Dogs pulling on leashes is an age-old problem – both for the dog-owning public and the dog themselves. For dogs, leash issues have been cited as one of the leading culprits for otherwise healthy dogs being left in animal shelters. Additionally, excessive pulling on the leach can potentially damage the neck and the eyes of the dog. Though for dogs and humans alike, dogs leash pulling can be potentially dangerous, especially if the dog is large. Despite thousands of online videos, training mechanisms, collars and harnesses aiming to address leash pulling, 69% of dog owners report that their dogs pull on the leash [1]. But what influences this pulling behaviour?

Investigating this problem PhD student and vet Hao-Yu Shih and team (Fillipe Georgiou, Robert A. Curtis, Mandy B. A. Paterson, and Clive J. C. Phillips), developed a pull direction and force device. This system measured whilst on a walk, the quantitative measurements of the animals’ movement in the X, Y and Z directions and the gravitational component (devices orientation, although this was not used in their publication). These measurements together measure both the force and the direction of the pull from the dog. They also measured the humans pulling behaviour of the max and mean tension by human and pulling frequency.

Hao-Yu Shih then tested their sensor over various dogs sizes and ages with the RSPCA. Here they found that heavier dogs have greater leash tension but pulled less on leashes. Younger dogs, on the other hand, were more active pullers but less predictable in their behaviours pulling more frequently. Naturally, dogs who were rated by the RSPCA walkers as better behaved (scored on levels 1-3) had lower leash tension.

For the humans walking dogs, they found that the humans pull was not correlated with the dogs pulling behaviours. Dog walkers pulled the leash harder (but less frequently) when walking large dogs, tugging less with smaller dogs.

From these findings, Hao-Yu Shih et al. suggest that a dog leash meter could be used to grade dogs behaviour when walking on a leash to improve both the human and dogs experience by matching their leash expertise together. However, they have yet to test this device with various environmental challenges. In the future, they plan to look further at how the demographics of the human walker affect the pulling behaviours.

For the animal-computer community, this leash tension meter provides a way to measure the leash reactivity in dogs. But the future implications are wider: devices like this could provide not only the tension of the dogs pull but also reveal more information about humans pulling behaviour to correct both. A system like this could help provide dog walkers with a way of understanding and highlighting the dogs’ and their pulling behaviours utilising AI and other analysis methods to dig deeper into walking behaviours. This could then improving both the dogs and the humans, experience of a walk by tackling the pulling issue at both ends of the leash.

Read more in their paper: “Behavioural Evaluation of a Leash Tension Meter Which Measures Pull Direction and Force during Human–Dog On-Leash Walks” https://doi.org/10.3390/ani10081382. You can also find out more about Hao-Yu Shih on his website and OrcaID.

[1] Pirner & McGlone, 2016. https://doi.org/10.3390/ani6050034