Fear & Emotion

Anything that sets off an emotional response in your dog. The skateboard, the doorbell, the man in the hat, the nail clippers, the sound of the smoke alarm being tested.

A trigger is a kind of antecedent, and the useful thing about naming it precisely is that triggers have dimensions you can turn down. Almost never is a trigger simply present or absent. It has a distance, a volume, a speed, a size, a duration, and a direction of movement, and every one of those is a dial.

The man in the hat at forty feet, walking away, is a different trigger from the man in the hat at six feet, walking straight at you. Same man. If you can only think of your dog's trigger as one thing, the work has no small steps in it, and small steps are the entire method.

So the first job with any trigger is to list its dials, and find the setting where your dog notices it and is still fine.

In Push Drop Stick

Plans turn the dials for you. A fear plan's early steps are the trigger at its mildest, and each step moves one dial by a little.

Often confused with

A distraction. A distraction competes for your dog's attention. A trigger changes how your dog feels. Something that makes your dog worried is not a distraction to be worked through.

Your dog can see, hear or smell the trigger, and is still able to function: eating, moving normally, able to look away from it, able to respond to you, loose in the body.

This is where all the work happens, and it is not the same as your dog being relaxed. Under threshold means your dog is coping. It includes mild, honest concern. A dog who notices the thing, checks it, and carries on is exactly where you want them.

If you are unsure whether you are under threshold, you are probably closer to the line than you would like. Add distance and see whether your dog changes. Distance is free and it is the most powerful dial you have.

In Push Drop Stick

Every step in a fear plan is written to keep you here. If a step keeps failing, that is the app telling you the setup is over your dog's line today, not that your dog needs to try harder.

The trigger has become too much, and the learning you want has stopped. Your dog is still learning, because classical conditioning never switches off, but what they are learning now is that this trigger is not safe.

What it looks like: barking, lunging, growling, snapping, spinning at the end of the leash. Or the quiet version, which is missed far more often: freezing, going flat, refusing food, hiding behind you, staring hard and still, trying to leave.

What matters is what you do next. Nothing you want can be taught here, and every extra second teaches the opposite, so there is no version of "push through it" that helps. Get your dog out: more distance, out of sight, into the car, back in the house. Leaving does not make the fear worse. You cannot reinforce a feeling, you can only help your dog feel safe, which is the foundation everything else is built on.

To be fully honest about it: getting away is what the barking and lunging are for, so the distance you give may reward the outburst a little. Leave anyway. Your dog's safety comes first every time, and the real fix is to leave sooner next time, at the first sign of worry, before there is an outburst to reward.

Then treat the episode as data. Something was set too high, and the answer is a smaller version of the same trigger next time, not a braver dog.

Expect the rest of the day to be harder than usual. The effects of a bad moment outlast it, so give your dog a genuinely easy evening and do not plan another exposure. See Trigger Stacking.

A feeling your dog has learned, attached to something that used to mean nothing.

Your dog was not born frightened of the vacuum or delighted by the sound of the treat pouch. Both are learned by prediction: one thing kept coming before another, and now the first one produces the feeling that belongs to the second. That is classical conditioning doing what it always does, in whichever direction the dog's history happened to point.

The reason the term is worth knowing is that it tells you what you are actually working on. When you counter-condition, you are not training a behavior and you are not teaching your dog to cope. You are changing what a thing predicts, and the feeling follows. A dog with a good CER to the nail clippers is not tolerating them; the clippers genuinely mean cheese now.

A useful test: what does your dog do when the thing appears and you have not asked for anything? Turning toward it, ears forward, looking for you expectantly, is a good CER. That is the goal, not compliance.

In Push Drop Stick

Comfort goals, shown with a party popper icon, earned on Classical steps. They record the one thing behavior alone cannot show you: that your dog's feeling about something changed.

Exposing your dog to a trigger at an intensity so mild it causes no fear, then raising it in small steps while your dog stays comfortable.

The word means "make less sensitive," and the whole method sits in that first sentence: so mild it causes no fear. Not manageable. Not "he settled after a minute." No fear. You might start with a recording played almost inaudibly, a strange dog visible at the far end of a field, or the nail clippers lying on the floor across the room while your dog eats dinner.

Only when your dog is genuinely comfortable do you turn one dial up a little. If you turn it up and see stress signals, you went too far, which is information rather than failure. Go back to where it was easy and make the next increase smaller.

Two rules that keep it honest. Move one dial at a time, because closer and louder and faster is a different exercise. And your dog can always leave. An exposure your dog cannot escape is not desensitization as we do it.

