valuables to his four-legged siblings.

Pat Miller, CBCC-KA, CPDT-KA, CDBC, is WDJ’s Training Editor. Author of numerous books on positive dog training, she lives in Fairplay, Maryland, site of her Peaceable Paws training center, where she offers dog training classes and courses for trainers.

Thanks to Sarah Richardson, CPDT-KA, CDBC, of The Canine Connection in Chico, California, for modeling with Otto and Peanut.

Comments (11)

My 9 month old dog just started this week, by resource guarding my food. If my children or I are eating and one of our other dogs come up she attacks them. No warning or anything. The behavior just started this week so I'm a little confused and afraid that one of them is going to get hurt. I have started making sure she is not in the room when we eat, but this isn't going to solve the problem. Any advice would be wonderful. Thanks

Posted by: momgove | August 29, 2014 12:08 PM    Report this comment

I've recently been helping a couple with their 2 shih tzu's. Bella is resource guarding toys (mainly 2 particular toys) from Ruby. However, she is often the instigator by taking the toy and sitting or lying right next to Ruby and then directing warning snaps at Ruby's face. Ruby looks away or shows some other calming signals but doesn't move away from Bella, nor does she try to take the toy. Has anyone else experienced this type of behaviour?

Posted by: CommuniCanine | July 28, 2014 12:12 AM    Report this comment

@jetdoglove Your problem is in management of your "pack" more so that resource guarding. I have been training professionally and doing group dog walks for many years and I always make sure to separate the dogs into crates or partitioned off sections of my car when I am doing group walks. In an open area the dogs can choose to avoid each other when necessary but in a car they are forced together and this is very uncomfortable for some dogs. Imagine having an acquaintance sit in your lap for an entire car ride, that is not something many of us could or would tolerate. For your little stress case: put her in a crate that is covered so she feels safe and pick her up last to avoid pushing her past her threshold.

Posted by: erika | May 27, 2014 8:58 PM    Report this comment

This is helpful, but one of the guarding (greedy) problems I have is my newer rescue going up and "snatching" toys out of the other dogs mouth. Luckily my dog B, (whose toys have been getting snatched) has been very tolerant. How does one stop dog A from being a "snatcher"!!

Posted by: haverescues | May 16, 2014 10:07 AM    Report this comment

This is helpful, but one of the guarding (greedy) problems I have is my newer rescue going up and "snatching" toys out of the other dogs mouth. Luckily my dog B, (whose toys have been getting snatched) has been very tolerant. How does one stop dog A from being a "snatcher"!!

Posted by: haverescues | May 16, 2014 10:07 AM    Report this comment

I'm a dog walker who has a fairly new dog in the pack, Winky. (I pick up up to 6 dogs and take them to off leash areas for adventures). Winky (Dog A) is always one of the first dogs to be picked up and is stressed in the (all) car (tail between her legs). The last couple of times we've gone to pick up a couple of the other dogs in her pack (who she plays well with during the adventures) she's growled/snapped at them (a different dog each time) (Dog B) getting in the car. I'm still not absolutely sure this is resource guarding (of the car), but I think so. How can I train her given she's guarding (assuming that's what this is about) the car that all the dogs need to ride in? I'm managing by bringing her out of the car before the other dogs get in so she doesn't associate it as 'hers' and she can then follow, but I'd like a conditioned emotional response. Thanks for your help.

Posted by: jetdoglove | March 25, 2014 12:43 AM    Report this comment

@Sandcastle: My suggestion would be to follow the modification training provided here with a borrowed dog. The one you dog sat for, or a friend or neighbor's dog. Different dogs if you can. This will help your dog to understand that what you are teaching him, is for his behavior with all dogs, not just the one you are dog sitting.

Posted by: ailurafil | January 9, 2014 4:40 AM    Report this comment

My dog is an only dog. I recently kept a friend's dog for the weekend and learned that my dog will engage in behavior similar to Dog A in "Now We're in Trouble Part I." What is your suggestion if I do not have a "Dog B" to help with my dog's training?

Posted by: sandcastle | July 22, 2013 1:47 PM    Report this comment

Many trainers steer clear of putting labels on dogs as it doesn't help them in the behavior modification process. What we (trainers) do is describe what the dog (or dogs)is/are actually doing so that we can create plans to modify the behavior. We look at antecedents, the actual behavior that you'd like to change, and then the consequences of that behavior. We look at the functions of the behavior - what does the dog "get" by doing what it does. It's really hard to determine whether or not a dog is doing something to be dominant or submissive. That's like saying you know my own motives for doing something. But we certainly can quantify behavior and by using solid training techniques as described in the article, we can measure whether or not the behavior we like or dislike is increasing or decreasing. It's easier than trying to figure out a dog's motive.

Posted by: hpytls | August 7, 2012 7:35 PM    Report this comment

It can also be medically related. My dog, now diagnosed with EPI, started guarding her food from our other dog a few months before her diagnosis. Turns out her pancreas was atrophying and she was literally starving.

Posted by: Paula G | May 18, 2012 4:46 PM    Report this comment

If your looking at possible dog to dog recource guarding how can you tell the difference between that and being dominant?

Posted by: ajastar79 | October 14, 2011 9:24 AM    Report this comment

New to Whole Dog Journal? Register for Free!

Already Registered?
Log In