Contact: LDRidgeway at gmail dot com

Tuesday, May 11, 2010

Walking Recall

Training with the Walking Recall

I've invented many recall programs over the years, in addition to using a variety of published recall training plans.

Recently, I invented a new plan for Laddie, and I'm pleased at how effective it has been in improving Laddie's recall in general, and field recall in particular. I call this plan the "Walking Recall". Here's a description (it's essential that you also read the following section, "Cautions"):
  • Equipment: yard collar and long line. I use a 40-foot line in the city, a 100-foot line in parks. (See "How I Made the Long Lines" below.)
  • Take the dog for a walk. In my experience, dogs don't like to walk at steady, human pace. They like to run ahead, then sniff and fall behind, then run ahead again. Let the dog do that, continuously looping the line up in your hand when the dog gets closer, letting the line out as distance builds. Dogs love this freedom, and I feel that it opens the dog's mind for the learning that is also taking place (next steps).
  • Call the dog as needed. This might happen because the dog is getting too close to a passerby, or you're about to run out of line, or the dog is about to go out in the street, or she's going on the wrong side of a tree and the line will get hung up, etc. Or maybe it's just because you feel like giving her a treat.
  • For each call, call exactly once, using your standard recall cue. For field dogs, that's "Here," using the same inflection you use in the field. If the dog comes, big party, whatever the dog loves. If the dog doesn't come, gently (no jerking, no recriminations) reel her in. I think rewards are good in that situation, too: classical conditioning to the effect that coming (even if the dog has no choice) predicts good things. In addition, you might want to reward the dog if she comes to you on her own without being called.
  • In my experience, a variety of high-value rewards — one or more treats, petting, a tennis ball tossed a short distance, an enthusiastic, "Yay! Good job!" — in unpredictable sequence from one recall to the next is the most powerful method of reinforcing, because it adds the additional powerful reinforcer of surprise. Even the smaller rewards gain additional value because the dog senses she's "due" for a bigger reward the next time. Unlike some trainers, I believe in reinforcing every time.
  • After each recall, as soon as celebration is over, resume walking and, when the situation permits, release the dog with an explicit cue such as "OK", again letting her set her own pace.
  • Time your recalls so that at least 70% do not require the dog to feel any pressure from the line. Too much of that will result in the line becoming a necessary context for a successful recall, producing a "line-wise" dog and significantly reducing the value of this training plan. Always responding correctly may mean your dog has fantastic recall, or it may mean you're not taking her into challenging situations often enough.
  • Don't lure the dog, for example with treats or a swinging tug toy. If you do, she'll be learning to come for that context rather than the context you want, which is the recall cue. For the same reason, don't reach for the reward until she arrives. You don't want her watching you as she comes and stalling if she doesn't see you reaching for a reward.
  • On the other hand, be prepared to reward her instantly when she arrives. If you need extra time opening the wrapper, etc., use happy talk as a conditioned secondary reinforcer, or bridge, so she can recognize the association of her behavior to the reinforcing outcome. Without the bridge, the delayed reward is significantly less effective as a reinforcer for the recall.
  • Do not say the dog's name when calling her, because in field work, she will learn that her name is how you release her to retrieve a mark. If you get in the habit of saying, "Princess, here!" one day you may say, "Princess, here!" when you don't mean to and inadvertently send her racing out into the field after a difficult honor or no-bird.
  • Handling 40 or 100 foot of line with a dog attached is a learned skill. It may not go smoothly at first. :0)
Cautions

Please take careful note the following:
  • Don't let the dog build up speed and suddenly hit the end of the line. That's a lot more dangerous for a dog than you might expect.
  • Don't grab the line while the dog is running outward. You'll get a rope burn.
  • Don't let the line get underfoot. The dog might flip you if she breaks into a run.
These problems are difficult to avoid. The only solution is to watch closely and try not let dangerous situations arise. When I first started the Walking Recall with Laddie, I needed to pay attention every second.

Mechanics

Here's my experience with the mechanics of the Walking Recall.

I've got two lines, one 40' and one 100'. Some evenings, I use the 100' in Prospect Park, a humongous park in the middle of Brooklyn. I use the 40' line for our mid-day walks every day we're there.