Does it work? Yes, and with realistic expectations. Reviews of noise fear find that counter-conditioning, relaxation training and desensitization with recordings all improve things (Riemer, 2023), and a survey of over a thousand owners found feeding or playing with the dog during fireworks was the one management strategy associated with improvement (Riemer, 2020). The only randomized trial of a home DS/CC program for veterinary fear called it mildly effective, and nearly half the owners did not complete it (Stellato et al., 2019). That last figure is the real lesson: the limiting factor is usually consistency, not the method, which is exactly the problem the app exists to solve.

In Push Drop Stick

The structure of every fear and handling plan: a ladder of exposures with the bottom rung set absurdly low on purpose.

Often confused with

Getting used to it. Repeated exposure at an intensity that frightens your dog is a gamble: sometimes the fear fades, and often it deepens instead. See Flooding and Sensitization.

The two above are one protocol, and they are almost always run together. Turn the trigger down until your dog is comfortable, then make it predict something excellent.

They need each other. Desensitization alone leaves you waiting for a feeling to change on its own, which is slow and unreliable. Counter-conditioning alone, without controlling the intensity, means pairing food with something your dog cannot cope with, and the food stops working because your dog stops eating.

That is what "under threshold" is for, mechanically. A dog who has stopped eating is past the point where the food can change what the trigger means, and is learning the old lesson instead. It is why food refusal is such useful information and never something to work around by finding better cheese.

One thing to watch for, because well-meant advice gets this backwards. In a lot of popular exercises, success is defined as your dog getting closer to the frightening thing: taking food from a stranger's hand, being coaxed forward, learning to greet. Food can pull a worried dog inside the distance they would otherwise keep, and then the food runs out and your dog is standing close to something frightening with no way out. That is where bites happen. In a series of 132 bites to the face, the thing that immediately preceded the bite was almost always a person closing distance: bending over the dog in 76% of cases, face-to-face proximity in 19%, mutual gazing in 5% (Rezac, Rezac and Slama, 2015). Distance should be free, never something your dog has to earn, and any closing of the gap should be your dog's idea.

Which points at a reframe worth sitting with: maybe your dog does not need to get closer to people. Maybe your dog needs to feel better about being approached. Vets approach. Children approach. Being approached is the skill no dog can opt out of, and it is the one worth building first. There is a plan for exactly that, called Being Approached.

In Push Drop Stick

Classical steps for the emotional work, Errorless steps where the setup should make failure nearly impossible. Both back off fast, on purpose.

Exposure that makes your dog more reactive rather than less: the opposite of getting used to something, and the outcome desensitization is designed to avoid.

Every exposure is teaching something. Done below threshold and paired with good things, it teaches that the trigger is fine. Done above threshold, the response grows, and on top of that your dog learns that the trigger is as bad as they suspected. Same activity, opposite outcome, and the only difference is intensity.

This is why "he needs more exposure to dogs" is such risky advice, and why a well-meant hour at a busy park can leave a worried dog worse than before. It is also why the answer to a bad session is a smaller version of the trigger and not a longer one.

If your dog's reactions are getting bigger, louder or starting at a greater distance than they used to, take it seriously. It is rarely stubbornness. Rule out pain with your vet first, because a dog who hurts reacts sooner and bigger. After that, the most likely cause is sensitization, and it means the current exposure level has been too high for a while.

Often confused with

Regression. Sensitization is not your dog forgetting the training. It is your dog learning something you did not intend.

Exposing a dog to a trigger at full intensity, with no escape, until they stop reacting.

The dog on a short leash held near the thing they fear. The dog put in a room with the visitor and the door closed. The puppy passed around a circle of strangers.

We never do this, and the reason is worth being precise about. Flooding can reduce fear: with escape blocked, a fear that never gets to run its course may fade, and a version of it is used in human therapy, with consent, by people who can stop whenever they choose. A dog cannot consent, and the fear often does not fade. What it does often produce is a dog who goes quiet, and that quiet is regularly reported as success. But behavior stopping is not the same as fear stopping. What frequently happens is that the dog runs out of options and shuts down, which looks far more like learned helplessness than comfort. And a shut-down dog has usually stopped giving the quiet warning signals, which makes them less predictable and less safe, not more.

Two things go wrong regularly. If the fear does not fade within the session, the exposure sensitizes instead, and your dog ends up more afraid than when they started. And if the session ends when your dog goes still, going still is what got them out. That is negative reinforcement, and it makes shutting down the thing that pays, whatever the dog still feels.

The alternative is the whole of this section. Turn the dial down until your dog can cope, pay them well, and let them leave whenever they like.

To be straight about the evidence: nobody has run a controlled trial of flooding dogs, and nobody should. The case against it rests on the wider experimental work on stress, escape and uncontrollable events, and on the consistent clinical experience of veterinary behavior specialists, who recommend the opposite approach of minimal restraint and letting an animal retreat (Riemer et al., 2021).

