# Operant Conditioning in Dog Training: The Four Quadrants, the Evidence, and How to Train

> Operant conditioning in dog training: the four quadrants, the evidence on reward-based vs. aversive methods, clicker training, timing, and a recall protocol.

- Source: https://operantconditioning.com/dog-training/
- Author: Ryan Martinson (https://operantconditioning.com/about/)
- Publisher: Operant Conditioning Inc.
- Published: 2026-09-07 · Updated: 2026-09-10
- License: https://operantconditioning.com/terms/#copyright (quote with attribution and a link; free to reproduce for non-commercial teaching)

---

*Applications · Animal training*

Every dog trainer uses operant conditioning, whether they know the vocabulary or not. Here are the four quadrants with dog examples, what the research actually says about reward-based and aversive methods, how markers, shaping, timing, and schedules work, and a protocol for a recall you can trust.

> **Definition**
>
> **Operant conditioning in dog training** is the deliberate use of consequences to change what a dog does: behaviors that are followed by reinforcement (a treat, play, release to sniff, or relief from pressure) happen more often, and behaviors followed by punishment (an added aversive or a lost reward) happen less often. Trainers arrange the [antecedent](https://operantconditioning.com/abc-model/) — the cue and the setting — and the consequence, and the dog's behavior changes in between.
>
> A "reward" only counts as a reinforcer if the behavior actually increases, and a "correction" only counts as a punisher if the behavior actually decreases. The dog, not the trainer, decides what works.[1]

**In brief**

- The [four quadrants](https://operantconditioning.com/#the-four-quadrants-reinforcement-and-punishment) are not equal choices: R+ and P− need nothing unpleasant present, while R− and P+ depend on an aversive the trainer introduces.
- No controlled study has found aversive methods more effective than reward-based ones, and aversive methods are associated with stress and aggressive responses.
- A reward counts as a reinforcer only if the behavior actually increases; the dog, not the trainer, decides what works.

## The four quadrants in dog training

Every consequence a dog experiences can be sorted by two questions: was something *added* or *removed*, and did the behavior go *up* or *down*? Trainers usually abbreviate the results as R+, R−, P+, and P−.

| Quadrant | What happens | Dog training example | Effect and notes |
| --- | --- | --- | --- |
| R+ Positive reinforcement | Something the dog wants is added after the behavior | Dog sits → treat, or a thrown ball, or the door opens | Sitting increases. The foundation of modern, reward-based training. |
| R− Negative reinforcement | Something the dog dislikes is removed after the behavior | Steady leash pressure → dog steps toward the handler → pressure released | Moving toward the handler increases. Requires an aversive to be present first; used in traditional and some "balanced" training. |
| P+ Positive punishment | Something the dog dislikes is added after the behavior | Dog pulls → leash pop; dog barks → spray bottle | Pulling or barking decreases, if it works at all. Associated with stress and aggressive responses (see the evidence below). |
| P− Negative punishment | Something the dog wants is removed after the behavior | Dog jumps up → person turns away; dog mouths hand → play ends for 30 seconds | Jumping or mouthing decreases. The mildest way to reduce behavior, and it pairs naturally with R+ for an alternative. |

Two things follow from the table. First, "positive" and "negative" describe adding and removing, not kind and cruel: the spray bottle is *positive* punishment. Second, the quadrants are not four equal choices. R+ and P− require nothing unpleasant to be present, while R− and P+ both depend on an aversive stimulus the trainer has to introduce, which brings side effects with it. That asymmetry is why most evidence-based trainers work almost entirely in the top-left cell of the [quadrant grid](https://operantconditioning.com/#the-four-quadrants-reinforcement-and-punishment) (R+) and reach for P− when a behavior needs to shrink.

## What the evidence says about reward-based vs. aversive dog training

Trainers argue about methods constantly; the research is smaller than the argument, but it points consistently in one direction.

