Sunday, June 30, 2013

Shedd Animal Training Seminar: Problem Solving

Problem behaviors in animals can vary widely; beyond aggression, which we’ve discussed in previous Shedd Seminar installments, behavior problems can range from inappropriate elimination to fear to resource guarding to a trained behavior simpy falling apart. Ken said that when solving problem behaviors, you should have some kind of system in place to ensure that you aren’t missing something. The particular system doesn’t matter so much as the fact that you use one that makes sense to you.

Don’t have one? No problem! Ken shared his five-step plan to solving problems. And the most important steps are the first three; most problem solving fails because people don’t spend enough time creating a solid plan before they try to implement it.

This picture has nothing to do with the post. I just like it.

Identify the Problem
It might seem obvious that the first step is to identify the problem, but this is not as easy as it sounds. Behavior is complex, with many parts, and often only certain parts need fixing. Break it down. Look for patterns. Ask others for their feedback (what do your kids, baby-sitters, housekeepers, friends, etc. think?). Write a thorough description, including what the behavior looks like. How is the animal misbehaving? What are they refusing to do or doing wrong?

You also need to accept responsibility. Labels often get in the way of finding solutions; they become excuses that trainers feel absolves them of responsibility. In turn, this makes it harder to fix. One way to do this is to shift your thinking. Instead of trying to figure out what’s wrong with the animal, ask yourself why you can’t train the animal.

Determine the Cause
There can be many, many reasons that a problem behavior begins. Although knowing the cause may not help you solve the problem, sometimes there is a very simple fix, so it’s worth trying to figure out. Ken presented a systematic method (developed by Desmond and Laule) that can help trainers look for the causes of a problem. There possible categories are: Environmental (weather, facility changes, prop changes), Social (dominance/submission, aggression from another animal, competition for resources, sexual activity), Psychological (boredom, superstitious behaviors, genetic issues), Physical (health, changing abilities in vision or muscles, illness), Trainer (skill level, attitude, emotions), Session Use (length, frequency, pacing), and Regression (which is a normal part of learning).

Consider the Balance of Reinforcement
Every problem has two ways of looking at it. On one hand, a desired behavior may not be happening if it is being punished. On the other, an unwanted behavior can be reinforced. These usually aren’t things the trainer is doing (and when it is, the trainer probably isn’t doing it on purpose). Often, the environment, including other people or animals, contributes to this.

Think about the problem in terms of what you would like the animal to do. Then make two lists: things that reinforce the behavior and things that punish it. Increase the reinforcers, and decrease the punishers. Doing so should tell you a lot about what motivates your animal. (Three big contributors to an animal’s motivation: the animal’s desire to control their environment, their selfishness/the “what’s in it for me?” question, and past consequences to their behavior.)

Develop and Implement a Plan
Only after you’ve completed the previous three steps can you move on to developing a plan. Keep in mind that you will need to evaluate your priorities as you do this; if the behavior was easy to change, you probably would have already done it. Often, changing a problem behavior requires, well… change. Are you willing to sacrifice something (time, energy, money, etc.) in order to solve the problem? If not, the problem may be difficult or even impossible to solve.

But if you are, you can develop and implement a plan. There are many ways to approach an undesirable behavior. Ken cited Karen Pryor’s book “Don’t Shoot the Dog” as a nice summary of the options. These include “shooting the dog” (where you give up on the animal entirely), using punishment (which always has possible undesirable side effects), using negative reinforcement, extinction protocols, putting the behavior on cue (this is not suited for all problems), training an incompatible behavior (teach a specific behavior that the animal can do instead of the problem behavior), shaping the absence of a behavior (anything except the problem behavior is reinforced), or change the motivation.

Constant Monitoring

Problem behaviors may be gone, but they are never forgotten. You can’t “untrain” an animal; the neural pathways are always in the animal’s brain, meaning there’s always a risk the problem will come back. Watch for any early signs that the problem behavior is reemerging and address it immediately. Good records can help with this process.

1 comment:

Ninso said...

Great post! Very clear method for thinking through solutions to behavior problems. The biggest one for me is planning the solution. Earlier in my dog training life I would get fed up with a behavior and in that moment of frustration start working on a solution that in theory could have worked, but that I hadn't really thought through, nor was I truly committed to following through with. Now I have realized that if I really want to change a behavior I need to be "all in," committed for the long haul. I also do a better job thinking through a real plan. This means I address fewer behaviors, but I usually get good results with those I do address. The ones I'm not willing to fix, I can often just manage well enough that it's not a big deal.