Running head: BEHAVIOR MODIFICATION THROUGH POSITIVE REINFORCEMENT AND SHAPING
Behavior Modification Through Positive Reinforcement and Shaping
Jane M. Doe
Minnesota State University Moorhead
The field of behavior modification is based on the notion that one’s behavior can be influenced and changed to become more socially acceptable and desirable. One’s behavior can be controlled and manipulated through reinforcement and punishment, which is known as the process of operant conditioning. This theory is also used in reversing habits such as smoking or overeating and in enforcing the engagement of healthier behaviors. Behaviors also can be learned through a process of shaping. The new behavior is formed by reinforcing closer approximations of the wanted behavior until the desired behavior is performed. The field of behavior modification has generated many research studies that provide evidence that positive reinforcement and shaping really do work to modify behaviors.
Behavior Modification Through Positive
Reinforcement and Shaping
The Behaviorism era began and grew in popularity when the psychoanalytic theories of Freud became dissatisfying (Kazdin, 1978). Behavior modification through operant conditioning occurs when reinforcements and punishments are used immediately after the behavior is emitted. However, ceasing the use of the reinforcers can decrease these reinforced behaviors. This process is called extinction. Behaviors will be discontinued when they no longer result in desired responses. Shaping is a procedure where closer approximations of the desired behavior are reinforced until the particular behavior is performed by the individual. Theorists have hypothesized that reinforcing socially acceptable behaviors will result in the continuation of them and punishing behaviors will result in the extinction of them. It is important to remember the steps needed to modify behavior. The reinforcement must by systematic and consistent. Varying the type of reinforcement may not lead to the desired out come. It is also important to promote generalization of the conditioned behavior to different settings. This means that the individual should be able to not only display the rewarded behavior in one setting, but should be able to carry those same actions to different situations. And lastly, the reinforcement must be delivered immediately following the behavior. If this does not occur, the individual may be unclear about which behavior is getting reinforced or punished. Therefore, the individual will be unlikely to continue that particular desired behavior. The reinforcers must be something desired by the participant otherwise the behavior would be unlikely reinforced. On the same token, the punishers must be something aversive for the subject. This paper explores research studies confirming and disconfirming the effectiveness of behavior modification through the use of positive reinforcement and shaping.
B.F. Skinner and Edward L. Thorndike are the most noteworthy theorists of the process called operant conditioning. These studies are the baseline for today’s ideas about behavior modification. These animal research studies can also be transferred to develop conclusions about modifying human behavior. Thorndike is known for his “puzzle-box” experiment. He conducted his experiment by placing a hungry cat in a box and recorded how long it took the cat to escape it by moving a barrier. He placed a small piece of meat outside of the box as an incentive for learning to escape. The cat used the process of trial-and-error to discover the solution to opening the box. Soon the cat was able to escape and consumed the meat. The meat was reinforcing to the cat because not only did it get out of the box, but it also got the opportunity to eat the meat. Thorndike repeated this process numerous times and recorded the time on each trial. He noticed that the cat took less and less time to escape after each trial. This shows that the cat was able to learn a task faster when reinforced as opposed to no reinforcement because it had something to work for. Therefore, learning and accomplishing the new task will result in a desirable reward. (Kazdin, 1989). This study formed the foundation of the theory on operant conditioning. By manipulating the use of reinforcement, one can control the rate at which participants can complete a task. This study by Thorndike can be transferred to the explanation of the modification of human behaviors. For example, children are more likely to complete a task if they know they will be rewarded afterward. This will cause them to finish it quickly in order to obtain the desired reward.