### Owner surveys

Hiby, Rooney, and Bradshaw surveyed 364 dog owners about the methods they used for common tasks and the behavior of their dogs. Owners who relied on reward-based methods reported higher obedience, and the use of punishment-based methods was associated with a higher number of problem behaviors.[2] Herron, Shofer, and Reisner surveyed 140 owners whose dogs had been referred to a veterinary behavior clinic. For several confrontational techniques — hitting or kicking the dog, growling at it, the "alpha roll," staring it down — roughly a quarter or more of owners reported that the dog had responded with aggression. Reward-based techniques rarely produced aggressive responses.[3]

### Reviews and welfare studies

Ziv's 2017 review of the literature concluded that aversive methods carry welfare risks and that there is no evidence they are more effective than reward-based methods.[4] Vieira de Castro and colleagues then went beyond questionnaires: they filmed dogs from training schools that used reward-based methods and schools that used aversive methods, sampled salivary cortisol, and ran a cognitive bias test. Dogs from aversive-based schools showed more stress-related behaviors and body postures during training, higher post-training cortisol, and in the cognitive bias test they approached an ambiguous bowl more slowly — the "pessimistic" pattern seen in animals in poorer welfare states.[5]

### Experimental comparisons

The strongest single piece of efficacy evidence is an experiment rather than a survey. China, Mills, and Cooper assigned pet dogs with known off-lead problems to training with remote electronic collars by industry-approved trainers, to the same trainers without collars, or to reward-based trainers. The reward-based group responded to "sit" and "come" more reliably and more quickly; the e-collar added no measurable benefit.[6]

> **The honest caveats**
>
> Most of this evidence is correlational. Owners who choose punishment may already have more difficult dogs, and owners who choose reward-based methods may differ in other ways too. Surveys rely on self-report; clinic samples are not typical dogs; and school-based comparisons cannot fully separate the method from the trainer. What can be said is this: no controlled study has found aversive methods to be *more* effective, several have found reward-based methods equally or more effective, and the welfare and aggression findings point the same way across designs. That is enough for professional bodies to recommend reward-based training as the default.[7] [More on aversives and their side effects ›](https://operantconditioning.com/positive-punishment/#aversives-in-dog-training-what-the-evidence-shows)

## The dominance myth

A great deal of popular dog training rests on the idea that dogs are trying to become the "alpha" of the household and must be shown their place. The idea traces to studies of unrelated wolves confined together in captivity in the mid-twentieth century, which showed constant fighting for rank. L. David Mech's 1970 book helped spread the "alpha wolf" concept; his own later fieldwork on wild wolves showed that a pack is simply a family — a breeding pair and their offspring — and that the parents lead the way parents do, without ritualized dominance contests. Mech published the correction in 1999, has spent years asking people to drop the term, and has said he asked his publisher to stop printing the 1970 book.[8][9]

Domestic dogs are not wolves in any case, and studies of free-ranging dogs and of dog–human interactions find little support for the notion that problem behavior is a bid for status.[10] The dog that pulls on the leash is not staging a coup; pulling has simply been reinforced by getting where it wants to go. The American Veterinary Society of Animal Behavior's position statement on dominance theory recommends against confrontational "dominance" techniques, and its 2021 statement on humane dog training recommends reward-based methods and advises against aversive tools.[7][11]

## Marker and clicker training

A treat that arrives three seconds after a sit reinforces whatever the dog was doing three seconds after it sat — usually standing up and sniffing your hand. A **marker** solves the timing problem. A clicker (or a short word like "yes") is paired with food until the sound itself becomes a **conditioned reinforcer**: through classical conditioning it comes to predict the treat, and through operant conditioning it can then strengthen whatever it follows.[12] Trainers call it a **bridge** because it spans the gap between the behavior and the primary reinforcer. [How classical and operant conditioning combine ›](https://operantconditioning.com/operant-vs-classical-conditioning/)

The technique is older than most people think. Keller and Marian Breland, two of Skinner's early students, left academia in the 1940s to train animals commercially and used a hand-held clicker as a conditioned reinforcer across dozens of species.[13] Skinner himself described the method for a general audience in 1951, explaining how to train a dog with a conditioned reinforcer and successive approximations.[14] Marine-mammal trainers adopted a whistle as their bridge, and Karen Pryor, a former dolphin trainer, brought the approach to dog owners through *Don't Shoot the Dog* and the clicker-training movement that followed.[15]

### Charging the clicker

Click, then treat; click, then treat — twenty or thirty times over a couple of short sessions, with the click always *preceding* the treat by about a second. When the dog's head whips toward you at the sound, the marker is charged. From then on the rule is simple: every click earns a treat, and the click marks the exact instant the behavior you want occurs. The treat can follow a moment later; the click has already done the teaching.

