But if You Use Negative Reinforcement Aren’t You Also Using Positive Punishment?

Minus plus

Surprisingly, no. Not necessarily.

You could actually get my answer to this question by reading this other post: Only If The Behavior Decreases!  But of course I’m writing some more anyway.

This issue was a big stumbling block for me when I first started studying operant learning. If negative reinforcement requires for an aversive to be removed, then it had to get there in the first place, right? That’s only logical. So whenever it appeared, there must have been punishment, right? I used to argue about this in my head all the time. But the answer is no, it doesn’t follow that there necessarily was punishment. There absolutely could have been, but it is not logically necessary after all. Here’s why.

Positive punishment: Something is added after a behavior, which results in the behavior happening less often.

Look at the second half of the definition. I’ve harped on this before, largely because I need a lot of reminding myself. The definition of punishment (and reinforcement also) is recursive. We can only know if punishment has occurred by traveling to the future. If the behavior didn’t decrease, there was no punishment.

Aversives, Negative Reinforcement, and Positive Punishment

Employing an aversive and using negative reinforcement do not mean one is also positively punishing a behavior. I have this from the words of Susan Friedman, PhD, in one of her lectures in her Living and Learning with Animals Professional Course. For punishment to have occurred, a behavior must decrease in frequency. That’s the definition. And if one is in a negative reinforcement scenario, the behavior the animal is performing at the onset of the aversive could very well be randomized, because that is not the focus of the training. The animal may not always be doing the same thing when the aversive (e.g. shock,  pinch, pressure, nagging, or appearance of scary monster) starts.

Examples

R- collage 2

Some typical implements and examples of negative reinforcement. Photo credits: Wikimedia Commons (cat o’ nine tails; man scratching), Alan Cleaver (alarm clock), Eileen Anderson (shock collar and stuffed dog with tight leash)

If I decided to teach my sensitive dog Zani to back up by consistently walking into her whenever we were both in a certain hallway (note: I don’t teach backing up that way), her behavior of coming into that hallway when I was there would definitely decrease. Coming into the hallway with me would have been positively punished.

But let’s say in another training scenario I play a mildly aversive noise whenever I want Zani to come get on her mat in the kitchen, and the noise stays on until she gets there. The mat is a “safe place” and getting on it turns off the noise. (Again, I would not do this.) Zani could be anywhere in the house when the noise starts. Her behavior at the onset is randomized, so nothing gets punished. But if I started doing it consistently when she was sitting in a particular chair, she would likely stop getting in that chair.

Since we humans fall into patterns so easily, it is very easy for positive punishment to start happening when we regularly use an aversive. But the point is that it does not have to happen.

Let’s face it, people use aversives all the time without behavior decreasing. That’s one of the many problems with using positive punishment: unless the aversive is strong enough (and well timed enough, and several other criteria), the original behavior may maintain its strength.

Ramifications

So does this mean that negative reinforcement is OK? No. An aversive is an aversive. Just because there is no positive punishment going on doesn’t mean that the training is humane.

If you wanted to reword the question in the title, you could say, “But if you use negative reinforcement aren’t you also using an aversive, just like in positive punishment?” As long as it is recognized that the aversive is used in a different way, the answer is yes.

But the funny thing is, I’ve heard the relationship between the two processes that use aversives used to make two opposite claims, neither of them true in my opinion.

First is the claim implicit in the title. It usually goes like this:

If you use negative reinforcement, therefore you are using positive punishment, so neeter neeter neeter.

I have dealt with that one above.

The second claim goes like this:

Yes, you can have negative reinforcement without positive punishment. And that’s the GOOD kind of negative reinforcement. As long as there is not positive punishment going on,  negative reinforcement can actually be kind of nice.

Excuse me? The existence or non-existence of concomitant positive punishment is irrelevant to how aversive a stimulus is. It is possible to train with a shock collar using a negative reinforcement protocol, for instance.   Again, as long as the point at which the shock is turned on is fairly random with regard to the dog’s behavior, you may not necessarily see a decrease in a behavior as a result of the commencement of the shock.

Negative reinforcement is only one notch up from positive punishment on the Humane Hierarchy, and for good reason.  It always involves an aversive, and employs escape and avoidance, pure and simple. So don’t anybody dare use my argument above showing that there is not necessarily positive punishment in order to say that negative reinforcement is OK.

Coming up:

  • Shut Down Dogs Part 2 (Finally! Coming Soon!)
  • How Skilled are You at Ignoring? (Extinction Part 2)
  • Threshold: It May Not Be What You Think
  • OMG Could She Really be Talking about the Continuum AGAIN?

