Leash Reactivity in Dogs: Why Your Dog Loses It on Leash

You know the scene. You turn a corner with your dog, and the walk is going well. Then another dog appears at the end of the street. Yours freezes, the tail goes up, there’s a low growl—or he barks, lunges, and pulls at the leash. You stay calm, hold out a treat, call his name, maybe a sharp “No!” None of it gets through to him. And you’re left standing there feeling like you have no control over your own dog.

The usual advice stays on the surface: stay calm, distract him, practice more. None of that is wrong in principle. But it misses two points that matter more than anything else in this situation: your dog’s state, and the question of what the behavior actually does for him.

What you’ll read here comes from two sources: the daily work of our dog training school, where leash reactivity is one of the most common reasons owners come to us, and the research literature. Where the two agree, the recommendation is solid. Where they diverge, we say so explicitly.

In brief

Your dog doesn’t lose it out of disobedience or dominance. He is in a state of high arousal. In technical terms, the word is neutral: in this state, the ability to process information flexibly and respond to familiar cues can be impaired. Joy and fear look similar from the outside—which is why the visible activation alone doesn’t tell you reliably what motivation lies underneath.

There’s a second reason the behavior sticks: it works. For some dogs, barking can help create distance; pulling can shorten the gap to the other dog. And the other dog walks past sooner or later anyway. From your dog’s point of view, the behavior can therefore seem successful, because the behavior is followed by growing distance from the trigger.

Three things follow from this. First, your dog only learns something new below his threshold; above it, less new learning tends to take place, and the existing reaction keeps being shown. Second, distance is the lever you actually hold in your hands—it’s easier to change than your dog’s state. Third, punishment can cause additional problems here—especially if your dog links the trigger with the unpleasant experience.

And one more thing that regularly gets overlooked in everyday training: pain can play a part in behavior problems and should be considered whenever behavior changes suddenly (Mills et al., 2020). More on that below.

1. Arousal is a state, not an emotion

When a dog blows up on leash, people often read it as an emotion. He’s angry. He’s scared. He’s frustrated. Some of that may be true. But it doesn’t explain why he can no longer hear you in that moment.

Something else is responsible for that. A dog who is overjoyed to see visitors behaves, in some respects, like a dog who is afraid: both are fast, both are scattered, both are harder to reach, both take food differently than usual. The emotions are opposites; the state is similar.

How worked up a dog is says nothing about whether he feels good or bad. It only says how much is going on inside him right now.

That doesn’t mean the emotion is irrelevant. For long-term work, it matters a great deal whether your dog wants distance or desperately wants to get to the other dog. That determines what you build. But for the question of what you can do in the next twenty seconds, a different question comes first: how high up is he right now?

This distinction takes some weight off your shoulders. You don’t have to assume there’s something wrong with your dog. And you don’t have to ask yourself what you did wrong in raising him.

What can lie behind the behavior

Once the acute moment has passed, it’s worth looking at motivation. Four patterns are common, and they aren’t mutually exclusive. In many dogs, several are mixed together.

He wants distance. The other dog is a stranger, or there have been bad experiences. Barking and growling are then warning signals, not eagerness to attack. Typical signs: tucked tail, crouched posture, trembling. For context: individual body signals are not proof on their own—what counts is the overall picture of behavior, situation, and the individual dog’s history.

He wants to get there. Your dog wants to sniff, play, run, but the leash holds him back. For a social dog, that is extremely frustrating. Typical signs: lunging forward, a stiff, straight tail, high-pitched barking.

He revs up quickly. Some dogs have a low arousal threshold. Scent, movement, proximity: everything is exciting, and the system ramps up fast. Typical signs: frantic body language, panting without exertion, fixating.

He sees no way out. Poor socialization, stressful past experiences, a difficult start in life. Every encounter is a risk, and on leash the option of walking away disappears.

That last point leads to something that is often attributed to the dog but actually belongs to the situation.

2. Room to maneuver makes a difference

When a dog ends up in an unpleasant situation, he basically has several options. He can move away, stand still and wait, do something else entirely, or go forward. Which one he chooses depends not only on him but on which options are available at all.

