What's the difference between a ski lesson and a puppy class?

February 27, 2012 2:03 PM

To what extent is it useful to apply dog training techniques to teaching people?

I've been thinking about the differences in teaching practices between my friend and I. She teaches physical skills to dogs. I teach physical skills to people (of all ages).

The more I teach, the more my lessons seem to have in common with dog training: I find that "tricking" people into doing the movement I'd like and immediately "rewarding" with praise or high fives, is usually more effective than explaining things. Particularly with children, of course, but also with adults.

I'm very interested to find out more about the science behind learning, specifically the differences between the "dog training" kind of learning and the more intellectual kind. I vaguely remember reading a book that referred to our "lizard brains" - being the primitive part of the brain that responds to dog training techniques - but I somehow doubt this is the technical term!

I'd love to know how other teachers of physical skills navigate these waters and whether there is some state-of-the-art opinion on when and how to employ which technique

I'm also interested in discussions about using techniques similar to dog training as a form of self-coaching when training a physical skill.

Unfortunately I don't know enough about the science of teaching to know the buzzwords to search for to find out more myself!

Can you help?

posted by emilyw to Education (22 answers total)

You might be interested in the NYTimes article What Shamu Taught me About Marriage which discusses using positive feedback to establish behaviors. This is a pop psychology take on some of what you're talking about.
posted by sciencegeek at 2:14 PM on February 27, 2012 [4 favorites]

I have enjoyed this comment on AskMe and referred to it many times. I had assumed it was from somewhere else and it may be but not under that name.
posted by jessamyn at 2:21 PM on February 27, 2012

I am actually working on a fpp on the science of learning. I haven't written the fpp yet but I could mail you the links.
posted by Foci for Analysis at 2:22 PM on February 27, 2012

Don't Shoot the Dog!: The New Art of Teaching and Training will change your life.

(Or at least reinforce what you have already noticed, that teaching physical skills through this kind of conditioning is fundamentally different than trying to explain it in words.)
posted by thehmsbeagle at 2:25 PM on February 27, 2012 [1 favorite]

Could you elaborate a bit on what you mean by "tricking" someone into doing something physical? My bachelor's is in physical education, so I have a lot of book learnin' in the area of of teaching physical skills to others, but the concept of tricking someone into performing a certain movement isn't something I've heard about.
posted by SuperSquirrel at 2:30 PM on February 27, 2012

It is called Behavior Modification and the overarching principals are the same - reward behavior to increase it, ignore behavior to decrease unless the behavior is destructive and you cannot reward opposite behavior in which case you punish behavior. In other words from best to worst way to increase a behavior-
positive reinforcement
extention (ignore)
negative reinforcement
Wikipedia entry:
Behavior modification is the use of empirically demonstrated behavior change techniques to increase or decrease the frequency of behaviors, such as altering an individual's behaviors and reactions to stimuli through positive and negative reinforcement of adaptive behavior and/or the reduction of behavior through its extinction, punishment and/or satiation. Most behavior modification programs currently used are those based on Applied behavior analysis (ABA), formerly known as the experimental analysis of behavior which was pioneered by B. F. Skinner.
posted by shaarog at 2:43 PM on February 27, 2012

I have seen sites on the net thag suggest this sort of dog training can be useful for young and new wives to apply to husbands.
posted by Postroad at 2:52 PM on February 27, 2012

The thing with behavior modification and physical skills is that behavioral mod usually implies some sort of choice to perform that activity or not, right? (I'm not well-versed in all its intricacies.)

But learning a physical skill is different. Let's say the skill in question is swinging a baseball bat correctly so that you can hit a pitched ball. No amount of rewarding a beginner is going to increase their skill level at this, because there is no choice involved in hitting a baseball, so there isn't really anything to reward. At first it's luck, and then with practice, it becomes a skill.

I suppose one could say that actually hitting a ball and seeing the ball travel away from you toward the field is a kind of reward. But the pleasant feeling associated with succeeding that time isn't going to influence your ability to hit the ball again on the next pitch, unlike Skinner rats who associate the pleasant feeling with something external, and learn to adjust their behavior to force that external trigger to repeat.

