I have been gone a bit, and I am not up to speed on this zero latency bit. Saying that, let me suggest that the animal (or human, for that matter) should be given some reasonable time to receive, interpret and then respond to the command. Assuming that an animal has many commands or cues to which it must respond, and that you don't want the animal being too context sensitive (that is, certain surroundings are most likely to particular cues), you want the animal to spend a reasonable time to sort out the correct response to the cue.
Thus, if you want the dog to respond to the FULL sit, and not just to any command with initial sibilance, you don't want the dog going down with the initial SSSSS, you want it to wait for the word S - I - T to be completed. If someone is trying to get Latencies less than one second, be sure that there is someone watching or a camcorder running to assure that the dog is not ANTICIPATING the cue.
Now, I would say (and I have NO data to support this) that properly conducted speed trials would be worthwhile. Why? For the same reason that our students become faster at clicking when we conduct speed trials. Simply put - sitting is a mechanical skill. If a dog has never done a lot of sits, and if the trainer has never pushed for speed, they both need lots of experience. The trainer can't ask for too much too soon, but, sooner or later a slower sit must not be reinforced and the dog is given the message that a slow sit doesn't pay off.
Also, do remember that extinction trials can lead to more vigorous (faster) action. HOWEVER, play the extinction card carefully. Gary Wilkes is a master at it, but he uses it a lot and he is willing to take the risks inherent in this approach. Please, please, those of you who are new at training, don't push the extinction button too enthusiastically. You can create lots of trouble, not just with the behavior in question, but other behaviors too.
Let me also add that I am pleased that some are very interested in short latencies, BUT, like other aspects of training, don't get too carried away. Sharp behavior does not have to measured in milliseconds - several tenths of seconds are really pretty good for the animal. The trainer should strive for short latency between the animals response and his or her reinforcement (or other response, such as turning around or averting eyes). Now, that response should be ASAP, or zero latency.
Copyright of all posts is the property of the original author. Please obtain permission from the original author before copying, quoting, or forwarding.
List and Site Owner: Melissa Alexander, mca @ clickersolutions.com