## Shaping, luring, and capturing

There are three ways to get a behavior to happen so you can reinforce it.

- **Capturing.** Wait for the dog to do the behavior on its own — lie down, make eye contact — and mark it. Slow for rare behaviors, excellent for common ones, and it produces behavior the dog "owns."
- **Luring.** Use a treat in the hand to guide the dog into position: raise it over the nose and the rear drops into a sit. Fast, but the lure must be faded within a few repetitions or the dog learns that the behavior happens only when food is visible.
- **Shaping.** Reinforce successive approximations — first a glance at the mat, then a step toward it, then a paw on it, then lying on it. Shaping builds behaviors that can't be lured and teaches the dog to experiment.[1] [How shaping works, step by step ›](https://operantconditioning.com/shaping/)

## What makes reinforcement work

### Timing: the one-second window

Laboratory work on delayed reinforcement shows that a consequence loses much of its power within a few seconds of the behavior, and a delayed reinforcer tends to strengthen whatever happened just before *it* rather than the behavior you intended.[16] In practice: mark within about a second, and deliver the treat where you want the dog to be (feed a "down" on the floor between the paws, feed heel position at your left knee).

### Rate of reinforcement

Early in learning, aim for many reinforced repetitions per minute — ten or fifteen is not unusual in a good session. A high rate keeps the dog engaged and out-competes distractions.

### Reinforcer value

Reinforcers are not interchangeable. Most dogs rank kibble below cheese, cheese below roast chicken, and, for some, a tug toy above all of it. Match the reinforcer to the difficulty: kibble for a sit in the kitchen, chicken for a recall past a squirrel. And remember motivating operations: a dog trained right after dinner is a dog for whom food has stopped being a reinforcer.

### Life rewards and the Premack principle

Anything the dog wants to do can reinforce something you want it to do — David Premack's [principle](https://operantconditioning.com/premack-principle/) that a higher-probability behavior reinforces a lower-probability one.[17] Sit, and the door opens. Look at me, and you are released to sniff that fascinating hydrant. Come when called, and you get sent back to play.

## Schedules of reinforcement in dog training

While the dog is learning, reinforce *every* correct response — a continuous schedule, which produces the fastest acquisition. Once the behavior is reliable, thin to a **variable-ratio** schedule: reinforce most responses, then some, in an unpredictable pattern, while keeping the best responses on a "jackpot."[18] This is the step most owners skip, and skipping it in either direction causes trouble. Never thinning produces a dog that works only when food is visible. Thinning too fast, or reinforcing so rarely that the behavior stops paying, produces a dog that stops offering the behavior.

> **Why occasional treats make behavior stronger, not weaker**
>
> Owners worry that "not rewarding every time" will erode the behavior. The opposite is true. Behavior maintained on an intermittent schedule is *more* resistant to extinction than behavior reinforced every time — the partial-reinforcement extinction effect that Ferster and Skinner documented in thousands of hours of cumulative records. A dog whose sit is reinforced unpredictably keeps sitting through long stretches without a treat, because long stretches without a treat are exactly what it has learned to expect.[18] The same principle works against you when counter-surfing pays off once a month. [Run the schedule simulator ›](https://operantconditioning.com/schedules-of-reinforcement/)

## Stimulus control, cues, and generalization

### Add the cue after the behavior is reliable

A cue is a [discriminative stimulus](https://operantconditioning.com/glossary/#discriminative-stimulus): a signal that the behavior will now be reinforced. Saying "sit" to a dog that does not yet sit reliably teaches it nothing except that "sit" is background noise. The cleaner sequence is: get the behavior (by capturing, luring, or shaping), reinforce it until the dog offers it readily, and then say the cue *just as* the dog starts to do it. Within a few dozen pairings the word predicts the behavior and the reinforcement, and the dog starts responding to the word alone. A behavior is under stimulus control when it happens promptly on cue, does not happen without the cue, and does not happen to other cues.[1]

### Dogs don't generalize well — so train everywhere

A dog that sits perfectly in the kitchen has learned "sit, in the kitchen, facing my owner, with the treat pouch on." Take it to the park and the behavior can vanish, not out of stubbornness but because none of the antecedent conditions match. Behavior analysts learned long ago that generalization has to be programmed rather than hoped for: train in many places, with many people, at many distances and levels of distraction, and reinforce across all of them.[19] Trainers summarize this as the three D's — distance, duration, distraction — and the rule that you raise only one at a time.