That said, rewarding a beginner for sticking with practice is different - that's rewarding a choice they've made, not the performance of the skill itself.
posted by SuperSquirrel at 3:13 PM on February 27, 2012

Just for fun, check out how Sandra Dee used a dog training book to train her husband in the 1962 movie If a Man Answers.
posted by CathyG at 3:32 PM on February 27, 2012

Response by poster: SuperSquirrel: For example, if I want a child to lean forward more while skiing, we will play at being rabbits digging holes in the ground while we are skiing (This is incredible fun when you are four). If I want an adult to lean forward more, I get them to make a series of tiny jumps while skiing. After a few wobbles they will correct their posture on their own.

In neither of these cases will the learner have any self awareness of the posture changes, until I start doing the Happy Ski Instructor Dance, complete with high fives, directed positive feedback, etc.
posted by emilyw at 3:37 PM on February 27, 2012

You mean like this?
Sheldon Trains Penny
posted by tuesdayschild at 3:41 PM on February 27, 2012 [2 favorites]

I disagree with SuperSquirrel - most physical skill is learned on a intuitive, nonverbal level. What the trainer wants to do is called "shaping" - reward any random event that happens to be somewhat closer to the desired goal than average. Once you get that preferred behavior more often than not,t hen you raise the bar and only reward the best examples, even they are still far from the desired goal.

While the person may not know what done right, he will gradually get a felt sense of what works and what doesn't. This is especially important if there isn't any natural feedback for me to use. If I miss the baseball every time, I have no idea which swings are closer to being right. If the trainer praises for me for swinging at the right time (but too low) or dead level (but too late) I begin to get an idea of what to do more of.

The pleasant feeling if hitting a ball IS a reward. So if you practice free throws there is an automatic feedback - the reward to seeing it go in and (assuming a reasonable number go in) the difference between ones that bounce in and one that drop cleanly. So eventually, you adjust to do more of what works even if you couldn't ever put it into words. However, if you can more detailed feedback, you can improve more quickly.
posted by metahawk at 3:41 PM on February 27, 2012 [1 favorite]

YOu might find this article interesting on how thinking to much gets in the way of muscle memory.
posted by metahawk at 3:47 PM on February 27, 2012

Response by poster: and, Foci for Analysis, I would love to see the links from your proto-FPP.
posted by emilyw at 3:52 PM on February 27, 2012

You might read about operant reinforcement, or Applied Behavior Analysis (ABA) in general. It doesn't speak to the tricking them part, but it does discuss how to use rewards.
posted by juliplease at 3:54 PM on February 27, 2012

Actually, metahawk, I think we agree. :) I was responding to shaarog's mention of classic Skinner-style behavior modification, which I don't believe is what your example demonstrates. (Again, not trained in Skinner, or anything like that.) I was thinking of this scenario as true behavioral modification: Every time you make contact with the baseball, I give you a cookie. What I am rewarding is your persistence (your choice to continue trying to hit or not to continue trying to hit). I'm not rewarding the physical act of hitting the baseball. But maybe the outcome is the same, regardless, so I guess we agree on that.

metahawk: "What the trainer wants to do is called "shaping" - reward any random event that happens to be somewhat closer to the desired goal than average.

This is exactly the technique I learned in school, especially for complex activities that are made up of a series of smaller behaviors, like hitting a baseball or doing a layup in basketball. We never used the term "behavior modification" to refer to it though. We did indeed call it shaping.

emilyw, the example you provide seems like it would be a form of transfer of learning, specifically motor transfer. (There are a lot of good search terms in that Wikipedia article that might help you find more info.)
posted by SuperSquirrel at 4:24 PM on February 27, 2012

I believe that TAG teaching is exactly what you're looking for.
posted by HotToddy at 4:28 PM on February 27, 2012

emilyw, you'll probably find a lot of good stuff in the various journals published by the American Alliance for Health, Physical Education, Recreation and Dance (AAHPERD), which is one of the professional organizations for physical education teachers and coaches.
posted by SuperSquirrel at 4:30 PM on February 27, 2012

Seconding that you will really like Don't Shoot the Dog!

