Do ultrasonic bark devices work, and are they safe?

Ultrasonic bark devices interrupt some dogs the first few times and most dogs stop responding within days to weeks. In the one controlled trial of commercial units, no device repelled all dogs at any setting. VCA describes them as usually neither unpleasant enough nor consistent enough to be a reliable deterrent, and veterinary bodies class them as aversive: they work, when they work, by being unpleasant.

What does the evidence say?

• Blackshaw et al., 1990: three ultrasonic units, 60 to 75 dB across 7 to 45 kHz, tested on 15 dogs. “No device repelled all dogs.” Some dogs showed aversion; others ignored them.
• Moxon et al., 2010: guide dogs exposed to ultrasonic deterrents noticed them and carried on working.
• No trial of a bark-activated ultrasonic box or collar appears in the veterinary behaviour journals. The efficacy percentages on sales pages don’t trace back to a publication.

Why does it stop working?

A sudden odd sound makes almost any animal pause. If nothing follows it, the dog learns it means nothing; that’s habituation, and owners typically report it within days. The AKC’s own advice on these tools is to switch the device off when you don’t need it, or the dog gets used to the tone. A smaller number of dogs go the other way and become more anxious about sounds in general, which the veterinary behaviourist Jennifer Summerfield describes as sensitization.

Do they work through walls or fences?

Poorly. High-frequency sound is directional and is absorbed and reflected by solid surfaces far more than low-frequency sound, which is why you hear a neighbour’s bass and not their treble. A unit aimed at a dog behind a wall or a wooden fence is mostly talking to the wall. And hearing itself isn’t on your side: dogs hear best around 8 kHz, and their sensitivity falls away steeply above 20 kHz.

Are they cruel? Are they safe?

They’re designed to be unpleasant, which makes them an aversive, and the American Veterinary Society of Animal Behavior advises reward-based methods only. The British Veterinary Association says it is concerned about the potential of sonic and ultrasonic devices to cause suffering, while acknowledging the research is thin. The bioethicist Jessica Pierce notes that manufacturers she contacted could not produce safety studies. Three practical problems follow:

• They fire at every dog in range, including the quiet one, puppies, and the neighbour’s pets.
• They misfire on other sounds, so the dog can’t work out what causes it.
• They do nothing about why he barks. A frightened dog is still frightened.

What about using one on a neighbour’s dog?

We’d advise against it. You can’t see the dog’s reaction, you may be adding stress to a dog that’s barking out of isolation, and the evidence says it probably won’t work anyway. Here’s what does.

What should I use instead?

Find the cause, manage the trigger, and teach a quiet cue with rewards. A sound becomes useful when it has a meaning you’ve taught: that’s the difference between a cue and a punisher, and it’s how DogHush uses sound, replaying your own trained word when your dog barks. For the collar versions, see bark collar alternatives; for phone apps, do dog whistle apps work.

This guide is general information about dog behaviour, not veterinary advice. A sudden change in barking, or barking with signs of distress, is a reason to see a vet.