In open ground with plenty of distance, he has every option. On leash on a narrow path, with a hedge on one side and a road on the other, those options are much more limited. They aren’t gone entirely: even on leash, dogs can walk a curve, look away, or create distance through body position—if the human gives them the room to do it.

Taking away a dog’s ability to create distance can increase the likelihood that he will use other strategies—moving forward, vocalizing, pulling.

This turns a common interpretation on its head. “He’s just a daredevil” describes the dog. In fact, the behavior in this situation often describes the position he’s in. The same dog can turn away calmly at the edge of the woods and go forward in the hallway of an apartment building. That isn’t a contradiction—it’s two situations with different room to maneuver.

How much the leash itself changes things is shown by an observational study of 1,870 dogs in public spaces: dogs off leash sniffed each other considerably more often than dogs on leash. Sniffing is the normal course of an encounter. It clarifies who the other dog is and usually ends the matter. The leash changes what is possible in an encounter: direct approach, moving aside, or breaking off can all be restricted.

What was observed was what the dogs did, not how they felt while doing it. The association is well documented; the interpretation remains an inference.

👉 Arousal Regulation in Dogs: Self-Control, Learning and Calm

👉 How to read dog body language

👉 Comforting a fearful dog: is it okay?

👉 Fearful dogs: the basics

👉 Reactivity Is Not Aggression: The Neuroscience of the Lunging Dog

👉 The Neurobiology of Frustration in Dogs

3. Why the behavior persists

Picture the situation from your dog’s point of view. Something unpleasant is approaching. He can’t get away. He gets louder, moves forward, makes himself big. And after ten or twenty seconds, the other dog is gone.

The unpleasant thing has stopped. After the behavior. Your dog doesn’t know that the other dog would have walked on anyway. He only knows what he did and what happened next.

It works the same way in reverse: a dog who pulls and gets closer in the process has learned that pulling gets him where he wants to go.

This kind of reward is often more powerful than a treat from your hand. It happens without a detour through you, without delay, without your timing having to be right. The dog controls the success himself. Your treat is competing with that under the worst possible conditions.

Why ten good encounters and one bad one don’t add up to zero

Here comes the point that drives many owners to despair. You practice for weeks, it works, ten encounters go calmly. Then one goes wrong, and the setback feels out of all proportion.

That feeling has a basis. Intermittently reinforced behavior can become particularly resistant to extinction—that is, it can persist even once it no longer pays off. That is one reason why occasional successful lunging or pulling may help maintain the pattern, even when most encounters now go calmly. Slot machines are a familiar illustration of this kind of persistence: people keep playing even though wins come only now and then.

For you, this means: avoiding failed encounters is not a side issue. It is just as important as the training itself, and it’s the reason management isn’t a substitute for training but the precondition for it.

👉 Learning theory in dog training

👉 Aggressive behavior in dogs: not simply innate

4. The threshold, and what happens above it

People like to imagine the threshold as a switch: fine up to here, gone from here on. That’s not how it works. It’s more of a slope:

• For the first stretch, your dog works normally
• Then he gets slower to respond and needs the cue twice instead of once
• Then he still performs it, but sloppily
• Then he still takes the food, but spits it out again
• Then he won’t take anything at all
• Then he stops looking at you

This gradation is the most important observation in the whole topic. If you can tell where your dog is right now, you can act before he reaches the bottom.

Above the threshold, it’s not that nothing happens. Something other than what you intend happens: your dog isn’t repeating the exercise—he’s repeating the experience. And that experience is: here comes a dog, it’s getting tight, I can’t get away. A dog who goes through an encounter ten times in that state hasn’t practiced ten times; he has confirmed ten times that encounters are unpleasant.

On top of that, the trigger becomes more distinct over time. At first it was a particular dog in a particular situation. After many bad encounters, the silhouette at the end of the street is enough. And you become part of the pattern, too: your tensing up, shortening the leash, your tone of voice—they come every time, just before the unpleasant thing. That explains why some owners report that their dog blows up before anything is even there. Something is there—just not what they think.

5. What the research says

Whether excitement impairs self-control was examined in 106 dogs (Bray, MacLean & Hare, 2015). They had to solve a detour task that requires impulse control. Then one thing was changed: the people cheered the dogs on enthusiastically.