Often confused with

Desensitization. They look related and they are opposites. Desensitization keeps the intensity below the level that frightens your dog and keeps escape available. Flooding does neither.

What happens when nothing a dog tries makes any difference, so they stop trying.

The term comes from laboratory work with dogs in the 1960s (Seligman and Maier, and Overmier), in experiments no ethics committee would approve now: dogs who could not escape a shock later failed to escape one when escape became available. Applying it to a pet dog who has gone still under handling is an analogy rather than a diagnosis, and it is one behavior professionals make routinely because it fits what they see.

A dog who has learned that moving away does not work, that the growl gets punished, that the handling happens regardless, eventually stops all of it and goes still. From the outside this reads as calm, tolerant, well-behaved, finally getting used to it. From the inside nothing has improved. The dog has simply learned that behavior does not change outcomes. (The original researchers have since refined the explanation: going passive under something you cannot control may be the default, and what dogs learn is that some situations can be controlled. The practical picture is the same.)

How to tell it apart from a genuinely comfortable dog: a comfortable dog is loose and offering things. Soft body, easy mouth, moving, checking in, willing to eat, willing to start the activity again. A shut-down dog is still, closed, low, often refusing food, and doing nothing that they have not been made to do. The tell is what happens when you release the pressure. A comfortable dog carries on. A dog who was only holding still leaves as soon as they can, and a dog who has truly given up may not leave at all, even with the door open. Neither of them was comfortable.

It is also the point at which training has quietly become unsafe, because a dog who has given up on the quiet signals may go straight to the loud one when they finally cannot take any more.

This one warrants professional help rather than a plan. If your dog goes still and blank during handling or exposures, please get a certified positive-reinforcement professional or a veterinary behaviorist to see them.

Often confused with

Calm, tolerant, and stoic. All three are words owners are given for this by people who are pleased with the result.

First: you are not in trouble, and neither is your dog. Almost everyone who used one was told it was the responsible choice, usually by somebody confident. The way forward is not guilt, it is knowing which parts of ordinary training will not work yet and what to use instead.

Four things may have changed. All four are fixable.

Your dog may not offer anything. A dog who has been corrected for guessing learns that the safest move is to do nothing, which is an accurate read of the old contingencies rather than stubbornness. It matters here because two of our five step types need your dog to volunteer a behavior, so with this history those plans start dead: Capturing has nothing to catch and Shaping has nothing to build on. Start with Errorless and Classical steps instead, which are arranged so your dog cannot get it wrong, and let offering come back in its own time. Prompting suits these dogs particularly well for the same reason: a food lure hands your dog the answer so they never have to risk a guess, and you can fade it once the movement is flowing. It is often the gentlest way back in. When it does, and your dog starts throwing behavior at you including things you never asked for, that is the recovery, even though it looks like your dog getting naughtier. It is the best sign there is, because offering means your dog feels safe enough to risk being wrong.

The warning signals may be missing. Growling, stiffening and backing away are usually the first things corrected, because they are the parts that upset people. Punish the growl and you have removed the warning without touching the reason for it, so your dog looks fine right up until they are not. Assume the quiet signals are unreliable and work further below where you think you need to be: lower handling floors, shorter exposures, more distance than seems necessary. Stress Signals is worth rereading with that in mind.

Your marker may be poisoned. If a word, a click or a beep reliably arrived just before a correction, that sound is now a warning rather than a promise. A marker that predicted something unpleasant cannot do a marker's job, and you will see it as a dog who flinches or checks out at the exact moment you meant to tell them they were right. You can simply change it: a marker means whatever it has been paired with, so pick a fresh word your dog has no history with, pair it with food, and use that one everywhere. Consider dropping the no-reward marker altogether while you are at it. No no-reward marker is ever entirely neutral, and an "ah-ah" or a "nope" that was ever followed by a correction carries far more than "no treat this time". For a dog coming off corrections that history cannot be checked — silence and no treat carries the same information at no risk. See No Reward Marker.

The equipment and the room may carry meaning. The collar going on, the remote in your hand, a particular doorway, the tone of voice that used to come first: any of these can have become a predictor. If your dog deflates when you pick up a lead, that is worth knowing before you conclude that a plan is too hard.

And whatever you work on, escape stays available and is never earned. Your dog must be able to leave, and leaving must never cost them anything. A dog who has to work, wait or hold still in order to end something is back in the old arrangement, whatever is in your hand. See Start Button (Consent and Opt-Out).

