Positive Reinforcement: Adding A Bit More Control
Positive reinforcement has been making waves in almost every behavior modification community. Even if you’ve only lightly dabbled in operant conditioning, there’s a good chance you’ve heard this phrase tossed around at least once. But what is it exactly? And why does everyone recommend it? Do nice words and coddling actually help train things? If your answer was “yes” to that last one, then this article is especially for you. But fear not, this article is here to go over what positive reinforcement is, why people use it, and how.
What is Positive Reinforcement?
Positive reinforcement is when you give a reinforcer to the subject so that they perform the behavior again. Just in case you haven’t read the previous article on reinforcement, a reinforcer is just something the subject likes, wants, or is willing to work for. The idea is, if someone or something performs a behavior and they get what they want out of the behavior (a pleasant consequence), then the behavior will most likely be performed again for the same or similar results.
But that’s all positive reinforcement is: giving something the subject likes (a reinforcer) to the subject so their behavior will happen again.
The classic example of this is when a dog gives you a desired behavior, like sitting, then you give a treat. Because the dog likes the treat, it will want to sit again so they can get another treat. Eventually, the dog will associate listening to you and sitting, with getting a treat, something the dog likes.
Another example is if you’ve ever said, “good job,” or even “thank you,” then you’re applying a positive reinforcer (adding a compliment) that will increase how often the behavior you’ve complimented will occur.
Real Uses With Animals
If you’re starting out, or just trying to train something, then positive reinforcement is one of the easiest concepts to use. For animals specifically, what’s important is that you’re trying to establish a line of communication with your animal. Ultimately, you are communicating that if they follow what you’re saying, then they get a treat.
A useful tool to help you communicate is a clicker; it’s there just because you can’t always feed your dog a treat when they do the behavior you want (for example, if you’re trying to reward a jump, you can’t give your dog a treat mid-jump), so a clicker will tell your dog the exact moment that you are rewarding.
A Few Quick Steps to Help Start Positive Reinforcement Training
First off, you want to start by clicking at random intervals for almost no reason. Every time you click, give the animal a treat. By doing this you’re establishing two important rules:
- A click always means something good. Early on, you want the animal to associate clicks with treats. Later you can ease up on the treats and focus more on the clicks. But remember, a click always means what they did was right. This turns into a rule that will be important later, but making sure your animal understands that this is a constant rule in training will help them understand what you want faster. Never click and then provide no reinforcer; this is dangerous to the learning process!
- Providing treats with clicks establishes the clicker as a secondary reinforcer. The clicker is a fast and expedient way to mark whenever the animal you’re training did something right in the exact moment they did something right. It’s like playing a high-speed game of hot/cold. Click means the animal is getting hotter toward what you want, while no click means colder.
After you’ve clicked and given a reinforcer (ideally food) for the first few times, you can now sit and wait for your animal to do something you like before you click. After that, every time they do that behavior, click and treat. Once the animal understands that you will reward that behavior, associate it with a command by saying the word you want to be associated with it before/during the behavior. After that, if you ask for the behavior with that word, then your animal can make the association between the word and behavior a little faster. This association will be talked about in a later article, but it’s worth mentioning here so you can apply this knowledge immediately.
What About With Humans?
Humans can be a little bit different. While I can’t help you too much if you’re trying to help someone else with a problem, I can give you suggestions on ways to help improve your own behavior.
A friend of mine used to spend a lot of money on eating out at restaurants. Once he realized how much money he was losing, he came up with the idea to reinforce cooking at home with money. Every time he wouldn’t eat out and instead cook food at home, he would put a $1 into an account for him to use on whatever frivolous buy comes to mind. While it isn’t the most efficient solution, it still worked. He built the habit of cooking at home over just eating out more. He did the math and figured out that, at most, his reinforcement schedule only added a $1 to each grocery shopping trip compared to the $10+ additional dollars he would have spent had he decided to eat out.
With humans, a schedule like my friend’s makes the most amount of sense. Reinforce yourself with the behavior you want with an incentive you want. For myself, I usually have a set of specific songs I really enjoy listening to, and I set these songs to play whenever I decide to start working.
But if you’re trying to test this on yourself, make sure you document the behavior you are trying to increase/decrease with data, and implement a change.
What Are The Downsides?
Reinforcement alone isn’t perfect, and positive reinforcement runs into a few issues that you should address. Be mindful of the downsides of what reinforcers you use and how best to manage them in a healthy way. For example, treats are a wonderful positive reinforcer, but too many additional calories can lead to health problems.
Another problem mentioned earlier is timing. It can be a problem making sure you time giving the reinforcer so that you can reinforce the correct behavior. While these articles will make it sound like subjects will just give you a behavior and see your response, usually what will happen is that your subject will throw a mess of behaviors out at once. Clicking on the behavior you’re looking for suddenly turns into a game of “did the clicker like this one? Or this one?”
Another problem with timing is if you short-circuit the behavior. This specifically hits people who are using these principles to improve their own lives, but if you are using positive reinforcement, make sure that you NEVER give the reinforcer before you actually do the behavior. While sometimes it may feel like you deserve it, or you want to give yourself a break, do not give in to temptation! This is known as short-circuiting a behavior, and will result with almost no results in improving your behavior. In those situations, you are positively reinforcing yourself choosing to be lazy instead of following through with your original plan.