In calm assistance dogs, this worked—their performance improved. In the family dogs, who were more excitable to begin with, it got worse, with exactly the same encouragement.

The same action, two opposite effects. The difference didn’t lie in what the human did, but in how worked up the dog already was.

For the leash situation, this means: once your dog has already ramped up, an excited tone of voice usually doesn’t make things better—not even a friendly one. That applies to the cheerful “Look over here!” just as much as to talking to him loudly.

A review on stress and cognitive function (Mendl, 1999) adds to this: stress can narrow and shift attention, and moderate levels of stress hormones can support memory formation, while high or prolonged levels impair it. The often-cited rule that a little stress helps with simple tasks and that things tip over sooner the harder the task (the Yerkes–Dodson law) is described there as a model with limited explanatory power. Walking past an unfamiliar dog is a hard task for your dog, harder than it looks. If you also ask for a sit, eye contact, or heel, you make it harder still.

Asking for the hardest exercise in the hardest situation means planning for failure.

For context: these studies took place under controlled conditions, not on a walking path. What is supported is the principle, not the specific instruction.

6. Check first: is your dog in pain?

This is the most frequently overlooked point in the entire field of behavior, and there’s a good reason it comes before the training section.

In a review of cases referred to veterinary behaviorists, a painful component was found in roughly one third; the proportion varied considerably between individual practices. These aren’t dogs who limp. They are dogs who were presented for behavior problems, and in whom it turned out that something hurts.

A dog with chronic pain often shows no lameness. What he shows are changes in behavior: lower resilience, quicker reactions, less tolerance for closeness. Exactly what stands out in the leash situation. To complicate matters, pain perception can be temporarily dampened under strong excitement. So a dog who shows nothing in the moment is not necessarily pain-free.

What should make you think of pain:

• The behavior has changed without anything else having changed
• He is noticeably different on some days than on others
• He is more sensitive to touch, brushing, being lifted
• He is over six years old and the problem is new
• Jumping, stairs, or getting up look different than they used to

For context: these analyses concern dogs who had already been referred to specialists. The proportion in the general dog population is likely lower. What they show is that pain is regularly overlooked when nobody looks for it.

Before you train for months, have your dog checked for anything that needs treatment. That’s not a delay to the training—it’s the precondition for it.

👉 Chronic Pain and Behavior: The Neurobiological Link Between Osteoarthritis and Aggression in Dogs

👉 Recognizing pain in dogs

We regularly send dogs for an orthopedic workup before starting the actual training, especially when a dog who was previously unremarkable suddenly starts reacting, or when his day-to-day form varies a lot. A training plan for a dog with untreated pain doesn’t fail because of the plan.

7. Management and training are two different things

A note on the practical part: The following steps are derived from well-studied learning principles—working below the threshold, gradual approach, reinforcing alternative behavior—and from our training practice. We know of no controlled studies that test them as a specific procedure for leash reactivity. The numbers in this part—distances, weeks, sessions—are practical guidelines, not study results.

This is where most of the confusion arises in practice, so here is the distinction spelled out:

Management is everything you do when the distance isn’t enough. Turning around, stepping aside, ducking behind a car, walking in an arc, positioning yourself between your dog and the other dog as a screen. The goal isn’t to teach anything, but to avoid the encounter before it escalates. Without comment, without rushing, without a guilty conscience.

Training is what you do at sufficient distance—where your dog is still with you. And there, a different rule applies, one that contradicts management: you don’t distract.

Both are right, just not at the same time. Distracting during training prevents the exercise. Waiting during management risks escalation. The distance decides which of the two modes you are in.

The first two to three weeks: management only

That sounds like a bad start, and it’s the most important part. As long as your dog keeps regularly experiencing encounters in which his behavior works, you are working against yourself.

In practice, that means changing routes and shifting times. Notice when most dogs are out on your routes—shifting your walk by half an hour can often avoid many encounters. Turning around is allowed. Use visual cover: two seconds behind a hedge replace twenty seconds of confrontation.

8. Determining the distance

Now it’s about finding a number—not estimated, but observed.

The question isn’t “at what point does he bark?” but: up to what distance is he still with me? That’s a different number, and it’s considerably larger. Between “still reachable” and “blows up” there are often twenty or thirty meters.