## Extinction and extinction bursts

When a behavior that used to be reinforced stops working, it does not vanish quietly. It gets worse first. The dog whose jumping has always earned a hand, a voice, or a push-off will, when the attention stops, jump higher, faster, and with more mouth — an **extinction burst** — before the behavior declines.[20] The same happens with demand barking: ignore it and the dog will bark louder for a while. Two rules make extinction workable. Be completely consistent, because reinforcing the louder bark teaches the dog that louder is the new price. And always reinforce an alternative — four paws on the floor, a quiet sit — so the dog has a behavior that *does* pay. Pure extinction without a replacement is slow and unkind. [Extinction, bursts, and spontaneous recovery in depth ›](https://operantconditioning.com/extinction/)

## Management: control the antecedent before you train

Every time a dog practices a behavior and it pays off, the behavior gets stronger. So before any training plan, arrange the environment so the unwanted behavior can't be rehearsed: baby gates at the front door, food off the counters, a leash on the dog when guests arrive, blinds down for the window barker, a long line for the dog whose recall isn't ready. This is antecedent control, the "A" of the [A-B-C model](https://operantconditioning.com/abc-model/), and it is what lets the reinforcement-based plan win: the old behavior stops being reinforced in the background while you teach the new one.

## How to teach a reliable recall, step by step

Recall is the behavior that keeps dogs alive, and it is the one most often ruined by good intentions. The protocol below uses nothing but positive reinforcement, management, and schedules.

1. **Choose a fresh cue.** If "come" has been shouted at a dog that ignored it, or used to end walks, it is contaminated. Pick a new word — "here," a whistle — and protect it: never use it when you can't make it pay.
2. **Charge the cue.** Indoors, with the dog beside you, say the cue once and immediately feed three to five tiny pieces of something excellent, one after another. Ten repetitions, twice a day, for three or four days. The cue now predicts a jackpot before the dog has done anything.
3. **Add the behavior at short range.** Say the cue when the dog is a few feet away and likely to come anyway. When it reaches you, take the collar gently, *then* feed. The collar touch becomes part of the reinforced chain, so a hand reaching for the collar never becomes a signal to dodge.
4. **Build distance and distraction separately.** Increase distance indoors, then move to a quiet yard at short distance, then a long line in a park. Raise one variable at a time. The long line is management: it guarantees the cue is never disobeyed successfully.
5. **Send the dog back to the fun.** Most recalls in practice should end with "go play." If coming when called usually ends the walk, the recall is being punished by the loss of freedom — negative punishment of the very behavior you want. Use the Premack principle so that coming is the price of more freedom, not the end of it.
6. **Thin the schedule, keep the jackpots.** Once the recall is reliable, reinforce with food unpredictably, keep the surprise jackpots for fast responses through distractions, and lean on life rewards. Never let the schedule thin to zero.
7. **Never punish a recall.** If the dog comes slowly, reinforce anyway — a slow recall is still a recall, and scolding it teaches the dog that arriving is dangerous. If the dog doesn't come, go and get it calmly, then make the next repetition easier.
8. **Maintain for life.** A few surprise recalls on every walk, always paid, keep the behavior strong.

## Common mistakes in operant dog training

- **Repeating the cue.** "Sit. Sit! SIT!" teaches the dog that the cue is "sit-sit-SIT." Say it once; if nothing happens, make the situation easier and try again.
- **Poisoning the cue.** A cue that is sometimes followed by reinforcement and sometimes by a correction becomes ambiguous — Karen Pryor's term is a *poisoned cue* — and dogs respond to it slowly and with signs of stress. Keep each cue attached to one kind of consequence.
- **Punishing the recall by ending the fun.** Calling the dog only to leave the park, get a bath, or be crated trains the dog to keep its distance.
- **Treat dependence from never thinning.** If every sit for two years has produced a visible treat, the treat has become part of the cue. Fade the lure early and thin the schedule once the behavior is reliable.
- **Bribing instead of reinforcing.** Showing the treat before the behavior is a lure or a bribe. Reinforcement comes *after*.
- **Reinforcing the wrong moment.** A treat handed to a dog that sat and then stood reinforces standing. Mark the sit; feed in the sit.
- **Training when the dog is over threshold.** A dog that is frantic about another dog across the street cannot learn. Add distance until it can eat and think, then train.

