Shock Collars Don’t Teach. Here’s What the Research Says
Twenty years of studies, one pattern: the zap costs your dog more than it buys you.
On this page
Shock collar sellers love the word “results”. Fair enough. Let’s look at the results.
Researchers have been putting e-collars under the microscope for more than twenty years. They’ve filmed dogs, swabbed saliva for stress hormones, counted how many times a trainer had to repeat “come”, and tested whether dogs expect good or bad things after training. The findings aren’t subtle, and they all point the same way.
Here’s what the studies actually say, with the numbers.
The Short Version
- Dogs react to shocks with behaviour that researchers read as stress, fear and pain.
- Those stress signals show up outside the moment of the shock, and even when no shock is given.
- In head-to-head recall training, reward-based trainers got results that were at least as good, and in the 2020 study, better.
- Veterinary behaviour bodies in the US and Canada now say aversive tools aren’t needed for training at all.
No study in this list found the shock collar to be the more effective option. The best it managed was a tie, and the tie came with extra welfare costs attached.
2004: The Dogs Remembered Who Held the Remote
The classic study comes from researchers at the University of Utrecht in the Netherlands. Schilder and van der Borg (2004) watched German shepherd dogs in guard dog training.
First, they recorded 32 dogs reacting to 107 shocks. The dogs lowered their body posture, gave high-pitched yelps and squeals, tried to avoid, redirected aggression and flicked their tongues. The researchers wrote that these reactions “suggest stress or fear and pain”. Most of the reactions lasted only a fraction of a second, which is why a quick glance can make it look like nothing happened.
Then they compared 16 dogs that had been shocked in the recent past with 15 control dogs that had similar training but never received shocks. They only used sessions where no shocks were given. The shocked dogs still showed lower ear postures and more stress behaviours, on the training ground and during a walk in a park.
The control dogs weren’t living in a treat utopia, either. The authors note those dogs were “also trained in a fairly harsh way”. The shock group still came out worse.
Their conclusion: receiving shocks “is a painful experience to dogs”, and the shocked dogs had learned that their owner, or the owner’s commands, announced shocks, “even outside of the normal training context”.
A shock collar doesn’t teach “come”. It teaches “the park hurts sometimes”.
2014: Same Results, Extra Stress
Next came a pair of UK field studies from the same research team. The first, Cooper et al. (2014) in PLOS One, took 63 pet dogs referred for recall problems and split them into three groups of 21:
- Trained by industry-approved trainers with e-collars.
- Trained by the same trainers without e-collars.
- Trained by members of the Association of Pet Dog Trainers UK, without e-collars.
A small preliminary study of 9 dogs found negative behaviour changes when the shock was applied, and higher cortisol (a stress hormone) after stimulation. Those dogs had “generally experienced high intensity stimuli without pre-warning cues”.
In the main study, trainers used lower settings with a warning cue. The reactions were less dramatic. Even so, the e-collar dogs “spent significantly more time tense, yawned more often and engaged in less environmental interaction” than the reward-trained group.
The results? 92% of owners reported improvement, with no significant difference between groups. Owners in the e-collar group were less confident about carrying on with the training at home: 76.2% (16 of 21) said they were confident, compared with 100% (21 of 21) in the reward group.
The authors’ summary: “no consistent benefit to be gained from e-collar training but greater welfare concerns compared with positive reward based training.”
2020: The Recall Rematch
Recall is the number one reason owners reach for a remote collar. The AVSAB statement says so directly. So China, Mills and Cooper (2020) looked hard at efficacy.
Same setup: 63 dogs with off-lead problems like poor recall, three groups of 21, up to 150 minutes of training over 5 days. The e-collar group was trained by manufacturer-nominated trainers. They measured how many commands it took to get a response, and how fast the dog responded to “Come” and “Sit”.
The reward-focused group:
- got significantly better responses to both “Sit” and “Come” after a single instruction,
- responded faster than the e-collar group, and
- needed significantly fewer commands overall.
There was no significant difference in the proportion of commands disobeyed between groups. So the collar didn’t buy obedience either.
The authors wrote that the findings “refute the suggestion that training with an E-collar is either more efficient or results in less disobedience, even in the hands of experienced trainers”. Their bottom line: “there is no evidence to indicate that E-collar training is necessary, even for its most widely cited indication.”
That’s the recall argument, tested with trainers nominated by the collar makers, and it lost.
The Wider Evidence on Aversive Methods
Shock collars sit inside a bigger family of “aversive” methods: anything that works because the dog finds it unpleasant, like leash pops, prong collars and yelling.