All talented teachers rely heavily on positive reinforcement. People love hearing that "Yes!" or "Exactly!" or whatever marker good teachers use to mark and reward good answers.

In my life, I use positive reinforcement far more with people than my animals. It isn't trickery -- it is helping people learn, adopt strategies that work, and remember things that they need to know. Really, why shouldn't people be praised when they get it right?

The other side of this, which Pryor and others who write about operant conditioning point out, is that punishment, negativity, and sanctions, though very commonly used to discourage unwanted behaviors, are far less effective. These techniques can work, but they can also backfire. I.e., with positive conditioning you are extremely likely to see the desired behavior, quickly and reliably, while with negative or punitive reinforcement, you don't know what the heck may happen.

One last very cool thing: you can use positive reinforcement to extinguish unwanted behaviors, too. By rewarding their absence. Works a treat.
posted by bearwife at 4:57 PM on February 27, 2012

Whoops, I also meant to give you this link to TAGTeach International.
posted by HotToddy at 5:21 PM on February 27, 2012

Back in the day when I was teaching piano and other music lessons for many hours per week (and also training other piano teachers), I had a little rule of them that went, "If you can find a way to teach something to X year olds, then it will work from X on up."

So in other words, if you can find a way to teach a 14 year old how to do some certain technique, then that teaching method will likely work from 14 year olds on up to adults.

If you find a way to teach 8 year olds, then it will work from 8 year olds on up to adults.

Same with 6 year olds, 4 year olds, etc.

This does NOT work in reverse--ie, you may find a technique that works perfectly for 12 year olds (and up!) but discover that it fails miserably for 10 year olds, 8 year olds, and so on. But if you find a technique that works for your 6 year old students it will most likely work on the 8, 10, and 12 year olds as well--and the adults, too!

(There are differences between older and younger students, of course, and even if using the same general teaching technique for a young and and old student, you will of course apply it differently. Most commonly, older students will be able to advance from step to step or technique to technique at a much faster rate than younger students. So in teaching a 12 year old and a five year old, you might follow the same general approach sequence of activities, but the five year old might take a month to step through them, taking them carefully one at a time and allowing days of rehearsal in between each step, while the 12 year old covers the same sequence in 15 or 20 minutes.)

Now that is obviously a bit of a rule of thumb and also meant somewhat tongue in cheek. But there is a lot of truth to it as well. We gain cognitive and physical skills as we grow and mature, but by and large, we keep all the senses, coordinations, and physical and mental skills we had at an earlier age and just keep adding to them as we mature.

So our human process of maturing is mostly a process of adding to, and only rarely and a few ways do we lose senses and abilities that we had before. So the way we learned when we were 2, still works now, and so does the way we learned when we were 4, and so on up.

And on this scale, the dog is something like a 1 year old--or maybe a 2 year old?

So yeah--anything that works on the dog will most likely work on almost any human older than 1 or 2. And sometimes, it might even work better on the human than the dog!
posted by flug at 6:51 PM on February 27, 2012 [1 favorite]

+1 Don't Shoot The Dog. It's an excellent book.

The author has a passage in there about how her daughter organized her grades school play. When the principal commented on what a great job her daughter was doing getting everyone motivated & working the mother replied "Of course she did a good job, she's a dog trainer." To which the principal balked - the idea of treating children like animals appalled him, but on a fundamental level there should be no difference except that you can communicate even easier with people.

I've clicker trained my cats & comprehension varies from cat to cat (connecting the action with the click with the treat), but they all take to it eventually.

Come to think of it, my coworker constantly brings candy to work - I wonder if this is a conscious attempt to curry favor. The most candy goes to the guy he gets the most favors from. Hmmm.
posted by MesoFilter at 7:43 PM on February 28, 2012