“The day came when the risk to remain tight in a bud was more painful than the risk it took to blossom.”
— Attributed to Anaïs Nin, possibly Elizabeth Appell.
During my years in graduate school, I became heavily involved in dog training and competition. It was a welcome distraction from the stress of graduate study – it connected me with a network of friends and mentors outside of academia, providing me with an escape from endless talk and fretting about classes, exams, and publishing. It was more than a distraction, though. Through dog training, I learned important lessons about interpersonal relationships and leadership. I learned a lot about the management of fear and anxiety through empathy, setting the stage for success, and minimizing the cost of failure.
I often think about how these lessons apply to helping people exit their criminal careers. It is awkward to talk about, because people don’t immediately understand the relevance of “dog training” work working with reentry populations – “You want to treat people like dogs?” No, I treat dogs like people – learning is learning, after all. The way that I learned to rehabilitate fearful/anxious dogs has nothing to do with “dog” and everything to do with creating a safe environment for learning and growth.
I learned that fear and anxiety are anathema to learning. If I may go full-on nerd for a moment, “Fear is the mind-killer.” The first step to helping a student (here, a dog) overcome fear is to start in a calm, non-distracting environment far from whatever causes the fearful reaction – another dog, for example. Trying to teach a student anything when he is already scared/anxious is destined to fail – no being can learn in that state. So remove the fear.
Another important lesson is that you can’t punish the fear/anxiety out of the student. Punishing the student for fear-based reactions is not going to resolve the fear or anxiety. The best outcome of such an approach is a student who has learned to hide his fearful behavior until he reaches breaking point – this is why you should never punish a dog for growling, for example. You have not taught the dog not to feel fear, you have simply taught him not to show that he is scared – at some point, backed far enough into a corner, he may bite “out of nowhere,” as so many dog bite stories go. You can’t punish away the fear.
It is also ineffective to train through punishment of wrong “answers.” When a student offers a behavior and you punish him for it, you have not taught the student anything useful. You have simply told him that this answer was wrong – but he is no closer to knowing the right answer. He now feels that guessing what you want is likely to be met with punishment. The smarter option may be to do nothing at all. Training through punishment is likely to lead to a “shut down” student who doesn’t know what to do, so does nothing. This makes future training even more difficult, because a student who won’t offer any “try” at all gives you very little to work with to shape your desired behavior.
What does pay off, where training is concerned, is to ignore wrong answers and reinforce the right answers. Ignoring the wrong answers minimizes the cost of failure – the student does not learn that trying might result in punishment, simply that not all tries result in reward. Set your criteria for success (“There is only one right answer,” my trainer would tell me – don’t confuse your dog through inconsistency) and mightily reward correct guesses while ignoring the incorrect guesses. The incorrect guesses are, to quote my trainer again, “just information.” Behavior, even unwanted behavior, is just information. If it isn’t the behavior you were looking for, try to understand why it happened and approach the training differently next time – maybe you pushed too hard, too fast. Maybe you were unclear. Maybe you asked for more than the student could give you in that context (environment, emotional state, etc).
By following these steps, you create an environment where failure is less painful/costly and effort is rewarded. Rather than “shutting down” a student, you end up with a student who will repeatedly offer behaviors in hope of hitting the right one and receiving positive reinforcement. A student who offers behaviors has near-infinite capacity for more learning, because you can seize on that effort and start shaping it toward what you want. For example, once my treats and clicker come out, my dog starts exploring his environment to see what will “pay off.” If I put down a new object, he will approach it and begin to interact with it, because this has paid off in the past – he will bop it with his nose, tap it with a paw, push it, circle around it, all while looking to me to see if he is “right.” Starting with these behaviors, I can start reinforcing and then shifting my criteria in the direction of my desired outcome – ringing a doorbell, standing on two legs on an upturned bucket, or any other seemingly complicated trick.
This all brings me back to reentry and desistance. The risk-oriented approach to community supervision treats clients as “risky” – risks to the state, risks to the public, risks to themselves. We assess their risks, we make decisions about how they will be treated, we give them lists of things to do (“find a job, go to school, get a house”), we punish them for failures. What we rarely recognize is the risk they undertake in reentry and desistance. Changing your friendship network is risky. Cutting ties with old relatives and friends is risky. Disclosing your supervision status to a potential employer or landlord is risky. Entering new social roles and reshaping your identity is risky. Going through therapy or treatment is risky. All carry the threat of discomfort, pain, shame, humiliation. If a client experiences pain and humiliation from trying and failing, perhaps she, too, will end up “shut down,” paralyzed by the thought of trying and failing again. Better off to not take the risk at all.
How, then, could we apply the lessons of positive reinforcement training to create less risky, more learning-supportive conditions for people on probation and parole? Following the lessons above, we could begin by minimizing fear and anxiety. Perhaps we could send stronger messages of support and encouragement. We could alter the physical environment of probation and parole offices to reduce visitor anxiety. We could minimize the “cost” of failure by not punishing small, relatively harmless violations. We could recognize the feelings of guilt, shame, and humiliation that are part of the parole experience – the “parole paradox,” to quote one of my favorite papers. We could do more to recognize effort in the right direction – making the “trying” worthwhile. Finally, we could celebrate successes, no matter how small, and boost clients’ self-esteem and desire to try, try again.