1) A method used to train a chained performance. The basic idea is to first train behaviour that is closest to primary reinforcement; once responding is established, links in the chain that are farther and farther from primary reinforcement are added. Each link in the chain is reinforced by the SD (which is also a conditioned reinforcer) that signals the next component in the sequence.

2) The establishment of the final link in a stimulus-response chain, with the addition of successive links until the first link is acquired.

3) A chaining procedure that begins with the last element in the chain and progresses to the first element.