Find a spot with a good overview where dogs pass by at some distance. Stand far enough away—the first time, eighty meters is perfectly fine. When a dog comes into view, watch your dog, not the other one. And go through the stages: Does he keep sniffing? Does he look at you on his own? Does he respond to his name the first time? Does he take food, and how?

Next time, a little closer, in steps of ten meters. Expect it to take two to three weeks before you have a reliable number.

In practice: in our experience, the distance owners name in the first consultation is considerably smaller than what turns up when it’s actually measured. Someone who says “about ten meters, then it starts” is often already at thirty when we observe. That’s not a criticism. From above, with the leash in your hand and your own tension in your neck, you simply don’t see the early signs. And the distance fluctuates: with wind, with rain, when your dog is tired, when the other dog heads straight toward you instead of walking past.

This distance isn’t a limit you have to respect. It’s the place where your training happens.

9. The basic exercise

You’re standing at your distance. A dog comes into view.

Your dog looks. You wait. Don’t talk to him, don’t lure him, don’t cheer him on. The reason was explained above.

He looks back at you on his own. That’s the moment. Now he gets the food.

If he doesn’t look on his own, the distance was too short. Don’t keep going; next time, move farther away.

That’s all there is to it. No sit, no heel, no eye contact on cue. Just: being allowed to look, coming back on his own, getting something for it.

It seems sparse, and that’s intentional. Every additional demand makes the task harder, and the harder the task, the less excitement it can tolerate.

Only move closer once it has worked without hesitation in four to five consecutive sessions. Then ten meters. If it doesn’t work at the new distance: one step back. That isn’t a setback, but information that the step was too big.

How often: this is where the data are surprising

One study with 44 dogs examined how the frequency and duration of training sessions affect learning. Two results: dogs who trained once or twice a week learned better than dogs who trained daily. And dogs with one session per day learned better than those with three sessions back to back. How far this goes: this is a laboratory study on acquiring one specific task with 44 beagles—not leash reactivity training. It doesn’t automatically follow that behavior training has the same optimal frequency; but the finding is a reminder that quality and recovery matter. These results come from a learning task under controlled conditions and cannot be transferred directly to every kind of behavior therapy. An earlier paper involving the same research group (Meyer & Ladewig, 2008) compared 18 laboratory beagles trained once a week or five times a week: the dogs trained once a week needed fewer sessions to learn the task, while the dogs trained five times a week reached it in fewer calendar days.

That runs counter to everyday intuition. But at least for acquiring a new task, “practice, practice, practice” was not the best approach in these studies.

In practice: two fixed sessions per week for distance training, five to ten minutes per session, three to five encounters from a distance. The rest of the week is management. Stop while it’s going well, not once it goes wrong.

For context: what was studied were laboratory beagles under controlled conditions, on a single task. Transferring this to everyday distance training is an inference. That two papers from the same research group point in the same direction supports the finding for task acquisition, measured by the number of sessions needed; whether it holds for behavior training has not been studied.

👉 Motivation in dog training

10. Why it comes back, and why that’s normal

At some point, usually after three to five good weeks, it suddenly goes wrong again. In a place where it had long been working, for no apparent reason. Many owners immediately interpret this the same way: so it didn’t work after all.

That’s the wrong conclusion. When a behavior is reduced, the original learning isn’t erased; it remains. What gets added is a second, newer experience, and that one is strongly tied to the circumstances in which it was made. Three patterns follow from this, and all of them are normal: it comes back after a break. It comes back in a different place. And it comes back after a surprise.

A relapse is not proof that the training isn’t working. It is to be expected. If you know that, you don’t give up.

In our work, this exact point is where people most often quit. Not in the first weeks, when it’s hard, but at the setback after a phase when things were going well. Owners who know in advance that it will come handle it differently from owners who take it for the end.

What you do then: don’t practice harder; go back to the last distance at which it reliably worked and rebuild from there. It goes faster the second time. And: practice in other places as soon as it works in one.

👉 Extinction in Dog Behavior: Why Learned Behavior Disappears—and Returns

👉 Desensitization and Counterconditioning in Dogs

These findings come mostly from basic research, a substantial part of it from animal experiments with rodents. The principle is considered well established; the specific transfer to leash encounters remains an inference.

