The “Bait and Switch” of Lure Training

What’s the best way to get your dog to perform the behaviours you want? Eric Brad explores the similarities and the differences between “Lure and Reward” training and “Mark and Reward” training.

One of the tried-and-true methods of dog training that continues to enjoy popularity is “Lure and Reward” training. The website Dog Spelled Forward describes a “lure” as “anything your dog will follow such as a piece of kibble or a toy”. It’s a method that’s worked for dog trainers for over half a century. So when I was introduced to the technology of Clicker Training, or “Mark and Reward” training, I expected to find not only a similarity in techniques, but also an enthusiastic endorsement of “luring” and “lures” among the best Mark and Reward trainers. As it turns out, this was not the case.

The primary difference between “Lure and Reward” and “Mark and Reward” training is that “Mark and Reward” uses a marker signal to indicate when the dog has done the correct behaviour.  A “click” or a “whistle” signals “that’s what I was looking for” and a reward is delivered. As the following video shows, “Lure and Reward” uses no such signal. The food or another lure is used to get the dog to do what is needed, and the delivery of the lure is meant to reward the correct behaviour.

So let’s take a closer look at “Lure and Reward” training. “Luring” is essentially holding a piece of food or desired item, such as a toy, in front of the dog to lead her through the behaviour you want her to do, then delivering the lure as a reward for compliance. This video illustrates the basic concepts of using a lure:

“Lure and Reward” has produced results for thousands of dog trainers, so I was curious why some of the best “Mark and Reward” animal trainers would even debate the usefulness of such a technique. It turns out that the objection was NOT that the technique didn’t work, but that it was inefficient and had the potential to create problem behaviours where none existed before.

The Nose Knows

In her book Inside of a Dog, researcher Alexandra Horowitz makes this startling contrast about the sense of smell in humans: “Human noses have about 6 million of these sensory receptor sites [olfactory, smell]; sheepdog noses over 200 million; beagle noses over 300 million.”  Those are some astounding numbers! Now let’s take a look again at how luring works: you hold a piece of food up to the dog’s nose to maneuver her into position.

Considering how acute and sensitive a dog’s sense of smell is, the experience of having food held that close must be nearly overwhelming. As the trainer deftly moves the food to get the dog into the desired position, the dog’s focus remains firmly fixed onto the food in the trainer’s hand. Once in position, the trainer delivers the food for the dog having performed the desire action. But has the dog performed the action?

Most trainers would say, “Of course she has; she’s sitting!”  But did the dog “sit” or did the dog follow the food, with the “sit” as just a by-product of following the food?  This may seem like a simple semantic distinction but it’s not. It speaks to “intentionality”. Did the dog intend to sit or did the dog intend to follow the food and just ended up siting?

The essential question here is: does the smell of the food so overwhelm the dog’s senses that she literally doesn’t know what her body is doing? Clearly, the experience is not so overwhelming that a dog cannot learn this way. But it’s a bit like teaching a child to turn her head on command by showing her a large brightly-colored sucker moving to her right and rewarding her afterward. The child will eventually learn to turn her head when asked, but does she really know that in the beginning when she is just watching the sucker? Unfortunately, we can’t ask the dog what her actual experience is so we will have to be satisfied with the limits of what science can tell us about dog behaviour.

What Else are They Learning?

So you start doing your “Lure and Reward” training. Before long, you may notice your dog has an unnatural fascination with your hands. Why would that happen? Because she has seen it presented to her as a training aid dozens, if not hundreds, of times! Many trainers discover that their dogs watch their hands a great deal of the time. In fact, some use an offered hand (without a treat) to gain the dog’s attention.

Classical Conditioning tells us that a dog will associate food not just with the trainers hands after a while but may also begin to recognize that certain gestures or positions of the hand are more likely to indicate the presence of food than others. The pinched fingers extended toward the dog becomes recognized as a signal to “come and get it” for example. But our dogs may pick up on even more subtle cues such has the hand reaching toward a bait pouch or pocket to retrieve the treat.

“Lure and Reward” training also requires that the trainer be within arm’s reach of the dog in order to deliver the treat. This can be limiting in two important ways. First, it limits the range of behaviours that you can “lure” a dog to do. For example, it’s impossible to “lure” a dog to go out and around the post located 20 feet away from the trainer. There is no way for the trainer to hold the lure for the dog and remain 20 feet away.

Second, the trainer’s ability to effectively keep the lure in the correct position in order to get the dog to do the proper movement can be an inhibiting factor. Does the trainer have a good plan for where to move the lure to get the desired movement? Move the lure too fast or too far away and the dog may begin reaching out for it, lunging to get the treat. Move the lure too slowly and the dog may begin “mugging” the hand to try to get you to give up the treat. So the trainer’s ability to handle the lure effectively is very important in this kind of training.