A study by A. Robert Sherman (1973) hypothesized that a person’s behavior will increase if followed by rewards and decrease if followed by punishments. His study was conducted in a natural environment of a first grade classroom. The researcher observed the behavior in the classroom and noticed some disruptive students who would call out answers without raising their hands. The teacher was then instructed by the researcher to praise the students for any good behaviors and completely ignore students for any bad behaviors. When students acted disruptively such as talking without raising their hand, the teacher just ignored it. Ignoring the disruptive behaviors is punishing to those students, because the attention is never brought to them. The two objectives of this study are to stop the students from disrupting the class by calling out answers and to get the students to raise their hands to answer questions. Disruptive students who see the other students getting praise for raising their hands now see what they can do to get that same positive attention. Sherman measured the behaviors emitted by the children after manipulating the different levels of rewards and punishments. He compared the results of the frequency of disruptive behaviors before and after the teacher changed her strategies. The results showed that students rewarded with praise for acting desirable continued that behavior frequently and engaged less frequently in the undesirable behaviors. This study demonstrates effective behavior modification. (Sherman,1973). This is a perfect example of modifying human behavior because the teacher was able to control the students’ behaviors by virtually eliminating the disruptive behaviors. Knowing what the students desire, in this case it was attention, the teacher only gave that to them after they showed respectful, obedient behaviors. The teacher made receiving the reinforcer (praise) contingent on showing respectful, obedient behaviors.
Nay (1976) explored a study by H. Leitenberg who set up a study using positive reinforcement in a clinical setting. A 21-year old male had been hospitalized for almost 4 years and began to avoid social interaction. He secluded himself in his room, closing himself from the outside world. The researcher first began working on his behavior by defining the targeted behavior as initiating conversation with the nursing staff. He wanted to enforce the boy to engage in three 90-minute conversations a day. The nurses were told not to initiate any conversations and were to respond appropriately when the boy initiated the conversation himself. Leitenberg noticed that waiting for the boy to initiate conversation would not work. He then developed a system of reinforcements to get the boy to engage in social interaction. The boy was told that for every two minutes that he talked during the three sessions he would earn 5 minutes of listening to the radio or watching television. He began talking to the nurses longer only after receiving the reinforcers. He was later given five minutes of listening to the radio or watching television for engaging in conversations with people other than the nurses. This slowing shaped him into overcoming his social avoidance problem. Leitenberg recorded the time that he engaged in social interaction. His results revealed that during the first two days of the experiment when no reinforcement was used, the boy on average engaged in 4 minutes of conversation. On day thirty of the experiment, 20 days of reinforcing social interaction had gone by. The young boy was now engaging on average 55 minutes of conversation during the 90-minute session. (Nay, 1976) This study clearly follows the behavior modification theory using positive reinforcement and shaping. The boy’s social behavior only began to emerge with the presentation of rewards. He understood that if he interacts with the nurses, he will receive reinforcement. He was reinforced precisely after three minutes so this method was consistent and reliable; he new exactly which behavior he was getting rewarded for. Also because listening to the radio and watching television was desirable to him, this method of reinforcement worked for him.
Lowitz and Suib (1978) were able to reduce the frequency of thumb sucking of an eight-year old girl. They did this by presenting the girl with pennies for every one minute of not sucking her thumb. They met with her for five days and for about an hour each time. Within about five sessions of the training, her thumb-sucking was eliminated. ( Kazdin, 1989). The young girl associated the elimination of the thumb-sucking and the presentation of the reinforcer; pennies. This process of reinforcement was aimed at eliminating a behavior by presenting a desirable reward for every response but the undesired behavior. This follows the theory of behavior modification through the use of positive reinforcement. The young girl only decreases the behavior when reinforcement is used. Eventually the use of positive reinforcement will eliminate the behavior completely. (Nay, 1976).
James K. Luiselli (1997) conducted a study involving an 8-year old boy, Don, with Pervasive Developmental Disorder. Don had problems getting potty trained. This study was performed in the classroom of his elementary school. The researcher had an assistant who escorted Don to the restroom at scheduled times throughout the day. The researcher first recorded Don’s behavior before reinforcing him. Don would refuse to go to the restroom. He also continued to urinate in his diaper. After recording these behaviors, Luiselli began the reinforcement phase. This phase consisted of singing fun songs to reinforce him in the bathroom. This helped make the behavior more enjoyable for him. Don was also given a drink of water out of his favorite water bottle for every successful restroom trip. This was extremely reinforcing for Don because it was exciting for him to drink out of his favorite water bottle. Luiselli documented the results of this study and revealed that after three weeks of reinforcing Don, he had a 100% success rate at becoming potty trained. This experiment provides evidence in favor of behavior modification through positive reinforcement. Don was able to learn a new association between successfully using the restroom and receiving a drink of water. Therefore, Don viewed the behavior of using the restroom more enjoyable than before the implementation of positive reinforcement strategies. What defined success in this situation, though, was the 100% success rate; the target rate did indeed increase.
