top of page

When Dog Training Doesn't Work

Red and white dog during dog training session

Sometimes, no matter how hard you try, your dog just doesn’t quite seem to get it. Or maybe they respond when you ask, but with a high latency (they take a long time to do it). Maybe they just look at you like you sprouted two heads and they just walk away.

There are quite a few reasons why your dog may not be responding when you’re trying to teach them something. You are using a newly learned cue in an unfamiliar location, you’ve switched to a new cue improperly, what you’re asking for is too difficult, or your reward is not valuable enough.

Cues are fairly easy to fix. Pay attention to the cues you’re using to ask for behaviors, whether they are visual or verbal cues. Dogs learn and respond much more quickly to visual cues. Verbal cues tend to be easier for us to use and remember. Dogs are capable of learning both, but it has to be done the right way.

When we teach a cue, it’s obvious to us what the cue is and what it represents. For the dog, it’s about what’s salient to them. We can never know what the dog determines is salient. In fact, it may not even be a single thing, but a combination of things. Let’s take the visual cue for sit. Most people raise a palm or fist over the dog’s head. While the dog is learning this in the living room, they may be noticing the couch, the coffee table, the tv, and anything else in the room. As you repeat your cue, your dog may be learning that everything in the living room plus your cue is the signal to sit. This is why when you take your dog to the local hardware store and ask them to sit, they look confused and don’t sit. You need to re-teach your sit cue in a variety of locations so your dog learns that there is only one relevant cue. Learning a single cue in multiple locations is called generalization. Some dogs generalize more quickly than others.

Teaching a dog a behavior is often done by luring. This is the process of putting a treat in front of a dog’s nose and then maneuvering the dog into a position or location. Because the luring process tends to introduce a hand or arm movement that translates into a visual cue for the dog, it’s an easy transition from luring to cuing. This is where it can get tricky adding a verbal cue. Our tendency is to say the verbal cue at the same time as the visual cue. When this happens during the learning process, the dog picks one cue to pay attention to and only learns that one. This is called overshadowing. After the dog has already learned the visual cue, when we give a verbal and visual signal at the same time, the verbal cue is ignored as irrelevant. This is called blocking. The way to overcome these mistakes to to introduce the new cue followed by the old cue. Each cue has to be distinct.

Complex or difficult behaviors take time to teach. They often need to be taught in steps. Let’s take teaching a dog to roll over. This may not sound like a very complex movement for a dog, but it can be, as dogs often dislike the feeling of falling as they move from a lying down position to a position on their side. The conventional way of teaching roll over is to lure the dog’s head over their shoulder. A dog will lean over further and further until they stop. Some dogs will make it all the way over, some will make it to their side, and some will stop as soon as they feel themselves “falling.” If the dog stops before completing the rollover, you may try luring the dog over the shoulder again. However, a dog that stops just short of falling may stump you and frustrate the dog. To get past this obstacle, you must “split” the behavior into achievable pieces. This means that instead of waiting to reward for the end goal, you will reward the dog for making tiny, incremental progress toward that goal. Eventually, all those little steps will lead to your desired outcome.

Many people complain about how long it takes for a dog to respond to them. The time it takes from giving the cue to the dog responding to it is called latency. Latency can be caused by a few things. Sometimes, we’ve rewarded latency. I remember tales of chicken training camp (yep, that’s a thing), where attendees who rewarded slower responses during agility course training got slower course times than those who rewarded faster responses. So how do you get lower latency? Only reward the responses that are faster than the others. This takes a good eye and patience, but as you practice, your dog will respond more quickly.

Another cause for high latency is that your reward isn’t valuable enough for the behavior you’re requesting. This is actually a more common problem than rewarding latency. It’s easy to underpay your dog for the work they do. We become complacent that our dog has always done this behavior for this reward, so we expect it to be sufficient every time. Or we aren’t mindful that a behavior that has been performed easily in the past has for some reason become more difficult or onerous for the dog to perform. Or we have established a history of underpaying and now our dog has decided it isn’t worth it to do the task. This is easily solved by increasing the value of the dog’s reward. Remember that the reward should always be commensurate with the difficulty of the job. Recall is a behavior that dogs often delay to perform. This is often an indicator that a strong enough history of high value reinforcement hasn’t been established. If you paid me $1 to interrupt something I was enjoying doing, I’d probably delay coming over, too. But if you paid me $1000 every time you wanted me to come to you no matter what I was doing, you’d better believe I’d be there.

There are many reasons you might not be succeeding in your dog training goals. This post only mentions a few common causes and their solutions. If you are having more complex difficulties, contact a force-free trainer to help you out.

21 views0 comments

Recent Posts

See All


Rated 0 out of 5 stars.
No ratings yet

Add a rating
bottom of page