## Common problem behaviors: what maintains them, and a reinforcement-based plan

The first question is never "how do I stop this?" but "what is this behavior getting?" Once you can name the reinforcer, the plan almost writes itself: manage so the old reinforcer stops arriving, and reinforce a behavior that can replace it.

| Behavior | Likely maintaining reinforcer | Reinforcement-based plan |
| --- | --- | --- |
| Jumping on people | Attention: eye contact, voices, hands — even a push-off is contact R+ | Manage with a leash or gate at the door. All attention stops the instant paws leave the floor P−; four-on-the-floor or a sit earns the greeting, generously. Recruit guests, and expect a burst. |
| Pulling on the leash | Forward progress toward smells and dogs R+ | A tight leash stops all forward motion (pulling no longer works); a loose leash makes the walk go on, plus frequent treats at your side. A front-clip harness for management while the new behavior builds. |
| Barking at passersby from the window | The passerby always leaves R−, plus the arousal itself | Block the view (film on the glass, closed blinds). Teach and heavily reinforce "go to your mat" when someone passes; reinforce quiet glances at the window. |
| Demand barking | Food, play, the door, or attention delivered to stop the noise R+ | Barking pays nothing, ever. Teach a quiet alternative request (a sit, a nose-touch) and pay it fast and often. Consistency from everyone in the house, and expect the burst. |
| Counter-surfing | Food, on an intermittent schedule — the most durable kind VR | Management is non-negotiable: clear counters, closed kitchen. Reinforce lying on a mat in the kitchen while you cook. One sandwich a month will maintain the behavior indefinitely. |
| Begging at the table | Scraps, from at least one family member, occasionally VR | Nobody feeds from the table, no exceptions. Give the dog a stuffed food toy on its bed during meals so lying there becomes the behavior that pays. |
| Lunging and barking at dogs on leash | Distance: the other dog goes away, or the owner retreats R−, usually driven by fear or frustration | Work at a distance where the dog can eat and think. Pair the appearance of other dogs with excellent food (counterconditioning), and reinforce looking at the dog and back at you. Do this with a qualified reward-based trainer. |

> **Aggression, fear, and resource guarding need a professional**
>
> Growling, snapping, biting, guarding food or objects, and severe fear are not obedience problems, and punishing them tends to suppress the warning while leaving the emotion intact — the dog that no longer growls may go straight to biting. Seek a certified reward-based trainer or, for aggression and anxiety, a veterinary behaviorist, who can also rule out pain and medical causes.

## Key takeaways

- "Positive" and "negative" mean added and removed, not kind and cruel. R+ and P− require nothing unpleasant to be present; R− and P+ both depend on an aversive the trainer introduces, which is why evidence-based trainers work almost entirely in R+ and reach for P− when a behavior needs to shrink.
- The research is smaller than the argument but points one way: no controlled study has found aversive methods more effective, reward-based methods were equally or more effective, and aversive methods are associated with stress, higher cortisol, and aggressive responses. The dominance idea rests on captive wolves and was retracted by the researcher who spread it.
- A marker is a conditioned reinforcer that bridges the gap between the behavior and the treat; without it, a treat three seconds late reinforces whatever the dog was doing three seconds later. Mark within about a second and feed where you want the dog to be.
- Reinforce every correct response while the dog is learning, then thin to a variable-ratio schedule and keep the jackpots. Intermittent reinforcement makes behavior more resistant to extinction, which is why a sit survives long stretches without treats and why one sandwich a month maintains counter-surfing.
- Cues are discriminative stimuli added after the behavior is reliable, and dogs generalize poorly, so train everywhere and raise distance, duration, and distraction one at a time. Before training, manage the antecedent so the old behavior stops being rehearsed, and ask what the behavior is getting before asking how to stop it.

