Leash Reactivity Isn't Aggression — Why "Watch Me" and Corrections Train the Wrong Response, and What the LAT + Threshold Protocol Actually Does

Your dog loses it on walks. Another dog appears across the street and yours erupts — barking, lunging, pulling hard enough to bruise your hand. You've tried treats. You've tried "watch me." You've tried corrections. Nothing sticks.

Here's what most guides won't tell you: the standard advice for reactive dogs doesn't just fail to help — in many cases, it makes the problem worse. Not because you're doing it wrong. Because the advice is wrong.

This post covers what leash reactivity actually is, why the three most common interventions don't fix it, and the evidence-based protocol that does.

Reactivity Is Not Aggression

This distinction matters because it determines the entire treatment approach.

When your dog barks and lunges at another dog on leash, most people — and many trainers — interpret it as aggression. The dog wants to attack. You need to establish dominance. You need to "correct" the behavior before it escalates.

None of that is accurate.

Karen Overall, veterinary behaviorist and author of the Manual of Clinical Behavioral Medicine for Dogs and Cats, frames reactive behavior as primarily fear-driven or frustration-driven — not predatory or dominance-based. Jean Donaldson, founder of the Academy for Dog Trainers, makes the same point in Mine! and her broader writing: most reactive dogs are emotionally over-threshold, not aggressive.

What "over threshold" means: Every dog has a trigger threshold — the point at which a stimulus becomes too intense to process calmly. Below that threshold, a dog can see another dog, notice it, and continue functioning. Above it, the brain shifts into a threat-response state: fight, flight, or freeze. Barking and lunging is usually fight-mode output from a dog that is overwhelmed and has no other trained response available.

Most reactive dogs, if you met them off-leash in a neutral space, would not attack. Many would play. The leash creates a constraint: they can't flee, can't approach normally, and the restrained, tense body posture of a dog on a short leash reads as threat posture to other dogs. The result is mutual escalation.

What's actually happening emotionally: Your dog has formed a conditioned emotional response (CER) to on-leash dogs. The sight of another dog predicts tension, restriction, and — in many dogs who've been corrected — pain or startle. The emotional response is negative. The behavior (barking, lunging) is just an expression of that emotion.

You cannot train your way out of a negative CER using behavior alone. You have to change the emotion first.

Why "Watch Me" Alone Doesn't Fix It

Eye contact cues are taught in nearly every reactive dog class. The reasoning sounds solid: if your dog is looking at you, they can't be reacting to the trigger. Get the attention, break the reaction.

The problem is the mechanism — or the lack of one.

"Watch me" teaches a behavior. It doesn't change the emotional state driving the reaction. Your dog looks at you because there's a treat in your hand, not because they feel differently about the dog across the street. The moment the treat disappears, or the trigger gets close enough, the original CER takes over.

Think of it this way: if someone with a severe phobia of spiders is told to stare at the wall instead of looking at the spider, their cortisol doesn't drop. The spider is still there. The fear is still there. You've just redirected their gaze.

Eye contact as a coping skill — something a dog offers voluntarily after genuine threshold progress — is valuable. Eye contact as a substitute for that progress is suppression.

This is why dogs who have been in "watch me" training for months can still explode the second they see a dog they weren't prepared for. The trigger hasn't been reconditioned. The underlying CER is intact.

Watch me can be part of the protocol, but it cannot be the protocol.

Why Corrections Make Reactivity Worse

Leash jerks, prong collars, and e-collar stimulation are still widely used for reactive dogs — often sold as "interrupting" the reaction or teaching the dog that the behavior has consequences.

The research doesn't support this. It's worse than that: the research shows it reliably makes the underlying problem more severe.

A 2009 survey study from the University of Pennsylvania School of Veterinary Medicine (Herron, Shofer & Reisner, Applied Animal Behaviour Science, 117: 47–54) surveyed dog owners about their use of confrontational training methods and the outcomes. The finding that gets cited most often: a significant proportion of dogs — roughly 25% or more — responded to confrontational methods with increased aggression. Dogs already presenting with aggression toward familiar people were more likely to respond aggressively to corrections than dogs presenting for other behavioral issues.

The mechanism is straightforward classical conditioning. If another dog appears, then a leash correction happens, the dog doesn't think "I shouldn't lunge." The dog thinks: other dog = pain/startle. You've added an aversive to a conditioned emotional response that was already negative. The association between the trigger and something bad has deepened.

The AVSAB (American Veterinary Society of Animal Behavior) 2021 position statement on humane dog training is direct: "Aversive training methods have a damaging effect on both animal welfare and the human-animal bond. There is no evidence that aversive methods are more effective than reward-based methods in any context."

The reason corrections sometimes appear to work short-term is suppression. The dog stops showing the visible behavior — the barking, the lunging — because it's learned that expressing the emotion causes pain. But the emotion is still there. In many cases it intensifies. What you end up with is a dog who is more afraid, more frustrated, and no longer giving warnings before snapping. The AAHA 2015 guidelines on behavior management put it clearly: "Aversive techniques are especially injurious to fearful and aggressive patients and often suppress signals of impending aggression, rendering any aggressive dog more dangerous."