Honesty about all of this: these are the patterns behavior professionals see and agree on, not measured findings from a study of dogs who came off shock collars. That research largely does not exist, because the experiment that would produce it is one nobody should run. What is documented is the cost while the tools are in use, and that is in Positive Punishment and Gear and Methods to Avoid.

If your dog has bitten, or has come apart suddenly after a long stretch of seeming fine, that is a case for a person in the room: a certified positive-reinforcement professional or a veterinary behaviorist. Suppression collapsing is one of the more dangerous moments in a dog's life, and it is not something to work through over a chat window.

In Push Drop Stick

Tell Coach the history. It changes which step types get chosen, how low the first step is set, and what your dog gets paid with. None of it is recorded as a judgement on you.

Often confused with

Well behaved. A dog who does very little is easy to live with and is often described as calm, respectful, or finally settled down. Learned Helplessness is the entry on telling those apart, and the question it asks is the right one here: what does your dog do when they are free to choose?

A behavior your dog offers to say "go ahead", used to begin each repetition of something they might otherwise not choose.

Your dog rests their chin in your hand, and the ear cleaning starts. They lift the chin, and it stops. That is a start button: a small, deliberate, easy-to-give signal that hands your dog the controls.

It sounds like a nicety. It is closer to the opposite. A large part of the fear around handling and grooming is not about the sensation, it is about the sensation happening unpredictably and without any way to stop it. Give your dog a way to start it and a way to end it, and the same procedure becomes tolerable, then boring, then something plenty of dogs actively volunteer for.

Two rules make it real. The stop signal has to work every single time, immediately, or it is not a start button, it is decoration, and your dog will stop bothering to use it. And ending it must cost your dog nothing: lifting the chin is enough, every time. Stopping when the chin lifts is negative reinforcement, used on purpose, and it stays the gentle kind because the exit is easy and always honored. What goes wrong in flooding is the opposite, an exit that is blocked or has to be paid for. The start button opens the door. It is not the toll.

There is a real argument against them, and it is worth knowing. A start button adds a trained behavior to a procedure whose whole point is emotional. Your dog is now doing something for food, and that can overshadow the association you were trying to build, which is the handling itself predicting good things. And with careful desensitization at an intensity low enough to be easy, you should not need a button at all, because a dog who is free to walk away is already voting with their feet.

The first has a mechanical answer: pay the handling, not the button. Keep the food glued to the ear touch, the paw hold, the brush stroke, and never to the chin rest itself. Food that arrives while the chin is down will still strengthen the chin rest a little, because it follows it, so watch for the one way that goes wrong: a dog working hard to keep their chin down while uncomfortable. Short reps and an exit that stays cheap are what prevent it. Then the handling stays the clearest predictor of good things, the classical association keeps building cleanly, and the button does one narrow job, which is deciding whether the rep happens at all. It is a gate, not the behavior you are paying for. The Start Button plans in this library are written that way on purpose, and so is the resetting between reps: the point is your dog choosing go ahead before each small piece, not holding still for a long one.

The second point is limited by two things, and between them they cover a great many household dogs. Desensitization done well is slow, precise work that depends on reading a dog accurately, and most owners are still learning that skill; a start button gives them something observable to score in the meantime. And plenty of dogs no longer vote with their feet. A dog whose attempts to leave were blocked, hurried through, or told off has learned that leaving does not work, so they hold still instead, and that stillness gets read as tolerance. For a stoic, tolerant dog the honest task is to hand the "no" back explicitly, because their feet have gone quiet.

The objection left standing is time. Working through a start button plan first is extra training, and in the short run it is less efficient than simply getting on with the handling. That is true and worth saying. Three things earn it back. The habit transfers, so one button serves every grooming and handling plan that follows. The deadline is usually softer than it feels, since nails need doing every few weeks rather than today. And the expensive outcome is a dog who starts leaving the room when the mat comes out, which costs far more time than it ever saved. If you genuinely cannot spare it this week, the honest options are a vet or groomer who handles well, or letting it wait. Not pushing a frightened dog through it faster.

So if your dog genuinely leaves when they have had enough and comes back on their own, use desensitization and skip the button. If your dog endures, start here.

Consent and start-button procedures have not been put through controlled trials in dogs, so this is professional consensus resting on a well-established principle: control and predictability reduce the impact of an unpleasant event. It matches what veterinary behavior guidance recommends for handling, which is minimal restraint, no cornering, and letting the animal move away (Riemer et al., 2021).

In Push Drop Stick

The Start Button plans in Grooming & Handling, and the reason the app's Errorless steps disable the fail button until the timer ends: your dog gets to be right by default.

Often confused with

Waiting for permission. A start button is not obedience, it is your dog choosing to begin. If your dog never presses it, that is a real answer and the setup needs to get easier.