In 2017, Gal Ziv’s review in the Journal of Veterinary Behavior examined 17 studies. It concluded that aversive methods (positive punishment and negative reinforcement) “can jeopardize both the physical and mental health of dogs”. Positive punishment can work, the review notes, but there was no evidence it works better than reward-based training, and some evidence the opposite is true. The review also flagged limits in the research, like small sample sizes.
Then came Vieira de Castro et al. (2020), also in PLOS One. They followed 92 companion dogs from seven training schools in Portugal:
- 42 dogs from three reward-based schools,
- 22 dogs from two schools using a low proportion of aversive methods, and
- 28 dogs from two schools using a high proportion of aversive methods.
The aversive schools used methods including leash jerks, yelling, pinch collars, e-collars and choke collars.
Dogs from the high-aversive schools showed more stress behaviours (like lip licking and yawning), spent more time tense and panted more during training. They had bigger post-training cortisol increases than the reward group. And in a separate test away from training, they were more “pessimistic”: slower to approach an ambiguous food bowl. The authors concluded that aversive methods, “especially if used in high proportions, compromise the welfare of companion dogs both within and outside the training context.”
What the Vets Say Now
The research has moved the professional bodies.
The American Veterinary Society of Animal Behavior (AVSAB) 2021 position statement says “there is no evidence that aversive training is necessary for dog training or behavior modification”. It recommends that “only reward-based training methods are used for all dog training, including the treatment of behavior problems”, and tells trainers to avoid tools that “involve pain (choke chains, prong collars, or electronic shock collars)”.
In Canada, the Canadian Veterinary Medical Association (CVMA) says reward-based methods “are highly recommended” and that “remote controlled shock collars are not considered a necessary method of training or behaviour modification.” It strongly discourages “choke, pinch, spray, prong or shock collars”.
What the Research Doesn’t Say
Being fair matters, so here are the limits.
- These studies are small. Sixty-three dogs is a solid field study, not a census.
- Some studies used working dogs (the 2004 guard dogs), and others used pet dogs. The pattern holds across both, but the settings differ.
- Cortisol is tricky. The AVSAB statement notes it rises with good excitement as well as distress, so researchers read it alongside behaviour. In the 2014 main study, salivary cortisol in the e-collar group wasn’t significantly different from the others.
- Owners in 2014 reported similar improvement across groups. The collar “worked” in the sense that behaviour changed. It just didn’t work better, and it came with extra stress.
None of that rescues the collar. When the best case is a draw with side effects, the smart choice is the one without the side effects.
What to Do Instead
You were sold a tool. Now you know better. Here’s the upgrade.
- Build a recall that doesn’t need a remote. Start with our step-by-step guide: A Recall That Works Without a Remote. A long line gives you safety while you train, with no zap required.
- Fix the pulling without pain. If the collar came out on walks too, go to Stop Leash Pulling Without a Prong.
- Reactive dog? Punishment can make it worse. Read Reactive Dog? Punishment Makes It Worse before you try anything else.
- Hiring help? Use How to Spot a Trainer Who Uses Pain to check for words like “stim”, “correction” and “balanced” before you pay.
- Gear swap. Our Pain-Free Training Kit covers the basics.
If your dog’s behaviour changed suddenly, or comes with other changes, get your vet to rule out pain first.
Treats, not zaps. The research has your back.
Your dumb questions
My dog only yelped the first time. Doesn’t that mean it’s fine now?
Studies found stress signals that outlast the zap itself. Dogs trained with shock showed more stress behaviour even on walks where no shock was used, and some linked the handler with the shocks.
Aren’t modern collars gentler than the ones in the old studies?
The 2014 UK study used lower settings with a warning cue, and dogs still spent more time tense than reward-trained dogs. The 2020 follow-up found the reward group got better results, too.
Is there any study where the shock collar clearly won?
Not in the sources below. The best case for the collar in these studies is a tie on owner-reported results, with extra welfare costs attached.
The receipts
- Training dogs with help of the shock collar: short and long term behavioural effects · Applied Animal Behaviour Science, 2004
- The Welfare Consequences and Efficacy of Training Pet Dogs with Remote Electronic Training Collars in Comparison to Reward Based Training · PLOS One, 2014
- Efficacy of Dog Training With and Without Remote Electronic Collars vs. a Focus on Positive Reinforcement · Frontiers in Veterinary Science, 2020
- The effects of using aversive training methods in dogs—A review · Journal of Veterinary Behavior, 2017
- Does training method matter? Evidence for the negative impact of aversive-based methods on companion dog welfare · PLOS One, 2020
- Position Statement on Humane Dog Training · American Veterinary Society of Animal Behavior (AVSAB), 2021
- Humane Training of Dogs · Canadian Veterinary Medical Association (CVMA), 2021
Keep reading