### Check yourself

**An owner calls her dog at the park, and when it arrives she clips on the leash and goes home. She never scolds it, yet over the weeks the recall gets slower. Why is the recall weakening?**

Coming when called reliably ends the dog's freedom, so the recall is being punished by the loss of play: negative punishment of the very behavior she wants. The fix is the Premack principle in reverse of what she has been doing: most recalls should end with "go play," so that coming is the price of more freedom rather than the end of it.

**A trainer applies steady leash pressure and releases it the instant the dog steps toward her. A student calls this punishment, since leash pressure is unpleasant. Which quadrant is it, and what is the real concern?**

It is negative reinforcement: something the dog dislikes is removed after the behavior, and stepping toward the handler increases. "Negative" means removed, not bad. The real concern is that R− requires the aversive to be present first, which is where the welfare cost lies and why it sits on the same side of the asymmetry as P+.

**A dog jumps on every guest, and every guest pushes it off with a firm "No." Months later it still jumps. What is maintaining the behavior?**

Attention: eye contact, voices, and hands, and even a push-off is contact. The intended correction is functioning as positive reinforcement, which the behavior proves by continuing. The plan is management at the door, all attention stopping the instant paws leave the floor, a generous greeting for four-on-the-floor or a sit, and an extinction burst to be expected first.

**An owner worries that reinforcing sits only some of the time will erode the behavior, so she keeps treating every one. Is her worry justified?**

No; the opposite is true. Behavior maintained on an intermittent, variable-ratio schedule is more resistant to extinction than behavior reinforced every time, the partial-reinforcement extinction effect. Reinforcing every sit for years makes the visible treat part of the cue. The errors to avoid are thinning too fast, or letting the schedule thin to zero.

**Explain it to a friend.** Explain why a clicker works, without using the words "reinforcer," "conditioned," or "bridge."

## Frequently asked questions

**What are the four quadrants of dog training?**

Positive reinforcement (add something the dog wants — a treat for a sit), negative reinforcement (remove something unpleasant — leash pressure released when the dog moves), positive punishment (add something unpleasant — a leash pop), and negative punishment (remove something the dog wants — turning away when it jumps). "Positive" and "negative" mean added and removed, not good and bad.

**Is positive reinforcement dog training effective?**

Yes. Owner surveys associate reward-based methods with higher obedience and fewer problem behaviors, a controlled comparison found reward-based training more effective than electronic-collar training for recall and sit, and no controlled study has found aversive methods to be more effective. Reward-based methods also avoid the stress and aggression associated with confrontational techniques.

**Do I have to give my dog treats forever?**

No. Reinforce every correct response while the dog is learning, then shift to reinforcing unpredictably — a variable-ratio schedule — while replacing many treats with life rewards such as play, sniffing, and going through doors. Intermittent reinforcement makes behavior more durable, not less. What you should never do is let reinforcement stop entirely.

**Is negative reinforcement bad for dogs?**

Negative reinforcement is not punishment — it strengthens behavior by removing something unpleasant. But it requires the unpleasant thing to be present first, which is where the welfare cost lies. Mild, brief pressure released the instant the dog responds is used by many trainers; methods built on shock, prong collars, or sustained discomfort are associated with stress indicators and are discouraged by veterinary behavior organizations.

**Is the alpha or dominance theory of dog training true?**

No. The "alpha wolf" idea came from unrelated captive wolves forced to live together. Wild wolf packs are families led by the parents, as L. David Mech, who helped popularize the term, later showed and retracted. Dogs are not wolves, and studies of dog behavior do not support the idea that misbehavior is a bid for rank. Problem behavior is almost always a reinforcement problem, not a status problem.

**What is clicker training and how does it work?**

Clicker training uses a distinct sound that has been paired with food until it becomes a conditioned reinforcer. The click marks the precise instant the dog does the right thing and bridges the gap until the treat arrives. It solves the timing problem — you can click within a fraction of a second even if the treat takes longer — and it lets you shape complex behaviors in small steps.

**Why does my dog only listen at home?**

Because that is where the behavior was trained. Dogs learn cues together with the context — the room, the person, the pouch, the absence of distractions — and do not generalize well on their own. Retrain each behavior in new places, with new people, at greater distances and with more distraction, raising one variable at a time and reinforcing generously as you go.