11. The typical mistakes

This list doesn’t come from the literature but from what we encounter most often in training. Most owners have done some of these things for a while. That’s no reason to beat yourself up—it’s the norm.

Getting too close. People think in meters and consider ten generous. For your dog, ten meters may be well above the threshold.

Making him tough it out. “Just let him be, he’ll get used to it” doesn’t work. Habituation only happens below the threshold.

Distracting him as soon as he looks. Sounds sensible, but it interrupts exactly the process that matters—at least in training. In management, when the distance isn’t enough, changing direction is the right call.

Punishment. Leash jerks, “No!”, startle stimuli. Your dog doesn’t learn “barking is wrong”—he learns that the encounter becomes even more unpleasant.

Constant chatter. “Sit… no… come… look… stay…” The flood of words turns into noise. Better: a single, previously trained cue, once or twice. Then act.

Judging the outcome instead of the process. “It went fine” is a statement about how it ended, not about your dog’s state. An encounter can look calm on the outside and still have been a close call.

12. Equipment

Equipment doesn’t decide anything, but it can make training unnecessarily difficult.

A leash that is too short takes away freedom of movement and the option to move aside, and you need both. A longer, sturdy leash of three to five meters and a well-fitting Y-harness with no pressure on the neck and airway are the uncomplicated choice.

Retractable leashes aren’t bad in principle; with reactive behavior, though, they usually get in the way, because they are under constant tension and can shoot out in a moment of fright.

Choke chains, prong collars, and correction chains cause pain or pressure. Depending on where you live, their use in training may be restricted or prohibited by law.

👉 How dogs understand us: words, gestures, gaze and tone of voice

👉 Dogs in public spaces: how to be a responsible dog owner

👉 The risks of retractable leashes

👉 The risks of tightening anti-pull harnesses

13. What the research doesn’t show yet

Some of what is considered settled in this area isn’t.

You often read that lip licking, yawning, blinking, and sudden sniffing are the dog’s way of defusing the situation. This interpretation is widespread, but it is not clearly established. In a study with 116 dogs (Firnkes et al., 2017), lip licking and looking away occurred less often in the highly threatening situations than in the less threatening ones. A pilot study of 24 dogs in off-leash encounters with other dogs (Mariti et al., 2017) points in the other direction: when a dog showed such signals after the other dog had behaved aggressively, the encounter de-escalated in most cases.

Individual signals such as yawning, licking, or looking away are not unambiguous. What counts is the context, the combination of several signals, and the individual dog’s history. They can indicate tension, but they are no proof that the dog is managing the situation. And the more dangerous point: their absence is not a good sign. If a dog who usually licks his lips suddenly stops doing so, that can mean the situation has become more serious for him.

Both papers have limited samples and examine test situations, not leash walks; the second is explicitly a pilot study. Taken together, the findings are mixed: they do not settle whether these signals defuse a situation—and they do not show that the signals are meaningless.

It also remains open how well the findings from controlled studies transfer to real leash encounters. Almost all of the studies mentioned here were carried out under laboratory conditions or in standardized tests. For the leash situation itself, there are hardly any robust intervention studies. What works in practice rests on the principles, not on direct evidence of effectiveness.

14. When an article is no longer enough

In some constellations, professional support makes sense and is sometimes necessary: after a bite incident, if your dog also reacts to people, if the behavior has changed suddenly within a few weeks, if you have worked carefully for eight to ten weeks and the distance hasn’t gotten any smaller, or if you start planning your routes so you won’t meet anyone.

That last point is not a sign of weakness. It’s the point at which it’s no longer just about the dog.

Incidentally, what someone else sees that you don’t isn’t a matter of skill but of position: you see your dog from above, and from that angle ear position, weight shifts, and muscle tension are barely visible. From the side and from a few meters away, you can see them clearly. Besides, you are part of the situation, and nobody can watch themselves tense up.

👉 A sign that your dog needs space: The Yellow Dog Project

15. Conclusion

Leash encounters are demanding, but they aren’t destiny.

Your dog doesn’t lose it out of disobedience, but because he is in a state in which thinking is hard, and because the behavior works for him. The lever is distance, not consequences. Training begins where your dog can still learn, and management is the precondition for it to work.