Philip A. Saigh (1980) conducted an investigation on a group of Lebanese children in Beirut, Lebanon that the teacher had a hard time controlling. She wanted to eliminate two behaviors: talking out of turn and leaving one’s seat without permission. The teacher had originally experimented with negative reinforcement without any success. After discovering that the students especially enjoyed swimming, Saigh developed a method of modifying behaviors by combining swimming with good behavior. The teacher would place a check on the board for every ten minutes the students were well behaved. The check stood for ten minutes of swimming time for the students. The results showed that before the positive reinforcement method was used, the students displayed on average 18 disruptive behaviors. On the first day of implementing the check system, behaviors dropped down to 4 incidents. This system was an easy and effective way of teaching the students to learn respectful behaviors. Because the activity was so enjoyable to the students, this system worked so well. The teacher was consistent with reinforcing the students by giving them a check every ten minutes if they deserved one. The students were then aware of exactly what behaviors were unacceptable and which ones were acceptable.
Behaviors can also be modified through a process of shaping. This process involves the presentation of a reinforcement after closer approximations to the desired behaviors are displayed. This is clearly shown in a study performed by Jackson and Wallace (1974). They conducted their study on an 8-year-old girl, Alice. Alice was mentally retarded and had a voice that was virtually inaudible. Jackson and Wallace designed a shaping method to get Alice to speak at a decibel level like her peers. They started by giving her a token redeemable for beauty products for every increase in speak volume she made. At first only slight increases of volume were reinforced. When Alice consistently met the volume requirements, she received a token. After she reached the volume requirement, she was not reinforced until she reached the next volume level. They continued this process until she finally spoke at the same volume as her peers. (Kazdin, 1989). Alice was continually reinforced until she displayed the desired behavior of speaking loud enough to hear. She was reinforced immediately after the behavior was emitted; therefore, she knew which behavior was getting rewarded. Alice was able to learn an association between her speaking louder and receiving a reward. After completing this therapy treatment, Alice was able to continue her audible voice without getting reinforced by token, presumabye because natural social reinforcers took over.
The use of positive reinforcement has been considerably useful in the modification of juvenile delinquent behavior. Delinquents tend to deviate from society to get recognized and draw the attention to themselves. To correct their behavior, researchers have tried punishing them for every behavior emitted. However, punishment has shown to be somewhat unsuccessful in changing their behavior. Researchers then tried methods of positive reinforcement aiming at teaching them socially acceptable behaviors. Fixen, Phillips, Phillips, and Wolf (1976) conducted a study at a group home of numerous delinquents. Researches began by defining the behaviors they wanted to enforce and the behaviors they wanted to eliminate. They wanted to encourage the adolescents to do chores, watch educational television, read the newspaper, and receive good grades. The behaviors they wanted to eliminate were poor grades, aggressive talk, lying, stealing, and fighting. Before implementing the reinforcement method, the adolescents only engaged in, on average, 55% of their time to the daily tasks required of them. The students were then given rewards such as going downtown, watching television, and receiving an allowance for every good behavior shown. After about thirty sessions of receiving rewards, the adolescent averaged 75% of their time to learned acceptable behaviors. Later results revealed that the students who participated in this method had fewer criminal offenses than did the juveniles that were not apart of this program. (Kazdin, 1989). These students realized that there are other ways to get the attention they desire. They were able to engage in socially acceptable behavior knowing that they will be rewarded for it in the end. This type of attention, along with the rewards, was desirable to the students therefore making this method successful.