This is not the outcome you want.

The Protocol That Actually Works: CC&D + LAT

The evidence-based approach to leash reactivity has two interlocking components: counter-conditioning and desensitization (CC&D), and Leslie McDevitt's Look At That (LAT) protocol from Control Unleashed.

Neither is complicated. Both require patience and consistent threshold management. Together, they address the actual problem — the conditioned emotional response — rather than suppressing its expression.

Step 1: Find Your Dog's Threshold

Before any formal training begins, you need to know your dog's threshold distance. This is the point at which your dog can notice the trigger without reacting — ears up, body stiffens slightly, but no barking, lunging, or hard staring.

Most owners don't know their dog's threshold because they've never deliberately worked at it. They try to pass other dogs on the sidewalk at 10 feet, which is well over threshold for most reactive dogs, and then wonder why the treats stop working.

Finding threshold:

• Walk your dog in a low-traffic area.
• When another dog appears, stop moving. Note the distance at which your dog notices the other dog but hasn't escalated.
• That distance is your starting point. Working distance is typically further than threshold — you want the trigger visible but at a distance where your dog can take food and orient back to you.

If your dog can't take treats when the trigger appears, you're over threshold. Increase distance.

Step 2: Counter-Conditioning — Change the Emotional Response

Counter-conditioning pairs the trigger with something your dog values highly. The mechanic is simple: other dog appears → high-value food appears. Other dog disappears → food disappears.

The timing is critical. The food must appear the moment the dog notices the trigger — before any reaction, before the emotional response escalates. The sequence needs to be: trigger → food, not reaction → food.

What this does: it starts to reprogram the CER. Other dog, which previously predicted tension/correction/fear, now predicts chicken or cheese. You're not teaching the dog to look at you. You're changing what the trigger means to the dog emotionally.

High-value reinforcement matters. Kibble doesn't cut it in high-arousal contexts. Use real food — small pieces of cooked chicken, cheese, hot dog. The reinforcer has to be meaningful enough to compete with the emotional arousal the trigger creates.

Session length: short. 3–5 minutes at most. Always end before your dog is exhausted or has reacted. A bad ending undoes progress.

The link between fear-based anxiety and this kind of threshold work is worth noting — the same mechanisms drive separation anxiety, and the protocol logic is similar: work below threshold consistently, never flood, let the emotional response shift before pushing the distance.

Step 3: Leslie McDevitt's Look At That (LAT) Protocol

LAT turns the trigger into a cue for a calmly offered behavior, with a positive outcome.

The core mechanic: click (or mark) the moment your dog looks at the trigger, then deliver the reward when your dog orients back to you.

The dog isn't being asked to ignore the trigger. The dog is being asked to acknowledge it and report it — and rewarded for doing so calmly. Over time, the sight of another dog triggers not a panic response but an operant chain: there's a dog → I look at it → I turn back to my person → I get rewarded.

Why this works, from the IAABC Foundation Journal analysis of LAT: unlike protocols that chain behavior around a trigger, LAT gives dogs decision-making power. The dog directs its own counter-conditioning process. The trigger becomes information rather than threat.

The verbal cue McDevitt uses — "Where's the dog?" — reinforces this open-ended acknowledgment. Eventually, fluent LAT dogs will give a casual ear-flick in the direction of a trigger and look back without the handler prompting at all. That's the sign the CER has changed.

How to introduce LAT:

1. Work at threshold distance or further.
2. Let your dog notice the trigger.
3. The moment they look at it — click or say "yes."
4. Reward by delivering the treat at your hip or chest, so your dog has to turn toward you to get it.
5. Repeat. The dog learns: look at trigger → look at human → treat.

Early LAT can be taught with neutral objects before progressing to actual triggers. If your dog is too aroused to function near dogs, start with a stationary bicycle or a person at distance.

What if your dog fixates instead of looking away? They're over threshold. Increase distance. If they're hard-staring and you click and they don't reorient, distance isn't enough — end the session. The protocol only works below threshold.

The CER mechanics here are directly analogous to what happens in resource guarding counter-conditioning: you're pairing the trigger with a positive outcome consistently enough that the trigger itself becomes a conditioned predictor of good things rather than a threat cue.

Common Mistakes That Stall Progress

Flooding. Going over threshold and hoping the dog "gets used to it." This is not desensitization — it's exposure without safety, and it deepens the negative CER. If your dog is reacting, you're not desensitizing. You're retraumatizing.

Using kibble. Kibble is appropriate for basic obedience in low-distraction environments. In the presence of a trigger that's causing cortisol elevation, kibble has no motivational pull. Use real food.

Ending sessions on a reaction. If your dog reacted and you're ending the session now, you've ended on a negative state. Push for one more successful look-and-reorient at increased distance, then end.

