How Do You Discipline a Puppy? Start With the Setup, Then the Timing

Discipline a puppy by removing the chance to rehearse the unwanted behaviour, then reinforcing a replacement within about one second. The American Veterinary Society of Animal Behavior set this out in its August 2021 position statement on humane dog training: common problems including jumping, barking and house-training "can be managed by arranging the environment appropriately and reinforcing desirable responses," while aversive methods "should not be used in canine training or for the treatment of behavioral disorders."

That one-second figure is measured. In trials reported in Clare Browne's 2015 doctoral thesis at the University of Waikato, 60% of dogs learned a novel task when food followed the correct response immediately. With the food one second late, 25% learned it. A marker sound given immediately, food still a second behind, recovered part of the loss at 40%. Browne's field observations at obedience clubs found owners delivering food an average of 0.98 seconds late, and dogs ignoring 44.20% of commands.

What "discipline" means once punishment is off the table

The word survives from an older model. In current veterinary guidance it names a procedure, not a consequence. The 2015 AAHA Canine and Feline Behavior Management Guidelines, from a task force chaired by Dr Karen Overall, define punishment as any consequence that makes a behaviour less likely in future, then build every recommendation around reinforcement instead.

That leaves a method any owner can run without equipment:

1. Record the behaviour for a week. Time, room, and what happened in the thirty seconds before. Counter-surfing at 5pm and at 8am are different problems.
2. Remove the antecedent. A bin behind a door cannot be raided. A puppy on a house line cannot reach the sofa arm.
3. Name the replacement. "Stop chewing" is not a behaviour. "Go to the mat" is.
4. Reinforce inside one second. Food, praise or a toy, while the puppy is still doing it.
5. Reread the record after seven days. If frequency has not moved, the antecedent was misidentified, not the reward.

The Generation Pup cohort at Dogs Trust reported in the Journal of Veterinary Behavior that the three most common problem behaviours in UK dogs at twelve months were pulling on the lead, jumping up at people, and poor recall. Each has an antecedent that can be removed before training begins.

Punishment against management: what the comparisons actually found

Several research groups have compared the two approaches and a mixed third.

| Approach | Owner-rated obedience | Welfare markers | Primary source | |---|---|---|---| | Rewards only | Highest of the three | Baseline stress behaviours; more owner-directed gaze | Hiby et al. (2004), University of Bristol, 364 owners | | Mixed rewards and aversives | Between the two | More stress behaviours and panting than reward-trained dogs; no better than aversive-only on cortisol | Vieira de Castro et al. (2020), PLOS ONE, 92 dogs, seven Portuguese schools | | Aversive-based correction | Lowest of the three | Higher post-training cortisol rise; more "pessimistic" on a cognitive bias task | Vieira de Castro et al. (2020); AVSAB summary of Hiby |

Hiby's survey also recorded what owners do: 66% used vocal punishment, 12% physical punishment, 60% praise and 51% food rewards. Obedience ratings tracked tasks trained with rewards (P < 0.01) and not punishment (P = 0.5); problem-behaviour counts tracked tasks trained with punishment (P < 0.001).

The sharpest figures on physical correction come from Meghan Herron, Frances Shofer and Ilana Reisner, who surveyed 140 owners at the behaviour service of the University of Pennsylvania's Ryan Veterinary Hospital. Hitting or kicking the dog produced an aggressive response in 43% of dogs it was tried on, growling at the dog in 41%, forcing an object from the mouth in 39%, the alpha roll in 31%, staring the dog down in 30%, the dominance down in 29%, and grabbing the jowls and shaking in 26%.

"Our study demonstrated that many confrontational training methods, whether staring down dogs, striking them or intimidating them with physical manipulation does little to correct improper behavior and can elicit aggressive responses," said Herron, a veterinarian in Penn Vet's Department of Clinical Studies and the paper's lead author.

AVSAB names the limit of this evidence rather than glossing it: survey work "cannot differentiate between causation and correlation." Its counter-argument is that if aversive methods eliminated problem behaviours, surveys would show fewer problems where they are used most. They show the opposite.

One controlled trial removes the survey problem. Lucy China, Daniel Mills and Jonathan Cooper at the University of Lincoln allocated 63 dogs with recall problems to three groups of 21, each given up to 150 minutes of training over five days. Latency to "come" was 1.13 seconds for the reward-based group against 1.35 for the electronic collar group, and 1.36 against 1.67 for "sit." Cues disobeyed did not differ significantly.

Do dogs connect the telling-off to the crime?

Alexandra Horowitz filmed 14 dogs at Barnard College, varying whether the dog had eaten a forbidden treat and whether the owner knew. The behaviours owners identify as the "guilty look" appeared when owners scolded, at the same rate whether or not the dog had offended.

Memory is not the limiting factor. Claudia Fugazza, Ákos Pogány and Ádám Miklósi at Eötvös Loránd University tested 17 dogs on unannounced recall of human actions, and found accurate recall after one minute and after one hour, fading across the interval. Puppies remember events. What they do not reliably do is attach a delayed human reaction to the right one, which is why the RSPCA is blunt: "Never punish your dog if you find an accident after the event or as it's happening."

Biting: telling ordinary mouthing from fear or pain

Mindy Waite, Mike Harman and Tiffany Kodak surveyed owners of 833 US dogs, defining mouthing as the dog's teeth or inner lips contacting human skin or clothing. Rates were highest under one year. The survey also killed a persistent piece of advice: dogs that played tug were no more likely to bite than dogs that did not.

