Escape learning is based on

Reinforcement theory explains how the consequences of a behaviour make that behaviour more or less likely to recur. A response is reinforced — strengthened — either by adding a desirable consequence (positive reinforcement) or by removing an already-present unpleasant, aversive consequence (negative reinforcement). When a response removes an aversive stimulus that is already acting on the organism, the resulting learning is called escape learning; it is one of the two classic behavioural paradigms (escape and avoidance) built entirely on negative reinforcement.

In escape learning, the aversive condition — for example a loud noise, an electric shock, or an uncomfortable situation — is already present when the organism performs the response, and that response ends or terminates the aversive condition. Because the consequence of the response is the removal of something unpleasant, the future likelihood of repeating that response increases. This is exactly the operational definition of negative reinforcement, so escape learning is grounded in negative reinforcement.

This distinguishes escape learning from the other reinforcement-related ideas:

• Positive reinforcement strengthens a response by adding a desirable consequence — escape learning involves no new reward being added, only an unpleasant condition being taken away, so this idea does not describe it.

• Delayed reinforcement is only about a time gap between the response and its consequence — it says nothing about whether the consequence is an addition or a removal, so it cannot define the mechanism behind escape learning.

• Omission of reinforcement withholds an expected reward to weaken a behaviour over time — this works in the opposite direction from escape learning, which strengthens a behaviour, so it does not fit either.

Hence, escape learning is based on negative reinforcement.