So regardless of what behaviours the dog is learning in Lure and Reward training, she’s also learning a great deal about the trainer and when, where, and how the trainer is or less likely to deliver a reward. “Stay close and watch the hands” would be a recipe for success for most lure and reward dogs.

Fading Fast!

One of the key components in any “Lure and Reward” training program is the instruction to “get rid of the food quickly”. In fact, Dr. Ian Dunbar suggests on his website that “food lures should not be used for more than half a dozen trials” and that “prolonged use of the same item as both lures and rewards comes pretty close to bribing…the dog’s response will become contingent on whether or not the owner has food in her hand.” Clearly the lure is only there for a short time. Then you must repeat the same action with an empty hand and hope the dog repeats the action.

Remember that highly sensitive and skilled nose we talked about earlier? If we’re dealing with food, the dog knows when you don’t have the treat. In fact, the reason Dr. Dunbar suggested that you get the food out of there as soon as you can is precisely to avoid the problems we discussed above — watching hands, staying close for the reward, etc.

Interestingly, what Dr. Dunbar suggests is well known to “Mark and Reward” trainers and “Clicker” trainers. Dr. Dunbar is describing what we “Mark and Reward” trainers call “fading” or making something important become gradually less important to shift the animal’s focus to what is really being asked of them. So from the perspective of “Mark and Reward” training, the food used as a “lure” is just another “prompt” used to get the dog to do what we want and, like any other prompt, it should be faded quickly so it doesn’t become an essential part of the behaviour. The difference is that a food prompt is REALLY, REALLY interesting and could be hard to fade without distracting the dog.

And this is where the trap lies. When the new dog owner uses a lure to get his or her dog to sit and it works, it’s “Woohoo! Party Time! Look what I can get my dog to do!”  It becomes very rewarding for the trainer and it may be difficult for that trainer to give up such a useful tool.  Fading out the lure is kind of scary because what if “Fluffy” stops doing the sit on command? So the lure stays a little too long and the training gets sloppy.

And that’s just one version of it. What if the lure is consistently moved too fast? The dog may lose focus and her attention may go someplace other than the training. What if the lure moves too slowly? The dog may become so focused on the lure that all awareness of the desired behaviour is gone in pursuit of the treat in the hand. And this is to say nothing of the difficulties that may arise if the dog doesn’t move in the way the trainer expects when following the lure, or if the dog gets “grabby” for the treat and nips the trainer.

All Training Works — Eventually

Ok, am I saying that “Lure and Reward” training is wrong? No! Am I saying that “Lure and Reward” training doesn’t work? Certainly not! Tens of thousands of dogs and their trainers would make me a liar instantly if I made that claim. “Lure and Reward” training, even with minimal instruction to the trainer, can produce great results in a relatively short amount of time in a very humane and dog-friendly way. But, like other training techniques, it can create problems if not done with proper technique and instruction.

“Mark and Reward” training (specifically, not using a “lure as a prompt”) is an effective and efficient alternative to lure and reward training. It requires no more or no less instruction or skill to do properly. The difference is that one more layer of confusion is removed for the dog. There is no question as to whether the dog is following the food or doing some other behaviour in order to get paid.

It seems that for some proponents of “Lure and Reward” training, it is an “easier way” to help the average dog owner work with his or her dog. There’s no technical terminology, no reference to science and psychology, no added equipment. In essence, they see it as a way to get people to stop yanking on the leash in an effort to train their dog and find a different method.  Proponents see it as a way to move dog owners gently into a better way to train their dogs.

Unfortunately, “Lure and Reward” training is only one small step away from “Mark and Reward” training. With just a slight change in the process and the introduction of a few different tools, dog owners could be working much more effectively with their dogs. What’s more, this whole notion that “positive training methods”, methods that use food (such as “Lure and Reward” and “Mark and Reward” training) are actually “bribing” dogs to behave seems to be based on poorly done “Lure and Reward” training.

So I am NOT calling for all “Lure and Reward” trainers to stop what they are doing. On the contrary, I would encourage them to go beyond the success they have already discovered and look at how they could be training without the lure but using a marker instead to make communication clearer.  From a “Mark and Reward” perspective, food is just one more prompt but it’s one that comes with its own set of problems. And those problems can be easily avoided with a change in methodology.

Next time we’ll look at the “markers” in “Mark and Reward” training and see how they can be so effective and how they can be a more powerful motivator than food.

Until then – have fun with your dog!

 

Photo & Video Credits

Airedale treats – Bogart Handsome Devil 2009 from Flickr
What’s in your hand? – Menage a Moi 2008 from Flickr
Standing up or checking it out? – Dogrando 2010 from Flickr

Video – Dog Training 101: Luring – WhenDogsTalk 2010 from YouTube

Recent Eric Brad CPDT-KA Articles:

• The Dog That Changed My Life
• Keeping Your Dog Informed
• Teaching Your Dog to Not-Behave
• The Simple Complexity of Dog Training
• What The Dog Wants