Dan Goldstein, Dog Whisperer LLC, Avon, CO asks:
Hey Kayce! I was hoping to get your viewpoint on a topic that I’m having trouble explaining to clients – the difference between classical conditioning vs. name & explain. I tell clients to use verbal markers as “predictors” in order to achieve classical conditioning (ex. Click, treat). But then later say that they can “capture” behaviors/teach the dog what it’s doing at the time of the behavior by naming it for them (“simultaneous” rather than “before/predictor”). Then we both get confused about how these two principles operate in relation to each other. Can you help clarify this concept for me? I hope you are well!
Kayce Cover answers:
Dan, I am doing great, thanks!
We are talking about two different kinds of information. A bridge (marker) tells the animal “that pays!”. It is a signal with little variability and is therefore fast to process. Bridges tell the animal to mark what is happening at this instant, but they don’t convey any information except that that instant brings payment. *
The words (language) give information about the environment and action relevant to the animal. They bring the attention of the animal to the specifics of what got paid, what distractions are looming, what the training goals are, and more. Words are not as simple as bridges and must be decoded. The animal must think about how the words pertain to what is going on.
Both bridges/markers and language can be given verbally, but they have different roles and are processed differently. Both are powerful So, the IB gives rich, immediate feedback, and is a reinforcer. In classical and operant conditioning, it is known as a special case of an Sd (Discriminative Stimulus) – the CS, or “conditioned stimulus”. That just means, it is a signal that has gained meaning by being paired with another event, in this case, food. The name and explain is different. it is not comprised of verbal markers. It gives INFORMATION. Classical and operant conditioning are all about unconscious learning. Behavior analysts argue that no conscious thought is required to change behavior, and operant conditioning gives no information to the animal to guide his efforts in training. The inefficiency of this practice has resulted in many people luring an animal into doing what is wanted, so that bridging can occur. In SATS, we do it differently. We don’t use lures. We use targets to demonstrate a behavior, and words to give additional information about the behavior and the environment in which the animal is working.
*modulated intermediate bridges do give additional information, which is “change is coming – look sharp”
AND
Wade MacVicar, wadesdogmanship.com asks:
I have a quote from you in it right after I talk about the 4 Quadrants, which I’d like to give you credit for.
“The 4 Quadrants of operant conditioning are incomplete because they leave no room for the stimulation to be classified as ‘information’”
Am I accurate in this quote?
Kayce Cover answers:
Thanks for the attribution, Wade!
Information delivery is not accounted for in the four quadrants analysis. The “Four Quadrants” only considers reinforcement and punishment. Information is not generally considered a source of reinforcement or punishment. I think it can be. Consider how you feel when you solve a puzzle or finish a good book or anticipate a favorite event). When I tell my dog where I placed his food bowl, his tail starts to wag immediately If I tell him he will get pumpkin, his tail starts to wag immediately. I tell him what is happening and he reacts congruently, showing excitement and pleasure, even when there is no other sign of feeding. By the way, we feed at varied times, so there is no expectation of food at a particular time.
It might be convenient to overlook information delivery as an important aspect of motivating animals to learn, but really folks, training is ALL ABOUT the delivery of information. Period. And here is the piece de resistance: I can often motivate animals to learn with NO tangible primary reinforcers or diminishers (“punishers”, in everyday language). I am not the only trainer who does this. Every trainer who has mastered Perception Modification does the same thing. We are working on a different playing field. It is a game where information. collaboration, and shared adventure are the currency of commerce. In general, we use, no food, no toys, no tugs, no physical corrections, in Perception Modification. Check it out! HERE!
Training without telling the animal what we are trying to accomplish is less effective, less efficient, and less motivating, than telling, and showing, the animal what we are working on. Even better, also tell him WHY we are working on this. Show him how this new skill we are learning will make his life better! For example, Rocky, when learning to give a ball to hand, is told that the ball must be in my hand for me to throw it for him. When he puts it in my hand, I then follow through, showing how this skill directly improves his life.
Always, don’t take my word for anything. TEST IT for yourself. Always.
Back to
Dan Goldstein·
In “pedagogy,” the study of teaching, Name & Explain is recognized as one of the most powerful strategies of education: creation of anticipation. In other words, we recognize how critical it is to excite a student about what we are to study, and alert the student to what information is critical to our goals. This is, in a nutshell, Name & Explain! We cannot actually teach anybody anything. We can only facilitate their learning. We must inspire human and animal students to engage in the process of learning something.
One of the most effective ways to inspire and facilitate learning is to create anticipation. Anticipation spurs learning in two important ways that I know of.
1) Anticipation raises a sense of suspense, excitement and curiosity. Over and over again, people miss fascinating and beautiful things, because they are not ‘interested’ in them.
2) It creates the interest that was missing, by telling the student what critical information will soon be of particular use. For example, if I arrive at a meeting with you, and say “tell me how many red cars you passed on the way here, and you get $1000 for the correct answer,” you will probably just be frustrated at me because there is little chance you will get that answer right because you were not paying attention to red cars on your way to our meeting, much less counting them. Better, I call you before our meeting and say, “count the red cars you pass on the way to our meeting, and if your answer is correct, you can have $1000.” NOW you are looking forward to this opportunity to learn about numbers of red cars, and you are looking forward to seeing how accurate your count was.
Find a link to a well written, clear, concise training plan for Operant Conditioning Here. It shows how to break requirements into steps. It does not tell how to demonstrate or explain the requirements, as we do.
I will write more about how the operant conditioning approach to training differs from the SATS approach soon. Please stay tuned.
A note about “science based” or “evidence based” training. It just means that a particular training method employs insights gain from research studies. However, there is still much to be studied and explained in training. For example, in 1990, when I wrote my manual, “Introduction to Bridge and Target Technique”, there was no study of targeting as a means of communicating task information to animals. There was no definition of targeting, in the scientific literature. “Target” was an adjective describing the lever mechanism in the Skinner box.* . Studies would say, ‘the target manipulandum…” meaning the manipulandum the animal needed to target to activate the food hopper. Nonetheless, targeting existed. It prevailed. All the professional trainers with whom I worked used targets with elegant sophistication and amazing speed to ‘draw’ behaviors in the air, for animals to see. For these trainers, a target is an extension of the hand which allows us to touch the animal in places we cannot reach, or five places at one, or 50 feet away. We can use targets to show, in a process like ‘connecting the dots’ exactly where the body needed to be and how it needed to move through space, in order to complete the training task. The trainers were fluent with using these tools to talk to animals. It is a shame that many new and upcoming trainers have abandoned targeting, for luring and free-shaping. This is apparently because targeting was not studied by science. These regressed the speed and efficiency of their work because they denied the animals the task information provided by the target.
Nothing in life is ‘scientific’. It just is. Training just is. Training happens when “reliable task performance on cue” happens. (Teaching can be seen as the transfer of information, training can be seen as the transfer of cued performance.) As a trainer, I would never want to confine myself to tools that have been validated by science. It would regress my training to a boring waiting game compared to what it can be with the many developments that are yet unstudied by science. These tools include the Intermediate Bridge, Name & Explain,
*Skinner box: A box used to house an animal temporarily during experiments testing an animal’s willingness to work by pressing a lever/manipulandum in order to obtain a consequence, such as sip of sugar water. These are often 1 foot square, made of metal mesh, and hang from a rack of cages with multiple animals, with the lever and feeder at the front of the set up so researchers can observe and access the animal and equipment.
Thanks for the great questions, guys! Keep them coming!