Punishment in Dog Training – Is It Worth It?

The question of whether or not to use punishment in training is an on-going controversy within the dog-training industry.  I’m a dog trainer by profession, so this is something I’ve been struggling with for some time.  I’m primarily a reward-based trainer and I rarely use punishment.  I use reward-based training because it works – not because I’m a soft-hearted, tree-hugger who thinks dogs can do no wrong.  I occasionally – well, maybe a little more often than occasionally — yell at my dogs; however, this is not training, it’s venting, and there is a difference. Anyone who is working to become a professional dog trainer or to become a certified dog trainer – and, most importantly, anyone who is working toward a dog training career and thinking of attending a school for dog trainers – must understand this philosophical issue to make an informed decision.

Without overloading you with technical stuff, there are some basic terms and premises you must understand.  The terms used in this article are scientific terms – they are used in everyday language as well, but they have different meanings.

Animals learn through behavior and consequence.  In other words, something happens (antecedent), the animal reacts (behavior) and something else happens (consequence).  For instance, you want your kid to take out the garbage so you say, “Billy, please take out the garbage” (antecedent); Billy takes out the garbage (behavior); when he comes back from taking out the garbage, you have a piece of cake ready for him (consequence).  This series of events causes Billy to learn that taking out the garbage is a good thing and will increase his garbage-taking-out behavior.  Another scenario could be that you say “Billy, please take out the garbage.”  Billy takes out the garbage; when he comes back, you say “I don’t know why I always have to ask you to take out the garbage!”  This series of events causes Billy to learn that taking out the garbage is unpleasant and will reduce his garbage-taking-out behavior.

When speaking, behaviorally, of punishment and reinforcement, positive means you add something, negative means you take something away, punishment means behavior decreases, and reinforcement means behavior increases.   There are five consequences to behavior:  positive reinforcement (+R); negative reinforcement (-R); positive punishment (+P); negative punishment (-P); and, nothing!  So, +R means something was added to increase behavior and +P means something was added to decrease behavior.  If nothing happens, that results in the behavior extinguishing (going away).  Another key thing to remember is that the one performing the behavior decides whether or not something is a reinforcer or punisher – not the one providing the consequence.  Therefore, once we decide on a plan of action to modify a behavior, we must periodically step back and ask ourselves if it’s working; if not, we need to revise our plan – this is what separates a professional dog trainer from a hobbyist.  If we are using what we consider to be positive punishment and the behavior isn’t decreasing, then it isn’t punishment and may very well be abuse!

In order for +P to be effective, it must meet the following four criteria:

• it must happen immediately
• it must be sufficiently aversive
• it must happen every time
• it must be associated with the behavior

There is also fallout associated with punishment.  The effects will vary from dog to dog and sometimes you might not even realize it is happening.  However, following are some things you might expect from applied +P:  It can cause physical pain/damage to your dog; if improperly applied, it can require stronger punishment to get the same results; it can cause emotional problems such as anxiety; the dog may become aggressive; the dog may develop learned helplessness; or the dog can develop a fear of the owner.

So, the obvious question is:  Why do we use positive punishment?  No one knows exactly, because it has been proven beyond a doubt to be an inefficient means of teaching behavior.  We have come up with some good ideas of why we punish.  One is that humans, being primates, are great imitators and we grew up learning about punishment – spankings, time outs, etc.  Another is that it is rewarding – when we punish, the behavior usually stops, for the moment, so we get immediate reinforcement and we get the feeling of having accomplished something; unfortunately, that doesn’t mean the animal we’re punishing learned anything other than to be afraid of us!

Up to now, everything in this article is fact.  So, here’s what I think!  I think aversive experiences are excellent teachers; however, the most effective aversives usually happen in nature. Perhaps the reason they’re so effective in nature is because when they happen in nature they generally meet the four criteria — it happens immediately, it is sufficiently aversive, it happens every time, and it is associated with the behavior.

As I see it, the problems arise when people try to administer punishment.  They rarely meet all four criteria.  Being human we are fallible — our timing is off, we don’t catch it every time, or something else goes awry.  If we can catch the embryonic behavior (the very first time it happens) and meet all the criteria, it is probably very effective. However, in my opinion, trying to stop an established unwanted behavior with punishment is likely to result in the fallout listed above. Those of us who have become a professional dog trainer do have a better understanding of the importance of the four criteria – timing, in particular – but it is very difficult for those less knowledgeable to efficiently administer punishment.

So, my conclusion is that, although punishment can be very effective, it is probably not the best way to approach training — particularly if it is an established behavior.  Further, I see absolutely no place for it when trying to teach a new behavior – it isn’t nearly as effective as reinforcement.  Possibly it has a place when trying to get rid of a bad behavior, but you can’t go off half-cocked; you must make a plan that takes into account the four criteria for using punishment, stick to the plan, and periodically review and revise it.

So, what am I going to do?  I’m already well on the road to not using aversives – I don’t use them in formal training but I do use them sporadically when I simply don’t want to deal with the situation.  So, I must make a plan for myself, including rewards and punishments for my behavior, and implement it.  I must periodically step back, review my plan and revise it if necessary.  Am I perfect?  No.  Do I expect to have set backs?  Yes.  But I have a goal and I know how to achieve it, so I have high hopes that I will someday be aversive-free except when I have made a deliberate, informed decision to use aversives in a situation where I think they will be extremely effective.

Raising Canine has a school for dog trainers which focuses on operant training for dogs, dog behavior, working with clients and addressing client compliance, and the science behind behavior modification.