Tuesday, November 8, 2011

Understanding Reinforcement

As I have mentioned several times, I have 2 loves in my life, my job and my dog. As for my job, I am a  behaviour analyst who works with individuals with autism and related disorders. My career path, studying and understanding behaviour and behaviour change principles, has given me an in depth knowledge of all the principles that govern behaviour in human and non-human species (a lot of research is still done with animals).

Given my unique background, I often forget that not everyone understands and uses reinforcement with their pets. I was reminded of this in one of my favourite blog's, Furry Four Legged Foster Friends http://www.furryfosters.net/, post yesterday. Mel talked about having a "life changing" moment when being at a talk that mentioned reinforcement. This has made me think I should write about reinforcement.

Reinforcement is anything that INCREASES the chances of the behaviour in the future.

There are 2 types of reinforcement:
Positive Reinforcement - adding something to the environment that increases the future of a behaviour in the future (in dog terms this likely is food, toys, pets, praise - for most of us it is $$$)
Negative Reinforcement - removing something from the environment that increases the future of a behaviour in the future (For my cat Bagheera, smacking Lee in the face when she comes to close allows him to stop/remove her butt sniffs - in human terms, the best example, is finally cleaning your room so your mom stops nagging you).

Punishment on the other hand is anything that DECREASES the chances of the behaviour in the future. (Again there is positive and negative, but I won't go into that now).

Common Challenges with Using Reinforcement:
1) You only know something is a reinforcer at another time in the future (do we see the behaviour increase later?) - just because the creature involved seems to enjoy it, it is only a reinforcer if later you see the behaviour increase.

2) Sometimes what we think are punishers actually are reinforcers: Sometimes (especially in my work with kids), we think we are punishing someone with a reprimand for example, but if the behaviour doesn't changes, or worse increases, well, the reprimand probably was a reinforcer!

3) Just because something is a reinforcer right now, it may not be in 2 minutes or tomorrow. The science explains the concepts of deprivation and satiation. Deprivation is when you haven't had something for a while and the value of it is much higher because you have been deprived (think about a long car ride and how reinforcing access to a bathroom is after it as compared to the bathroom's normal value). Satiation is when you have had too much of something and the reinforcing value of it declines (think of the value of food after a holiday dinner pig out).

4) New behaviours may need to have access to the reinforcer a different, undetermined number of times that you can't know before you start:) So giving your dog a cookie once when they sit doesn't mean they will sit in the future. They may need to be reinforced for this behaviour many times.

Hints for using reinforcement with your dog:
1) You need to reinforce any behaviour you want to see more of!

2) If you are seeing more of a behaviour you don't want to see, you need to figure out what is the reinforcer that is maintaining this behaviour and stop it! (Easier said then done!)

3) Have a variety of potential reinforcers/preferred items. By mixing it up, you can create deprivation and avoid satiation.  People laugh at the number of different treats/toys I have for Lee, but I do it to allow me to manipulate her motivation to get her to do what I want/need her to do.

4) Save the most powerful potential reinforcer/preferred items for specific behaviours.  Lee only gets access to her current favourite, beef liver, when I am doing her nail desensitization
program.

5) Often very small pieces can be reinforcing. By using small pieces, you can reinforce more often without satiation.

6) Reinforce often!

Happy Reinforcing!


3 comments:

  1. Yikes. It scares me that people still don't know this.

    One of the better discussions of this I've read is http://www.clickersolutions.com/articles/2001/ocguide.htm

    From what I've seen, the best trainers out there work exclusively with positive reinforcement, and lots of it.

    There are still "old school" trainers out there who pride themselves on never resorting to "crutches" like food or toys. Avoid them, please.

    ReplyDelete
  2. Regarding the above comment:

    With so much information available on the internet it's hard to figure out who is right and who is wrong as far as training methods, and unfortunately not everyone has the money to try a ton of trainers before finding the right one.

    Thankfully I happened to attend the right class and had the opportunity to listen to two very educated trainers/behaviorists.

    ReplyDelete
  3. It really doesn't help people that many of the popular dog training shows are done by people who do NOT use reinforcement.

    I feel lucky to understand how it works, even if sometimes I reinforce behaviours I shouldn't. At least I know I am doing it! (And some days when I get home, I am tired of using behaviour change procedures LOL!)

    ReplyDelete

Thanks for reading and leaving us a comment. We love your comments!