Who can define these terms?

Question

The terms reinforcement and punishment get thrown around a lot here, and they way in which they're used makes me think that most people don't understand them. Who can give good definitions of these terms, in the context of learning theory, since when you're training a dog, you're using one form of learning or another. What are your definitions?

I know the answer. I'm trying to see how many out there know.

Punishment is the technical term. Correction is something different.

Yes, they are mutually exclusive terms.

Answer 1

Positive reinforcement: adding/giving something good in order to get the dog to continue doing the right behavior, such as giving the dog a treat when you say "sit" and the dog sits.

Negative reinforcement: taking away something bad in order to get the dog to continue doing the right behavior. I can't think of an example for a dog. It would be like holding someone's hand above a flame until they do something right. When they get it right, you remove the negative. Next time when faced with the flame, they will understand that by doing the right thing, they will avoid the pain.

Positive punishment: adding something bad in order to get the dog to stop doing a wrong behavior. When in a heel and the dog wanders off, you give a quick snap to the leash.

Negative punishment: taking away something good in order to get the dog to stop doing a wrong behavior, such as when you stop giving the dog attention in order for it to settle down.

Answer 2

I have a different take on the term reinforcement. Most would relate "positive reinforcement" with cookie training. I feel that it is more correct to say it was "positive reinforcement" only when it produces the desired result, and negative reinforcement is when there was nothing accomplished.

Making a dog sit in the corner solves nothing. So punishment is a complete waste of time.

I don't believe in sugar coating, so point blank, I "force train". This in no way means that I train abusively. It is just calling an apple an apple.
There is far more to Pavlov than just ringing a bell to get a cookie. As a pro trainer I fully understand that when someone pays me they want to see some results for their money. So I use ALL of Pavlov theory instead of just the part that is politically correct.
I use the term "pressure" rather than the word "correction" because in training the application of pressure has nothing to do with the dog performing the task right or wrong. It is only about teaching the dog the meaning of the correction. For instance with a trained dog a person should be able to tell the dog to speed up or slow down by just tapping their index finger against the front or back side of the lead. Turn left or right by just tapping the side of the lead. So in other words the correction it's self becomes the command.

Answer 3

Positive punishment: changes the surroundings by adding an aversive stimulus following a behaviour in order to decrease the likelihood of the behaviour occurring in the future. An example is shocking an animal whenever it pressed a lever pressing which had been previously reinforced.
Negative punishment: changes the surroundings by removing a stimulus that is a reinforcer. An example is removal of a food supply contingent on undesirable behavior.
Positive Punishment: In an attempt to decrease the likelihood of a behavior occurring in the future, an operant response is followed by the presentation of an aversive stimulus. This is positive punishment. ex. If you stroke a cat's fur in a manner that the cat finds unpleasant, the cat may attempt to bite you. Therefore, the presentation of the cat's bite will act as a positive punisher and decrease the likelihood that you will stroke the cat in that same manner in the future.
Negative Punishment: In an attempt to decrease the likelihood of a behavior occurring in the future, an operant response is followed by the removal of an appetitive stimulus. This is negative punishment. ex. When a child "talks back" to his/her mother, the child may lose the privilege of watching her favorite television program. Therefore, the loss of viewing privileges will act as a negative punisher and decrease the likelihood of the child talking back in the future.

Basically though, the best way to train a dog is rewarding them when they do good (by saying yes and giving them a treat) and ignoring bad behaviour (ex. if you were walking by something outside and they wanted to go see it, just keep walking and ignore the fact that they are trying to get over there) or just saying no or making a noise to distract them from what they are doing.

Answer 4

Let's see if I can remember how this goes without looking it up.

Reinforcement is something done or not done to increase the motivation for repeating a behavior. Punishment is something done or not done to discourage a behavior.

Examples of positive reinforcement would be praise, petting, food rewards. Negative reinforcement would be the *removal* of something unpleasant, like an ear pinch.

Positive punishment could be a reprimand, collar jerk, slap. Negative punishment would be the removal or withholding of something the dog finds desireable (no praise, no treat).

Answer 5

They're not exclusive terms. Punishment is a tool of reinforcement.

When you reinforce an action, you are trying to get the subject to repeat the action or series of actions (push a lever, run a maze) or stop doing an action. There are two ways of doing this. Reward and punishment.

Answer 6

Demand Characteristics is when the participant gives sociably desirable answers to conform with what is expected of them. For example, if the participant was given a questionnaire they may tick answers that would made them look good to the researcher or answers that are what it considered to be "normal" or what they think the researcher wants. Investigator effects are things in the experiment that are caused by the researcher such as being biased towards certain participants. The single blind technique is when the researcher knows the aim of the experiment and what is happening such as control groups but the participants do not know so that demand characteristics are less likely but investigator effects can still happen. The double blind technique is when neither the researcher or the participant know the aim of the study and therefore demand characteristics and investigator effects are cut down upon.

Answer 7

In psychology, reinforcement is something you do to make an action by someone or something more common. There is positive reinforcement ("Good job! Have a cookie") and negative reinforcement ("If you don't get an A, you are grounded.")

Each have their place. Works for people and for animals. Reinforcement could ba actions, words, or even just a smile or frown.

Answer 8

to me, reinforcement means to positively reward proper behavior as soon as it happens to keep the behavior going. Punishment is just that. When bad behavior happens a consequence must happen imediately that is unpleasant and more unpleasant than the good feeling they get from the bad behavior, my favorite consequence is the swift put down at the neck just like the Apha leader in a pack does it.

Answer 9

reinforcement is either positive or negative in my understanding of training. Punishment I think is not appropriate wording at all, as you cannot punish a dog. Correction is a better wording.

Answer 10

With myself and my dog, Reinforcement is immediate reward following correct behavior and punishment is a brief firm acknowledgement of wrong behavior followed by a void of reward.