Even the 40' line is unwieldy, but I've gotten more skillful over time. I'll try to describe it:
  1. I put my left hand thru the handle at one end so that it's around my wrist as a safety measure. Of course I attach the snap hook at the other end to Laddie's collar.
  2. Starting with my right hand at my left wrist, I slide the line over my right thumb as I stretch my hands apart.
  3. Lightly holding the line with my right hand, I move it closer to my left hand again and loop the line over the inside of the fingers of my left hand.
  4. I repeat steps (2) and (3) over and over again, like an accordion, until I've taken up all the slack.
Now we begin our walk. Of course, when Laddie moves away from me, I let loops of line fall off my left hand one loop at a time. When he moves toward me, or when he stops and I'm catching up with him, I go back to the accordion motion, again taking up the slack. I proceed in this way the entire walk.

The key here is that the line is never slack. I'm essentially a human retractible lead.

Another key is that the distances I permit are based on the control I have confidence in. Yes, I want Laddie to occasionally need the line for difficult distractions, but I'm not going to let him get 30' from me and 3' from a kid. If I see a kid coming, I call Laddie and take up the slack before there's any risk of him having enough line to get to the kid. Every once in a while, someone will surprise us and step unexpectedly out of a doorway. I just have to hope he's not scared of sweet little Golden Retrievers.

Walking Recall versus Off-lead Walking

A similar approach to the Walking Recall is to walk the dog off-lead, only calling her when you're willing to bet $100 that she'll come. It may seem as though that approach would be similarly effective, but my experience is that the Walking Recall works better.

The advantage of the long line is not only as a safety net when you guess wrong, but more importantly, that you get to practice situations where the dog is not yet ready to come on her own. I didn't realize in advance what a significant effect that would have on the training, but when I thought about it later, it sort of made sense.

For example, the dog sees a squirrel and starts toward it. Without a line, you can't be sure the dog will respond to a recall so, being smart, you don't try, and the dog runs after the squirrel. Here was a golden opportunity to practice a difficult recall, but you had to pass it up. The dog didn't rehearse a failed recall, which is good, but the dog also didn't rehearse a successful recall, which is a missed opportunity. By contrast, with the line attached, the dog does rehearse a successful recall, either by choice or by no choice.

This sort of thing could happen a dozen times in a single walk. It may not sound like it would matter that much, but I'll tell you that just a few sessions of Walking Recall substantially improved Laddie's field recall after years of training recall in other ways.

Benefits of the Walking Recall

I've found that most recall programs produce skill in particular settings, with insufficient generalization to other contexts. By contrast, the Walking Recall seems to improve the dog's overall recall response. Here's what I found with Laddie, and what I'd expect with other dogs as well:
  • Naturally the dog will get better and better walking on a long line with practice. For example, she will become less and less likely to approach strangers or other dogs, because every time she does, you'll call her, and either with or without the line, she'll come to you instead. With no pay-off for attempting to approach others, she'll gradually stop trying. Similarly, she'll become increasingly less likely to try chasing a pigeon or going into the street.
  • She'll become better at walking on a short lead. Though walking on a long line may be more enjoyable for her, she'll be gaining skills that enable her to walk more comfortably on a short lead as well.
  • She'll become better at walking off lead. Again, the same skills will kick in, and her whole style of interacting with you on a walk will change. In fact, I would suggest that you not walk the dog off lead until you can depend on her recall, even in confined space. You don't want to go back to having her learn that she can ignore you.
  • Most importantly from my point of view, her field recall will dramatically improve. Granted, a field recall is different from walking on a lead, but my experience is that the dog's responsiveness to "Here", even from hundreds of yards away, improves remarkably after just a few sessions of the Walking Recall, despite years of previous recall training.
How I Made the Long Lines

I'll end with a note on how I made the long lines:
  1. I went to an outdoors store and purchased climber's webbing, great stuff for this purpose. I got a little extra length to allow for the sewing.
  2. I went to a hardware store and bought a snap hook.
  3. I brought the materials to a shoe repair shop, and asked them to attach the snap hook at one end of the webbing, and to make a handle at the other end. I brought along a commercial leash to show them what the stitches looked like.
I'm sure the materials are available online if you don't mind waiting a few days to receive them.

1 comment:

  1. Good, sound advice. I've been teaching recall this way for as long as I can remember. You make it very clear and have noted the details nicely including a cautions section is valuable. If you can actually get anyone to take your advice and start this recall training before their dogs have self reinforcement history it is even better. Thanks for putting it in writing.

    ReplyDelete