Dear Mr. Stallings,
This commentary is being shared in response to your article, “Teacher to Parent – Positive Reinforcement Doesn’t Work in the Long Run.” In the article, you assert that positive reinforcement is something that does not work in middle school and high school, and that extrinsic rewards are not an effective way to instill good conduct. You state that one of the problems is that rewards for good behavior can’t keep pace with children’s changing desires. You also describe how when an extrinsic reward is removed, the behavior you want to maintain doesn’t always stick.
Thank you for bringing attention to such an important topic that is often misunderstood: reinforcement. Reinforcement is defined as the response-contingent presentation or removal of a stimulus that results in an increased frequency of behavior in the future. Oppositely, punishment is the response-contingent presentation or removal of a stimulus that results in a decrease in the behavior (Cooper, Heron, & Heward, 2007). Human beings may be influenced by reinforcement and punishment in everything they do—people behave due to previously reinforcing or punishing consequences. We avoid a particular route home from work because of previously experienced traffic (punishment); we continue to speak in a meeting when a colleague smiles and nods her head (reinforcement).
Reinforcement can occur in four ways: by gaining access to a preferred item or activity, by escaping or avoiding a non-preferred situation, by gaining access to some form of social attention, or through automatic or sensory enjoyment of an environmental stimulus (Cooper, Heron, & Heward, 2007). Thus, you are correct when you claim that extrinsic rewards are not always effective for “instilling good conduct.” We suspect that you are paid for your work in US dollars and this form of extrinsic reinforcement is motivating to you whereas payment in the form of arcade tokens or baseball cards would not be. Clearly some reinforcers are potent and others are not!
To plan ethical and effective ways of improving behavior, we must consider the variety of ways behavior might be reinforced. Essential to analyzing and changing behavior is examining the function of the current behavior: why is the person is doing what he or she is doing? We must also identify what the individual’s unique reinforcers are: while listening to heavy metal music might be reinforcing for one person, it might be quite aversive and punishing for another. Once we know the function of a specific behavior and the person’s individual motivators, we can create an informed plan, implement the procedures, and monitor the effectiveness of our intervention by collecting data.
Another important point you make is that when extrinsic rewards are removed, the desired behavior doesn’t always continue. When you describe a scenario where a man stops providing reinforcement abruptly and desired behavior immediately stops, you refer to a situation marked by “ratio strain.” This is a situation where reinforcement is removed too quickly, and the desired response ceases to occur (Cooper, Heron, & Heward, 2007). An essential component of effective behavior change programs incorporates gradual removal of a contrived reinforcer as naturally occurring consequences take its place. For example, a parent might give their child $5 every week for practicing guitar. Eventually, the child may find playing the guitar intrinsically reinforcing. Thus, the behavior will be maintained when the extrinsic reward is slowly decreased by $1, and finally eliminated altogether.
Behavior analysis is a detailed, and at times complicated, science, but it is our hope that this response helps explain some of the common confusion regarding reinforcement and rewards. As has been highlighted, reinforcement is certainly not synonymous with extrinsic rewards, and care must be taken to utilize effective reinforcers based on the individual’s unique preferences and interests.
One final point: You advocate for the use of an extrinsic consequence in your article when you support the teacher’s removal of recess as a consequence for talking in class. Based on the information that has been presented, we hope you can now identify that talking in class and removing recess are not functionally linked, and that the teacher is attempting to use a punishing extrinsic consequence to modify behavior. We should consider that this student might not care one bit about missing recess; his behavior might be more strongly guided by the immediately reinforcing consequence of peer attention in response to talking in class. Instead of using punitive measures, it would be helpful (and likely more effective) for the teacher to consider the positive reinforcement that is currently maintaining the student’s talking behavior. She could then identify a way to remove the current consequence, and instead reinforce silent reading in class. You see, when we take the time to analyze behavior, it allows us to plan systematic, evidence-based interventions to improve behavior in ethical, informed ways!
Sincerely,
Caitlin Lostan, MA, BCBA
David Celiberti, PhD., BCBA-D
Association for Science in Autism Treatment (www.asatonline.org)
Reference
Cooper J. O., Heron, T. E., & Heward W. L. (2007). Applied behavior analysis (2nd ed.). Upper
Saddle River, NJ: Pearson.
