>Do you know if they train the underground fence >starting at the lowest level possible then work >up to higher levels as needed? When I tried a no->bark collar with my sheltie, that was what was >recommended and even with a shaved neck, I still >needed to go to a very high level.
I don't have a lot of experience with underground fencing, but I do have rudimentary knowledge of how punishment based training methods work. Starting with low-level corrections and slowly increasing the intensity level can desensitize the dog to the correction. (I've witnessed this first hand with the bark collars).
Pam Reid discussed this in her book "EXCEL-ERATED LEARNING, Explaining in plain English how dogs learn and how best to teach them."
Here is the quote from her book regarding effectiveness of the intensity of an aversive stimulus:
"Generally, the more intense the aversive is, the more the response will be suppressed. Low-intensity punishment produces only moderate suppression and the behavior often recovers despite continued punishment. High-intensity punishment, on the other hand, can produce complete and long-lasting suppression. Animals are able to adapt to punishment. In one study, it was found that rats trained to run down an alley for food would continue running even when severe shocks were given in the goal box, provided the rats had been exposed to a series of shocks of gradually increasing intensity. Other rats that received the intense shock from the outset stopped running completely."
So anyway, she summed up saying that it's better to start out strong and then back off.
FYI: she wasn't recommending punishment-based methods; however, if one chooses to use bark collars or underground electric barriers her remarks are worth keeping in mind.