Frequency is stubborn. A 2025 Frontiers in Behavioral Neuroscience trial by Sonowal, Hall and Stellato tracked 29 puppies aged four to ten months on a six-point scale topping out at "more than three times a day," and found no significant change across a week of intervention. Chewing damage was mostly cosmetic: 65.5% of owners reported minor damage to off-limit items, 13.8% major.

What separates ordinary mouthing from trouble is the rest of the body. AVSAB lists the stress signals recorded in observational studies: tense body, lowered posture, lip licking, tail lowering, a lifted front leg, panting, yawning and yelping. A puppy mouthing in play is loose and comes back for more. One that freezes, lowers itself, licks its lips and then snaps needs a veterinary appointment before a training plan; pain presents this way.

Trigger distance has no validated figure; any article quoting one should be read carefully. A 2026 Applied Animal Behaviour Science analysis structured canine reactivity by behaviour type rather than distance, into oral attack, frustration and posturing. Practitioner sources cite three to fifteen metres with no trials behind them. The usable test is behavioural: where the puppy still takes food and turns to its name, learning is possible. Where it stops eating, it has stopped learning.

House-training, sleep, and two numbers the sources fight over

The RSPCA advises taking a puppy out "every 45 minutes to an hour, depending on the puppy's age," plus after waking, meals, play, excitement, departures and arrivals. RSPCA Victoria puts the interval at one to two hours. The rule repeated across US veterinary and trainer writing is one hour per month of age, which for a nine-week puppy gives roughly two. These do not reconcile. The RSPCA figure is the one to start from: an accident rehearses the behaviour, and a spare trip outside costs nothing.

Sleep is the wider gap. The American Kennel Club, quoting chief veterinary officer Dr Jerry Klein, states that puppies sleep 18 to 20 hours a day. Rachel Kinsman and colleagues measured it for Generation Pup in Animals in 2020: owner reports for 2,332 puppies at 16 weeks gave a mean of 11.2 hours per 24, of which 7 were at night. At 12 months, 10.8 and 7.3.

The AKC figure describes a requirement, the Generation Pup figure what owners observe, and owners do not watch sleeping puppies all day. A follow-up analysis of the same cohort found dogs were less likely to develop separation-related behaviours when, at 16 weeks or younger, they were shut in a crate or room overnight and slept at least nine hours. That behaviour was reported in 46.9% of the cohort at six months.

Session length, treat calories, and the eight-month wall

Short and infrequent beats long and daily. Helle Demant and colleagues at the University of Copenhagen ran 18 sessions of six trials across four groups and found dogs trained once or twice weekly acquired the task significantly better than dogs trained daily, and one session a day beat three back to back. Retention four weeks later was unaffected.

Reinforcing at that rate has a calorie cost. The Nutrition Support Service at the UC Davis Veterinary Medical Teaching Hospital sets the ceiling plainly: treats and additional food should not exceed 10% of daily caloric intake, so a puppy eating 100 calories a day gets no more than 10 from training, out of the meal rather than on top of it. Growth is tracked against WALTHAM puppy growth charts and the nine-point body condition score developed by Dottie Laflamme and adopted by the WSAVA, on which four to five is the target.

Then there is month eight. Lucy Asher, Gary England, Rebecca Sommerville and Naomi Harvey followed 69 Labradors, Golden Retrievers and crosses through puberty and reported in Biology Letters that at eight months the dogs were more than twice as likely to need repeated commands before responding to a known "sit" as at five months. The effect was specific to the primary caregiver and absent with a stranger. The people who lived with the dogs rated trainability lowest at eight months; the trainers working with them rated it higher during adolescence than before.

"It's very important that owners don't punish their dogs for disobedience or start to pull away from them emotionally at this time," said Dr Lucy Asher, Senior Lecturer in Precision Animal Science at Newcastle University and the study's lead author. Adolescence is also when dogs most often arrive at rehoming centres.

Frequently asked questions

How do I discipline a puppy who will not listen?

Check the age before changing method. Asher et al. (2020) found puppies at eight months were more than twice as likely to need repeated commands as at five months, and only from their own caregiver. Shorten the session, raise the reward value, stop repeating the cue.

What month is hardest with a puppy?

Month eight, on the only dataset that has measured it. Asher et al. recorded reduced responsiveness to a well-known "sit" cue at eight months compared with five in 69 dogs. Owners rated trainability lowest then; the professional trainers working with the same dogs rated it higher.

At what age can I discipline my puppy?

Reinforcement works from arrival, usually around eight weeks. AVSAB's puppy socialization statement identifies the first three months as the primary window and says puppies can join classes at seven to eight weeks, given one set of vaccines seven days beforehand. Punishment has no recommended starting age.

Do dogs remember when you discipline them?

Not in the way most owners assume. Horowitz (2009) filmed 14 dogs and found the "guilty look" appeared in response to scolding, whether or not the dog had taken the forbidden treat. Fugazza et al. (2016) showed dogs recall observed actions for at least an hour.

How do I discipline a puppy for biting?

Treat mouthing as normal development and manage the trigger. Waite et al. (2021) surveyed 833 dogs and found mouthing most frequent under one year, with no link to tug games. End the interaction the instant teeth touch skin, offer a chew, check the puppy has slept.

How do I handle a puppy peeing inside?

Clean it and change the schedule. The RSPCA advises trips outside every 45 minutes to an hour depending on age, plus after waking, meals, play and excitement. Never punish an accident found after the event; the puppy will not connect the two.