Not only can the use of positive reinforcement be used to promote socially acceptable behaviors and eliminate disruptive and undesirable behaviors, it can also be used to save the life of a patient suffering from anorexia. Clinicians have been using this method to enforce the patient to eat and gain weight. They do this by presenting rewards every time he/she successful eats. Kadzin (1989) described at study conducted by Pertschuk, Edwards, and Pomerleau in 1978 involving severely anorexic patients. They administered the program by giving patients the opportunity to earn rewards for every one-half pound they gained. About two weeks into the program the patients gained an average of 9.3 pounds. Patients who do not agree to this program continued to lose weight. Patients who did gain weight had a goal and were able to reach that goal by earning incentives. This pushed the patients even more to overcome their sickness. The results of this study correlate with the theory on behavior modification. The researchers were able to control the behaviors emitted by the patients through rewarding them every time they performed the desired behavior. Like in the previous studies, these patients were able to develop a relationship with gaining weight and obtaining a reward. The more they ate, the more reward they received making the behavior of eating more desirable for the patient.
Another important behavior that needs to be enforced is wearing safety belts while in a moving vehicle. Kazdin explored an example of enforcing the usage of safety belts by Sowers-Hoag, Thyer, and Baily, (1987). These researchers conducted a study on 158 children attending day care. Assistants would escort the children to the car and noted whether or not the child belted himself or herself before leaving the day care. Only about 1% of the children successfully did this. After recording the baseline statistics, the researchers implemented a method where they informed the children about the importance of wearing the safety belt and explained that they would be eligible for prizes every time the buckled-up on their own. Knowing this, 75-86% of the children began buckling themselves on their own. Children knew that every time they wore their seat belt when leaving day care, they would get a prize. This made them enjoy the behavior, making it more likely they would perform the task. This continuation of receiving a reward for wearing a safety belt will eventually lead the children to develop a habit of wearing a seat belt every time they are in a vehicle.
Through investigation of ten different studies researching behavior modification using positive reinforcement and/or shaping, it is safe to say that this method is quite successful. Researchers were successfully able to control participants’ behaviors by promising them a reward for performing the particular behavior. By doing so, participants then learned a relationship between performing the desired behavior and receiving a reward. Researchers thoroughly followed the requirements of presenting a reward desirable to the participant, giving it systematically—contingent on the desired behavior, presenting the reward immediately after the behavior, and promoting generalization. With these steps, the participant finds the behavior desirable because of the result of gaining rewards. Knowing these procedures, people can enforce socially acceptable behaviors in an enjoyable manner.
Agras, W.S. (1972). Behavior modification: Principles and clinical applications.
Boston: Little, Brown and Company.
Bijou, S. & Ribes-Inesta, E. (1972). Behavior modification: Issues and extensions. New York: Academic Press.
Hamerlynck, L. A., Handy, L. C., & Mash, E. J. (1973). Behavior change: Methodology concepts and practice. Champaign:Research Press.
Kazdin, A E. (1989). Behavior modification in applied settings. Pacific Grove: Brooks/Cole.
Luiselli, J.K. (1997). Teaching toilet skills in a public school setting to a child with pervasive developmental disorder. Journal of Behavior Therapy and Experimental Psychiatry, 28, 163-168.
Martin, Garry and Pear, Joseph. (1983). Behavior Modification: What it is and how to do it. Second Edition. Englewood Cliffs: Prentice Hall.
Nay,W. R. (1976). Behavioral intervention. New York: Gardner Press.
Saigh, P.A. (1980). The effects of positive group reinforcement on the behavior of Lebanese school children. The Journal of Social Psychology, 110, 287-288.
Sherman, A. Robert. (1973). Behavior modification: Theory and practice. Belmont: Wadsworth.
Without Miracles: The Adaptive Modifications of Behavior. Retrieved: Nov.11, 2005. http://faculty.ed.uiuc.edu/g-cziko/wm/07.html.