Working too long. Threshold management is cognitively taxing for dogs. Thirty minutes of reactive walks accomplishes less than three focused 5-minute threshold sessions. Short, structured, always below threshold.

Inconsistent distance. One session at 50 feet, next session at 15 feet because there's no room to maneuver. Progress requires consistency. If you can't maintain working distance in a given environment, change the environment. Management (avoiding known trigger zones during training phases) is not failure — it's preventing setbacks.

When to Call a Professional

Most leash reactivity responds well to this protocol with consistent application. But some cases warrant professional support:

• Bite history. If your dog has bitten another dog or redirected onto a handler during a reaction, you need a board-certified veterinary behaviorist (DACVB) or a certified applied animal behaviorist (CAAB). Bite history changes the risk profile.
• Redirected aggression onto handler. If your dog, when reacting to a trigger, is spinning and biting the leash or your hand, this is a safety issue that requires professional assessment.
• Multi-trigger reactivity. Dogs reactive to dogs, people, bicycles, strollers, and cars simultaneously are presenting with generalized anxiety, not simple leash reactivity. The protocol still applies, but the scope is broader and medication evaluation may be warranted.
• No progress after 8–12 weeks. If you've been working consistently below threshold with high-value food and LAT mechanics and seeing no change, consult a professional. Some dogs need medication (typically SSRIs or TCAs, prescribed by a DACVB or veterinarian) to lower baseline arousal enough for learning to occur.

Look for credentials: IAABC Certified Dog Behavior Consultant (CDBC), CCPDT Certified Professional Dog Trainer-Knowledge Assessed (CPDT-KA), or a DACVB. These designations require demonstrated knowledge of learning theory and behavior modification.

The puppy socialization window post covers the prevention angle — the reason socialization between 8–16 weeks matters so much is that it directly shapes the CER baseline dogs bring into adult life. Many reactive dogs didn't have fearful genetics; they had an undersocialized critical period. That context matters if you have a second dog coming.

What You're Actually Building

This protocol isn't about making your dog perform a different behavior when they see a trigger. It's about changing what the trigger means to them.

Done consistently, the dog who used to explode at the sight of a dog across the street starts doing something different: they notice the dog, glance at you, and keep walking. Not because they're suppressing the urge to react — because the urge isn't there. The trigger is no longer a threat signal. It's a predictor of good things.

That shift takes weeks, not days. It requires managing your environment, working below threshold every session, and resisting the shortcuts that feel faster but aren't. The corrections feel decisive. The prong collar gets a response. But what it gets you is a dog who's more afraid and less able to tell you when they're scared.

The protocol takes longer and demands more from you. It's also the one that works.

Sources

• Overall, K. (2013). Manual of Clinical Behavioral Medicine for Dogs and Cats. Elsevier.
• McDevitt, L. (2007). Control Unleashed: Creating a Focused and Confident Dog. Clean Run Productions.
• McDevitt, L. (2019). Control Unleashed: Reactive to Relaxed. Clean Run Productions.
• 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(1–2), 47–54.
• American Veterinary Society of Animal Behavior. (2021). AVSAB Humane Dog Training Position Statement. https://avsab.org/wp-content/uploads/2021/08/AVSAB-Humane-Dog-Training-Position-Statement-2021.pdf
• Donaldson, J. (1996). The Culture Clash. James & Kenneth Publishers.

FAQ (Schema Markup)

Q: Is leash reactivity the same as aggression? No. Most leash-reactive dogs are responding from fear or frustration, not predatory or dominance-based aggression. Many are perfectly calm off-leash. The leash creates constraint that amplifies emotional responses.

Q: Can leash reactivity be cured? "Cured" overstates it. With consistent counter-conditioning and LAT protocol work, most reactive dogs reach a functional baseline — able to pass other dogs calmly at reasonable distances. The CER rarely returns to a truly neutral state, but it can shift dramatically.

Q: Why doesn't "watch me" fix reactivity? "Watch me" redirects attention but doesn't change the emotional response to the trigger. The dog still feels the same about the other dog — they're just looking at you instead. When the treat disappears or the trigger gets too close, the original response returns.

Q: How long does the LAT protocol take? Varies by dog and history. Most owners see meaningful change in 6–12 weeks of consistent threshold work. Dogs with longer reactivity histories or higher baseline anxiety take longer. Consistency matters more than duration of any individual session.

Q: What if my dog won't take treats near other dogs? They're over threshold. Increase distance until they can accept food. If no distance exists in your current environment where they can take food, you need a controlled training environment — a large empty field or parking lot — to start.

Q: Do I need a professional trainer for a reactive dog? Not necessarily for mild to moderate reactivity with no bite history. The protocol in this post is well-documented and owner-implementable. Seek professional help if there's bite history, redirected aggression onto the handler, or no progress after 8–12 weeks of consistent work.

FetchCoach is an AI coaching app for dog training. For leash reactivity, the app walks you through the CC&D and LAT protocol step-by-step, adapts based on your dog's current threshold distance, and helps you track progress over time.