Eileenanddogs on YouTube

About eileenanddogs

Passionate amateur dog trainer, writer, and learning theory geek. Eileen Anderson on Google+
This entry was posted in Behavior analysis, Negative Reinforcement, Positive Reinforcement, Punishment, Reinforcement, Terminology and tagged . Bookmark the permalink.

14 Responses to But if You Use Negative Reinforcement Aren’t You Also Using Positive Punishment?

  1. Robin J. says:

    Professor Rosales Ruiz (a frequent Clicker Expo presenter) and his animal behaviour students have done quite a bit of academic study on negative reinforcement: the good, the bad, and the ugly. You might find his work interesting.

    http://www.clickertraining.com/node/280

    In addition, both Alexandra Kurland and Katie Bartlett have done a huge amount of practical work on this with regard to training horses. Bartlett has a long article synthesizing current horse trained thought on the topic. Again, I think you’ll find it interesting.

    http://www.equineclickertraining.com/training/negative_reinforcement.html

    As always, thanks for the great work you do on the blog!

    • Robin, thanks for the great resources. I was not familiar with the Bartlett article. There was an interesting bit of synchronicity though. I follow along with several horse trainers who try to use as much positive reinforcement as possible. Recently I read something, it may have been by Melissa Alexander, pointing out that negative reinforcement was unavoidable with horses because of the particular behaviors we want horses to do: be ridden by a human rider, taking cues from the rider’s body. Ms. Bartlett says the same thing. I had never thought it all the way through before. That article is comprehensive and thoughtfully written and I’m so glad you linked it.

      • Robin J. says:

        I think it’s also helpful to remember that, as with the horses being ridden, sometimes our ultimate goal is to countercondition the animal to something that, at the moment, this particular animal finds mildly aversive but that another animal might find pleasureable, and where our hope is that this animal will eventually change her opinion as well.

        Some of Rosales Ruiz’s work, for example, deals with shy dogs in animal shelters who are initially terrified of people. A person moving close to them is highly aversive to them–but another dog in the same shelter might be actively soliciting passers by to come closer, even crave a human connection. The same with eye contact, and eventually touch. If RR’s students reward the shy dog by moving farther away, just feeling in control may make the dog able to tolerate a slightly closer approach the next time.

        This is a situation where the trainer is not introducing a new aversive–it’s an internal assessment by this dog. And for many animals, that assessment CAN change over time.

        This type of approach is also common in horse training. Learning that just a human presence DOESN’T mean actual pain and increasing the distance so the animal feels more in charge can help change the assessment over time.

        So when something is aversive not because it is objectively unpleasant, but because of that particular animal’s expectations of what will follow it, you can use negative reinforcement in a very humane and ethical manner simply by demonstrating over time that the expectation was false. A person coming close doesn’t have to mean pain. You prove that by moving away again. Changing the experience history.

        Indeed, I would say many situations where the handler hopes to earn the trust of a wary animal often benefit from a judicious use of negative reinforcement as part of a counterconditioning protocol. Just ending the training session can be a form of negative reinforcement for an animal who is still wary.

        • I support it when it is simply the only feasible method, as with wild animals, or when less intrusive things have been tried in a thorough and skilled way, and failed. Susan Friedman uses two examples of situationally humane R-. One was with wild parrots coming through customs. Time was of the essence to get them handleable. Doing approach and retreat got the trainers close enough to drop a treat in a cup and then switch to R+. The other was with a giraffe in a zoo that was afraid to go through a certain gate (into an enclosure where the other giraffes were). They designed an elegant R- protocol involving slow and gentle herding by humans carrying sheets of canvas stretched between poles. It was well designed in that even though humans carried the structures, the protocol probably didn’t have a side effect of making the giraffe more wary of humans in general. It was so thoughtfully designed and executed that they only had to do it once. I think it’s interesting that both her examples involve wild animals.

  2. Kathleen Kemp says:

    Thanks for all the info you share Eileen! It is always interesting, informative, based on science and sound research!!
    I am currently starting up my own Dog Training business (been doing it for free for years) having just completed my certificate in Professional Dog Training Science and Technology through Companion Animal Science Institute (CASI). One of the concepts that has been the hardest to get across to people is that we can actually train our dogs using only positive reinforcement (particularly with the people in my town!) so this post is so relevant to me right now!! 👍😃👌🐾

    • Thanks for the vote of confidence, Kathleen! I think using only positive reinforcement (and other ultra humane stuff like antecedent arrangement and classical conditioning) is a great goal. I do think if one lives with one’s animals it’s pretty hard to not have at least some negative punishment in the mix. But I’m with you all the way in not using aversives!

  3. Should I respectfully disagree or just maybe I am just on that “well for an aversive to cease the you have to activate it first right?” I do understand what you are saying, obviously it makes a lot of sense. Is there “good” negative reinforcement at all? Murray Sidman would disagree, how can or should we ignore the consequences of the use of a quadrant that bases itself on avoidance? Still is avoidance after all.