**Should I punish my dog for growling?**

No. A growl is information — the dog is telling you it is uncomfortable. Punishing it can suppress the warning without changing the feeling, producing a dog that bites without growling first. Move the dog away from what is bothering it, and consult a certified reward-based trainer or veterinary behaviorist to address the underlying fear or guarding.

## References

1. Cooper, J. O., Heron, T. E., & Heward, W. L. (2020). *Applied Behavior Analysis* (3rd ed.). Pearson.
2. Hiby, E. F., Rooney, N. J., & Bradshaw, J. W. S. (2004). Dog training methods: Their use, effectiveness and interaction with behaviour and welfare. *Animal Welfare, 13*(1), 63–69.
3. 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.
4. Ziv, G. (2017). The effects of using aversive training methods in dogs — A review. *Journal of Veterinary Behavior, 19*, 50–60.
5. Vieira de Castro, A. C., Fuchs, D., Morello, G. M., Pastur, S., de Sousa, L., & Olsson, I. A. S. (2020). Does training method matter? Evidence for the negative impact of aversive-based methods on companion dog welfare. *PLoS ONE, 15*(12), e0225023.
6. China, L., Mills, D. S., & Cooper, J. J. (2020). Efficacy of dog training with and without remote electronic collars vs. a focus on positive reinforcement. *Frontiers in Veterinary Science, 7*, 508.
7. American Veterinary Society of Animal Behavior. (2021). *Position Statement on Humane Dog Training*. AVSAB.
8. Mech, L. D. (1999). Alpha status, dominance, and division of labor in wolf packs. *Canadian Journal of Zoology, 77*(8), 1196–1203.
9. Mech, L. D. (2008). Whatever happened to the term alpha wolf? *International Wolf, 18*(4), 4–8.
10. Bradshaw, J. W. S., Blackwell, E. J., & Casey, R. A. (2009). Dominance in domestic dogs — useful construct or bad habit? *Journal of Veterinary Behavior, 4*(3), 135–144.
11. American Veterinary Society of Animal Behavior. (2008). *Position Statement on the Use of Dominance Theory in Behavior Modification of Animals*. AVSAB.
12. Williams, B. A. (1994). Conditioned reinforcement: Experimental and theoretical issues. *The Behavior Analyst, 17*(2), 261–285.
13. Breland, K., & Breland, M. (1951). A field of applied animal psychology. *American Psychologist, 6*(6), 202–204.
14. Skinner, B. F. (1951). How to teach animals. *Scientific American, 185*(6), 26–29.
15. Pryor, K. (1984). *Don't Shoot the Dog! The New Art of Teaching and Training*. Simon & Schuster.
16. Lattal, K. A. (2010). Delayed reinforcement of operant behavior. *Journal of the Experimental Analysis of Behavior, 93*(1), 129–139.
17. Premack, D. (1959). Toward empirical behavior laws: I. Positive reinforcement. *Psychological Review, 66*(4), 219–233.
18. Ferster, C. B., & Skinner, B. F. (1957). *Schedules of Reinforcement*. Appleton-Century-Crofts.
19. Stokes, T. F., & Baer, D. M. (1977). An implicit technology of generalization. *Journal of Applied Behavior Analysis, 10*(2), 349–367.
20. Lerman, D. C., & Iwata, B. A. (1995). Prevalence of the extinction burst and its attenuation during treatment. *Journal of Applied Behavior Analysis, 28*(1), 93–94.


## About the author

Ryan holds a master's degree from UCLA, where he studied animal behavior in Daniel Blumstein's lab and was part of the university's Evolutionary Medicine Program, which applies findings from evolutionary biology and animal behavior to human health. He founded Operant Conditioning Inc. and built the Operant habit app (https://operantconditioning.com/app/). Every page here is written from the primary literature and cites it. How pages are checked: https://operantconditioning.com/about/#editorial-standards

## Related

- [Positive reinforcement](https://operantconditioning.com/positive-reinforcement/): The quadrant modern dog training is built on — timing, contingency, and reinforcer types.
- [Shaping](https://operantconditioning.com/shaping/): Successive approximations and chaining — how complex behaviors are built one step at a time.
- [Schedules of reinforcement](https://operantconditioning.com/schedules-of-reinforcement/): When to thin the treats, and why intermittent reinforcement makes behavior last.