Friday, November 19

Quadratic Confusion... and Morning Coffee


Mass confusion usually follows anytime I am permitted to make up the questions to test the mettle of our class on their knowledge of operant conditioning. 

I will admit, I'm seriously obnoxious. I come up with statements that intentionally force our students to think critically, using the method that my brother and I made up to analyze and figure out in which of the four quadrants the answer lies. 

Here's my most annoying example to date: 


A very cute dog walks up to his owner and paws her in the leg. The owner smiles and tells her dog to sit, a treat hidden in her palm. When the dog puts his butt on the floor, the owner smiles wider and gives the dog the treat. The dog never sits again.

WHAT?! usually follows this statement. 
Surely I screwed that up, saying that a dog that is given a treat for sitting would never sit again. That isn't possible; it's not even probable! 

Don't worry; I'm just being a jerk. I'll admit it.

I also threw in a lot of meaningless babble, such as "very cute dog," the pawing of the owner's leg, the smiling, the fact that the dog was given a treat, etc. 

Well, how do you figure out where this falls on the grid? Let's analyze and check it out!
(*Note that nowhere do I say "desired" or "undesired" behavior; I simply note that a behavior is performed. Whether or not we like it has little to do with the science.)



Forgetting anything about our preconceptions about what dogs like or should do in accordance to blahblahblah, let's take the sciency way!

(A very cute dog walks up to his owner and paws her in the leg. The owner smiles and tells her dog to sit, a treat hidden in her palm. When the dog puts his butt on the floor, the owner smiles wider and gives the dog the treat. The dog never sits again.)

1.) Identify the behavior
In this case, the behavior is sitting. 

2.) Identify the consequence.
The dog is given a treat.

3.) What happened to the behavior as a result of the consequence?
The behavior decreased.

So, something was applied that made a behavior decrease. So, if you go back up and check out the section of the grid where behaviors are decreased and then check out in which of those two something is applied, you should come up with Positive Punishment or P+.



If you're anything like the people who take my class, you're probably all but ready to jump through the computer and argue with me that a treat cannot be a form of positive punishment. 
(Funny note: That's uncritical anthropomorphism you're using; because you believe that giving a dog a treat is positive, therefore the dog must think that getting a treat is positive. This is not so, and this kind of thought process will bite you in the butt... If a dog doesn't get there first!)

So, with that note: 
Can you critically and scientifically analyze these statements and tell me which of the four quadrants they fall under?
1.) A dog barks while running agility. His handler takes him off the field and plays fetch. The dog learns not to bark on the agility field.
2.) A dog barks while running agility. The owner takes him off the field and puts him in a crate. The dog learns to bark on the agility field.
3.) A dog steps out of heel position and is collar corrected. The dog learns not to step out of heel position.
4.) The leg bands on a front pull harness loosen when a dog chooses to walk closer to his handler.


8 comments:

  1. Fascinating stuff! It's interesting the kind of assumptions we make. When forced to let these go and think critically, how easy it all seems.

    ReplyDelete
  2. Ok, you officially managed to confuse me also.

    I have not seen a dog who didn't consider getting a treat a good thing, nor I've seen a dog who would not glue their butt to the floor when they thought a treat might be coming ....?

    Perhaps just my experience?

    The only reason I could see why the dog would not wanna sit after that would be the teeth baring on part of the human ...?

    ReplyDelete
  3. By the way, in case anyone reads this: The person in question (not the person in the fake question, however) was making a pained face at her dog when he took the treat. For whatever reason, that look was found aversive and connected to the taking of the treat, which was presented upon sitting.

    The point of the article was kind of to challenge your brains. We won't always know why the dog is doing what he's doing, but it took me to be able to step back and realize that it was the presentation of the treat - any treat - by the owner that was getting noncompliance. Worse yet, she was only making that face randomly because she didn't want anyone to see her doing it. (The why of that is anyone's guess.)

    We taught him to take treats nicely, and gave her a technique to ensure that even if he didn't, her fingers would not be snapped off.

    =]

    That's it in a nutshell.

    ReplyDelete
  4. So I wasn't that far off with the baring teeth theory LOL

    ReplyDelete
  5. ooooh ooooh can I play!!!???


    1.) A dog barks while running agility. His handler takes him off the field and plays fetch. The dog learns not to bark on the agility field.

    I wanna say negative reward? As in agility's a pain for that dog and when he stops barking owner takes home away from it to play fetch? Buuuuut that is assuming that the owner systematically takes the dog off the field when it stops barking. I assume that isn't practical in agility (you have to complete the trial). So am gonna for this one instead (I know, I know, make up your mind, dearie): positive reward? Stop barking = taken off to play fetch = reward. Hang on, the timing problem's there too. Right, back for #1 option then.

    Essentially, I am turning it around. I know that it results in the dog barking less, which some people could consider punishment (less of that behaviour), but I consider not barking as a behaviour too (for which the dog gets rewarded).

    2.) A dog barks while running agility. The owner takes him off the field and puts him in a crate. The dog learns to bark on the agility field.

    This is a definite example of, aherm, negative reward or positive reward. The dog is rewarded for barking by being taken away from the agility field which, presumably in this example, the dog does not enjoy.

    3.) A dog steps out of heel position and is collar corrected. The dog learns not to step out of heel position.

    Positive punishment?

    4.) The leg bands on a front pull harness loosen when a dog chooses to walk closer to his handler.

    Negative reward (assuming the tightening harness is annoying).

    How did I do, JJ?

    ReplyDelete
  6. @Laure-Anne
    You think way too hard about these. Turn it into a math equation.
    Did you know that if I turned the squares into colors that represent the terms and use the same flow chart as above, you have an 85% (not statistical, but based on experience) chance of getting the correct answer. With words like "positive" and "negative;" "punishment" and "reward," you have around a 40% chance of success.
    40% is being kind.

    1.) Not a chance. Turn it to math, and forget the rest.
    Behavior: dog barks on field
    Consequence: dog is removed from field. (application of fetch is irrelevant. punishments must happen before 3 seconds after the conclusion of the behavior or they will not be connected. the time it would take to get off the course and start playing tennis renders the fact that tennis was played strictly moot.)
    Result: the behavior was eliminated, or decreased, as a result.
    So, it's punishment in some way.
    being removed from the floor. negative.
    P- (negative punishment)

    2.) R+ or positive reinforcment. correct.

    3.) correct. P+

    4.) You don't have to assume anything - that's the problem with analyzing these things in the first place.
    Think of it this way. If a dog is collar corrected for getting out of a sit, and the next time... he chooses to sit in order to avoid the correction, the sit is less aversive. sit=good.
    So, by that logic, if the dog is choosing to stay in heel position to avoid the tightening of the leg bands, he must find it more aversive than walking nearer to the owner.
    R- or negative reinforcement.

    You get 75% lol. =]

    ReplyDelete
  7. Many small business owners to seek out those Student Loans job opportunities.

    Your target market needs customize and specialize products or
    services. What sociological or economic factors may affect your triglyceride levels.


    Feel free to surf to my web page :: Private Student Loans for People

    ReplyDelete
  8. if you like mobile gaming then you may sure like simcity buildit game, visit the blog on simcity buildit where you can find free game tips on this game.

    ReplyDelete