Increasing Behavior With Conditioned Reinforcement Learning Objectives After studying this chapter, you will be able to: > Discuss the differences between conditioned reinforcers, unconditioned reinforcers, generalized conditioned reinforcers, backup reinforcers, and token reinforcers. > Describe factors that influence the effectiveness of conditioned reinforcement in behavior modification pro- grams > Explain how those who are unfamiliar with the principle of conditioned reinforcement may unknowingly mis- apply it. Don't be so rude! Be nice! ERIN'S POINTS! PROGRAM “Erin, don't be so rude," exclaimed Erin's friend, Carly her homework that evening, she allowed herself time on “You're so unpleasant fo everybody, even your friends. social media according to her “points menu." Her menu ‘Why don’t you try being nice for a change?” As Carly looked like this: walked away, Erin decided that she had to modify her behavior. She wanted to be nicer to her friends. But boeing rude was such 2 habit that she knew she needed > PRINS OUT some extra motivation to change. After reading about 4 points 40 minutes of time self-management strategies in her psychology course, enn 660 minutes of time she decided to put herself on a points program. She really liked spending time on social media after doing her homework, but from naw on she would have to earn More than 6 points _As much time as | want the opportunity to do so. Every time she left her house in A week later when Carly and Erin were having lunch, Carly the morning, she carried an index card and a pen with said, “I can't believe how nice you've been lately It's ike her. Each time she said something nice to her friends, youte a different person.” Erin replied jokingly, "Yeah, its she gave herself a point on the card. Then after finishing that personality operation | had.” I Unconditioned and Conditioned Reinforcers We have inherited the capacity to be reinforced by some stimuli without prior learning, Such stimuli or events are import- ant for our biological functioning or survival asa species and are called unconditioned reinforcers, which are stimuli that are reinforcing without prior learning or conditioning, They are also called primary or unlearned reinforcers. Examples include food fora hungry person, water for a thirsty person, warmth for someone who is cold, and sexual contact for some- tone who has been deprived of such contact, Other stimuli become reinforcers because of particular learning experiences. ‘These stimuli called conditioned reinforcers, are stimuli that were not originally reinforcing but have become reinforcers by being paired or associated with other reinforcers. They are also called secondary or learned reinforcers. Examples of conditioned reinforcers include praise,a picture ofa loved one, books that we lke to read, our favorite television programs, and clothes that make us look good, Most of the reinforcers that influence us on a daily basis are conditioned reinforcersChapter 7 Conditioned Reinforcement 83, a the type of training conducted with dolphins at Sea World. Early on, the trainer paits the sound from a hand-held clicker with the delivery of fish to a dolphin. A fsh is a backup reinforcer, and after a number of pairings, the clicking sound becomes a conditioned reinforcer. Later, when teaching a dolphin to perform a trick, the sound of the clicker is presented as an immediate conditioned reinforcer, and the clicker sound continues to be intermittently paired with fish In this example of training the dolphins, the backup reinforcer—the fish—was an unconditioned reinforcer. However, backup reinforcers for a conditioned reinforcer could also be other conditioned reinforcers, T illustrate, consider Erin's program. The points that she awarded herself were not primary reinforcers, We doubt that she would have worked very hard fat all, just to obtain them for their own sake. The points were conditioned reinforcers because they were paired with the backup reinforcer, the opportunity to go on social media. In this example the backup reinforcer for the points was also «conditioned reinforcer. Erin was not born withthe stimuli provided by social media being unconditioned reinforcers for her. Rather, these stimuli would have become conditioned reinforcers by being paired with other things such as attention from adults and friends. Thus, backup reinforcers that give a conditioned reinforcer its strength can be either unconditioned reinforcers or other conditioned reinforcers. Questions for Learning Explain what an unconditioned reinforcer is. Give two examples. Explain what a conditioned reinforcer is. Give and explain two examples. Explain what a backup reinforcer is. Give and explain two examples. ‘What were the backup reinforcers in Erin’s program? Describe a target behavior of yours that you would like to improve that might be amenable to a points program like Erin’, What would you use as backup reinforcers for the points? Tokens as Conditioned Reinforcers ‘token system can be implemented with one or more individuals. The term foken economy usually refers to a token system implemented with a group of individuals. For example, a first-grade teacher might implement a token economy in which the children could earn stamped ‘happy faces for various behaviors, such as one stamped happy face per child for playing cooperatively during recess. At the end of the day, the children might be allowed to cash in their stamped happy faces for backup reinforcers to play a computer game or to have an extra 5 minutes of story time. Just about anything that can be accumulated can be used as the medium of exchange in a token economy or system. In some token economies, individuals earn plastic discs (like poker chips), which they can retain until they are ready to cash them in for backup reinforcers. In other token economies, people receive payment as “paper money” on which is written the amount earned, the indi- vvidual’s name, the name of the employee who paid the individual, the date, and the task the individual performed to carn the token, This information is used to control use and facilitate record keeping. In other token systems or economies, as in Erin's program, individuals receive points, which are recorded on a chart, on an index card, or in a notebook. (Token economies are discussed further in Chapter 24, Also see Boerke & Reitman, 2011; Hackenberg, 2009, 2018.) Related to the concept of conditioned reinforcement is the concept of ‘Moreover, punishing84 Part II Basic Behavioral Principles and Procedures tokens as well as reinforcing ones are possible. The demerit system used in the military is an example of a punishment token system, There are, however, problems with the use of punishment (as will be discussed in Chapter 15). Simple Versus Generalized Conditioned Reinforcers A stimulus can become a conditioned reinforcer because of pairings with a single backup reinforcer. In the days when ice cream vendors drove through a neighborhood and rang a bell to attract people’ attention, the sound of the ice cream vendor's bell became a conditioned reinforcer for the children in the neighborhood. After a few pairings of the bell with receiving ice cream, achild’s probability of making similar sounds—such as ringing a bicycle bell—increased, at least for a while (see the section called “Loss of Value of a Conditioned Reinforcer” below, ‘A mother who expresses pleasure at her child's good behavior is disposed to smile at, hug, or play with the child. Sometimes a treat, toy, or other things the child enjoys may accompany the mother’s praise. Normally, praise is established as a generalized conditioned reinforcer during childhood, but it continues to be maintained as one for adults, When people praise us, they are generally more likely to favor us in various ways than when they do not praise us. ‘Therefore, we are highly likely to engage in behaviors that are followed by praise, even when not deprived of any specific reinforcer. Common examples of simple and generalized conditioned reinforcers are given in Table 7.1, ‘Tokens can be both simple conditioned reinforcers and generalized conditioned reinforcers. