ive had this discussion before on a theoretical level. disclaimer, i have not done clicker training before so i’m not experienced in that, however, on a daily basis I do intervention based on the principles of applied behavior analysis so this covers clicker training.
We should go over the basics so we have a fundamental understanding of what is going on. Firstly, clicker training is operant conditioning (Skinner box, not Pavlov dogs). Basically in a nutshell, it is the learning through consequences. The basic learning paradigm is A-B-C, the 3 term contingency. A is the antecedent or what comes before the behavior. B is the target behavior you are trying to teach and C is the consequence (either reinforcement or punishment).
The premise of this is that you will give a command (A), then rabbit will do the behavior of choice (B), and then you will provide a reinforcement, a treat (C). When you do this reliably, the consequence (C), will increase the likely hood of the behavior (B) from occurring. This is how you will teach a behavior or a skill. Using a layman’s example, Your boss tells you to go to work on time (A), you get to work on time (B), your boss gives you $100 bonus (C). If you value the bonus (C), then tomorrow when you are getting ready to leave, you are more likely to leave and get there on time. Now flip this around, instead of getting a bonus, instead your boss embarrasses you in front of everyone. Now in the future, you will be less likely to get there on time. Makes sense? I know its not really “realistic” but thats just an example.
How does click training come into play? If you’ve done a token economy or a token chart, then clicker training is just that. If you havent, well some background. Primary reinforcers are things that have value and will always have value with the exception of getting too much of it (then it decreases, but will come back after some time off). Examples are food, water, sleep. A secondary reinforcer is something that does not have value, but must be paired with a primary reinforcer and will take on value, it will lose value if you stop pairing it. Examples are praise, tokens, money (yes money has no “inherent” value), clicks, etc. What needs to be done is to pair the click with the reinforcer (C).
For example: Tell bun to sit down (A)—> Bun sits (B) —> deliver treat (C) + click
what happens is that the click becomes associated with the reinforcer (treat) and therefore you can fade out the delivery of treats. Everytime you click, it should have some reinforcing value. However, you must put the treat on an intermittent schedule so that the click can always retain value.