A dog who blows up on leash is not a bad dog. And you are not a bad owner if you get help with it.

References

• Řezáč, P., Viziová, P., Dobešová, M., Havlíček, Z. & Pospíšilová, D. (2011): Factors affecting dog–dog interactions on walks with their owners. Applied Animal Behaviour Science 134(3–4): 170–176. https://doi.org/10.1016/j.applanim.2011.08.006 (1,870 observed dog–dog interactions in public spaces; what was observed was behavior, not well-being)
• Firnkes, A., Bartels, A., Bidoli, E. & Erhard, M. (2017): Appeasement signals used by dogs during dog–human communication. Journal of Veterinary Behavior 19: 35–44. https://doi.org/10.1016/j.jveb.2016.12.012 (116 dogs; specifically examined looking away and lip licking)
• Demant, H., Ladewig, J., Balsby, T. J. S. & Dabelsteen, T. (2011): The effect of frequency and duration of training sessions on acquisition and long-term memory in dogs. Applied Animal Behaviour Science 133(3–4): 228–234. https://doi.org/10.1016/j.applanim.2011.05.010 (44 beagles, laboratory study on task acquisition – not behavior training)
• Mills, D. S., et al. (2020): Pain and problem behavior in cats and dogs. Animals 10(2): 318. https://doi.org/10.3390/ani10020318 (commentary with clinical case series, not a prevalence study)
• Herron, M. E., Shofer, F. S. & Reisner, I. R. (2009): Survey of the use and outcome of confrontational and non-confrontational training methods in client-owned dogs showing undesired behaviors. Applied Animal Behaviour Science 117: 47–54. https://doi.org/10.1016/j.applanim.2008.12.011 (survey study, clinical sample)
• Bray, E. E., MacLean, E. L. & Hare, B. A. (2015): Increasing arousal enhances inhibitory control in calm but not excitable dogs. Animal Cognition 18(6): 1317–1329. https://doi.org/10.1007/s10071-015-0901-1 (106 dogs, assistance dogs and pet dogs; detour task under controlled conditions)
• Mendl, M. (1999): Performing under pressure: stress and cognitive function. Applied Animal Behaviour Science 65(3): 221–244. https://doi.org/10.1016/S0168-1591(99)00088-X (review, not a study of its own; summarizes findings from humans and various animal species)
• Meyer, I. & Ladewig, J. (2008): The relationship between number of training sessions per week and learning in dogs. Applied Animal Behaviour Science 111(3–4): 311–320. https://doi.org/10.1016/j.applanim.2007.06.016 (18 laboratory beagles, trained once vs. five times a week; shaping task with a clicker – not behavior training)
• Mariti, C., Falaschi, C., Zilocchi, M., Fatjó, J., Sighieri, C., Ogi, A. & Gazzano, A. (2017): Analysis of the intraspecific visual communication in the domestic dog (Canis familiaris): A pilot study on the case of calming signals. Journal of Veterinary Behavior 18: 49–55. https://doi.org/10.1016/j.jveb.2016.12.009 (pilot study, 24 dogs in off-leash encounters with other dogs)

Key takeaways on leash encounters

Arousal is a state, not an emotion. Joy and fear look similar from the outside on leash—the visible intensity says little about what lies behind it.

The lack of room to maneuver is at the core of the problem. Off leash, dogs could regulate distance and end the matter; on leash, exactly that option disappears.

The behavior can be maintained by its consequences. The other dog walks past anyway—the behavior is followed by growing distance from the trigger.

Ten good encounters and one bad one don’t add up to zero: intermittently reinforced behavior can become particularly resistant to extinction, so occasional successful lunging or pulling may help maintain the pattern.

Pain comes before training. Pain doesn’t always show up as obvious lameness—more often as lower resilience and quicker reactions.

Management and training are two different things. In the first weeks, the only goal is to avoid encounters so the old behavior doesn’t keep getting practiced.

Distance is the most important adjustment—and in our experience, it is usually greater than it feels.

Practicing less can be better. A laboratory study on task acquisition argues against daily sessions—but that doesn’t automatically transfer to behavior training.

It’s normal for things to get worse again after a few good weeks, and it isn’t a step backward—that is exactly where people most often quit.

Frequently asked questions about leash encounters