|
Bailey & Bailey - Suggestions for better training
Training is a mechanical skill: Practice training procedures without the animal. Practice clicking. Practice giving the food, toy, or praise. Practice moving and training at the same time. Practice, practice, practice. Then, train the animal.
Think -- Plan -- Do (three separate behaviors, don't mix them): First let your imagination soar; consider every training idea without concern for practicality. Next, filter out the wild, impractical schemes and build a detailed doable pathway to the desired behavior. Finally, carry out your plan.
Training is simple, but not easy: Training should be the application of simple principles. Your demands on the animal, plus the animal's own wants and needs, can make training complex. The more complex the behavioral task, the more vague your direction, the more complicated the training. Simplify each step of training, give unambiguous direction and precise reinforcements. Most trainers should be behavioral "splitters" rather than "lumpers."
Solving behavioral problems: Look for simple solutions rather than complexity -- consider first problems of Timing, Criteria, Rate of reinforcement
TIMING: In casual training, clicker timing is forgiving. In precision training, clicker timing is NOT forgiving. You get what you click, not what you want. Imprecise timing of reinforcement (clicking) is, by far, the most frequent trainer error, in the experience of the Brelands and the Baileys.
CRITERIA: Trainers should decide, in advance of the behavior, what will and will not be reinforced. If the trainer waits until the behavior occurs before deciding, the click will be invariably late and inconsistent.
RATE OF REINFORCEMENT: Training must be worthwhile for all -- for the animal, and for the trainer. The animal must get something it wants or needs. The trainer must get behavior or some other satisfaction out of training. Reinforce often, especially early in training.
RATIOS: Don't use a ratio in training unless you need to- the trainer has a choice. Ratios are overused and often improperly used. Properly used continuous reinforcement, plus fluency training, give moderately strong and very precise behavior. Reserve ratios for highly repetitive, long duration or other special behaviors.
PRIMARY REINFORCEMENT AND REINFORCERS: Food, play, social, and other stimuli filling a need or want may be useful in training. The primary reinforcer strengthens behavior. It can matter what, when, and how the primary reinforcement is delivered. As much thought should go into the delivery of the primary reinforcer as goes into the secondary, or bridge. Generalization: Click For Action, Feed For Position.
|
|