    Also when you say we would have to travel to the future to know if the behaviour has decreased, therefore confirming the P+ application, you are so right, which then makes me think that we should admit all the time that P+ is there shouldn’t we? An aversive is an event/stimulus the animal wants to avoid, the animal does want to avoid it therefore increasing the alternate behaviour that arises from the avoidance, so hence we could just say the is R- is working then P+ worked as well, however the difference being, that one chooses to increase a behaviour instead of working on the decrease of a behaviour.

    Thank you for making my wheel turn. I love your blog. From Portugal Claudia!

    • Hi Claudia, so nice to hear from you! Do you know, you hit on two things I was thinking about but didn’t write about. First, “Is there good negative reinforcement at all?” I had a sentence in there saying that R- was “never nice.” I took it out, since once in a blue moon it might be the right choice, if less intrusive methods didn’t work and if it did solve the problem. In that case it could be “nice” if it made the animal’s life better with a minimum of trauma.

      And your second paragraph! I got to wondering this morning about alternative behaviors. In an R+ scenario, when the non-reinforced behaviors decrease under DRA, it’s not from punishment but is extinction. I got to wondering if you used differential

        negative

      reinforcement of an alternative behavior, whether the other behaviors would decrease from extinction or from punishment because of “leakage” of the aversive used. Probably the answer is, “It depends”!

      I do think that someone who is using

        only

      aversives in training is probably going to have plenty of P+ even if it is not intended. Humans fall into patterns too easily!

      Thanks for writing. Lots of food for thought here.

  4. Bruce Wilson says:

    lets take the scenario of a parrot that wont step off a hand, some people squeeze the parrots toes to get it to step down. now if you look at it from the angle of stepping down increases, to escape the negative stimulus, then it is R-. but looking at the behavior from another angle, staying on the hand decreases, because that behavior is punished by the application of an aversive (squeezed toe) P+. So both (R- and P+) seem to be happening in this scenario, it just depends what you are looking at as the target behavior, stepping off the hand or staying on the hand.
    Yes? No? (i am still trying to get my mind around the finer details of operant conditioning)

    • I agree. I think you have a good example of when both P+ and R- are happening. The toe squeezing happens only when the parrot is on the hand. It’s not randomized, like in the examples where P+ may not be occurring. It happens consistently when the parrot is on the hand and has been cued to get off. You’ve got a behavior that is decreasing because of an aversive. And the aversive continues until the parrot gets off the hand. I vote yes. Great example!

  5. Pingback: punishment… | Etymon

  6. Great blog and discussion! I am trying to think of an example of what you described above Eileen when you said “I got to wondering if you used differential negative reinforcement of an alternative behavior, whether the other behaviors would decrease from extinction or from punishment because of “leakage” of the aversive used.” If you have the time an inclination, can you think of an example for me…?? Thanks!

    • Let me think on that and get back to you. When I read the words, I couldn’t even trace my thought processes back right away. But I’ll make a note to take a look. It’s a great question.

    • OK I came up with an example. Don’t take any of this as gospel; I am thinking out loud.

      Sorry to use a shock collar but it is the easiest purveyor of negative reinforcement to talk about. Let’s say that when visitors come the dog is supposed to go to his mat instead of jumping on people at the front door, rushing around, or barking. His owner has previously taught him a “go to mat” behavior with the shock collar by a duration shock that only ends when he gets on his mat. So he knows this cue in easy situations but not when people come to the door.

      So when people come to the door, the owner cues him to go to his mat and turns the shock on. It stays on until he goes to his mat. So he might be barking, jumping, running, or just standing around when the shock first turns on. The question is whether during the learning process, these other behaviors are punished by the initiation of the shock or go extinct.

      My thought is that since shock is such a potent aversive, the behaviors that are most commonly taking place when the shock first turns on will get punished. That’s what I meant by “leakage” of the aversive. So running to the door and jumping on people may well get punished. Other possible behaviors that aren’t as common, say, standing and looking out the front window, may never be taking place when the shock goes on. So they will go extinct in this scenario but they were not punished (when go to mat has been cued).

      I am thinking out loud here, and if anyone has some more informed opinions about this, I’d love to hear them.

      The part that I can’t get my mind around is that when we have DRA using positive reinforcement, we have different competing positive reinforcers. Barking is fun, but so is going to the mat and getting great food. But I don’t know what happens when we have a previously positively reinforced behavior (jumping) competing with a negatively reinforced behavior (go to your mat to avoid shock). If extinction is happening, is the extinction process different? Can anyone think of a situation where a negatively reinforced behavior is competing with a different, previously negatively reinforced behavior?

      Thanks for a good question, Lyndsey!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s