How we change what others think, feel, believe and do
A conditioned reinforcer usually is a signal of approval that lets the subject know that their actions are correct. It may also signal that they should keep going and not stop for a reward. In its broadest sense, it can be anything that is learned to reliably gives the subject any feeling, including via an accidental stimuli.
The signal can be anything that is sensed, including a word, phrase, sound. Initially, the signal means nothing and may be lost in the background of environmental noise. In animal training this can be words like 'good boy' and the animal's name. It can also be a whistle, a clicker or toys.
Setting up a conditioned reinforcer is a very useful initial activity when training an animal. Do this by pairing it with a positive reinforcer. For example you can connect 'good' and food by repeatedly saying 'good' while giving them a morsel. Setting up this reinforcer may take significant effort, but with persistence it can be reliably instilled. The strength of the reinforcer can be increased by pairing it with highly desirable rewards such as tasty food.
A school teacher regularly says 'well done' when they see their pupils working well. The students know that this means they should continue and not stop.
A horse rider makes clicking sounds to let the horse know that it is cantering at a good speed.
A dog owner says 'hello' when they come home. The word leads to wagging of the dog's tail whenever they hear the word.
It is not always possible to give a positive reinforcer to a subject while they are doing what you want them to do, even though this is the ideal of positive reinforcement. Typically this is because they would have to stop what they are doing to receive the reinforcement. This is particularly true when training a sequence of actions or where an action takes a while to complete. The conditioned reinforcer allows you to send a signal of approval without stopping to give a reward.
The conditioned reinforcer is also called a 'secondary reinforcer'. A 'primary reinforcer' (also an 'unconditioned reinforcer') is something that is already liked, such as food. The secondary nature of the conditioned reinforcer is because it is new, but has a similar effect to the primary reinforcer as something that the subject comes to like just for its own sake.
In its broadest sense, a conditioned reinforcer is any learned stimulus. It is a reinforcement that has been conditioned. In his original work with dogs, Pavlov noticed that they salivated when they saw researchers wearing white lab coats. The coats had become conditioned reinforcers as they triggered desire in the same way as food. An important factor about this is that anything can become a conditioned reinforcer, often through being paired with a primary reinforcer.
Conditioned reinforcers can also be negative (sometimes called 'negative conditioned punishers'), for example when a dog associates its owner preparing to go out with fear of being left alone.