Shaping is used when a behavior is NOT in a learner’s repertoire but we want to provide them reinforcement for their best attempt to engage in the behavior. For example, if we are working with a two year old child with severe language delay to request “bubbles,” it could take months or even years for them to be able to say “bubbles.” But, that DOES NOT mean, we do not want to reinforce their attempts to request until they can say the terminal behavior (bubbles). If the best they can do is to say “uh,” we would accept “uh” and provide them bubbles when they say “uh.” Our goal remains for them to say “bubbles” so when they can say something closer like “buh” we stop reinforcing “uh” and reinforce “buh.” We keep doing this until they can engage in target behavior “bubbles.”
Many people confuse shaping with chaining because they believe there are “steps.” However there are no steps. We are always simply reinforcing the child’s best response to the target behavior. For example, the child might jump from saying “buh” to “buh-uh” or “buh-buh.” We do not predetermine the successive approximations before they occur.
This example shows shaping across topographies. That means each successive approximation of the behavior looks different. Other examples might include a baby learning to run. First they crawl, then they take a few steps, then they walk and then they run. We use shaping across topographies when the learner cannot engage in the form of the terminal behavior.
Sometimes the learner can engage in the form of the terminal behavior but cannot do a sufficient magnitude of the behavior for it to function. For example, a child can hold a pencil and write, but cannot press hard enough to produce a legible response. Another example might be, a person can swing a golf club but cannot get the ball far enough to make a shot. Then we would use shaping within topographies. This means we want the behavior to look the same but do it harder, faster or stronger. Just like with shaping across topographies, we always reinforce the best approximation of the terminal behavior at the time. If the farthest a person can hit the golf ball is 100 feet, we provide them reinforcement for hitting it 100 feet. But if one day, they hit it 150 feet, we stop reinforcing 100 feet and only reinforce them hitting it 150 feet until the terminal behavior (final goal) is achieved.
Citation: Cooper, J. O., Heron, T. E., & Heward, W. L. (2019). Applied Behavior Analysis (3rd Edition). Hoboken, NJ: Pearson Education.
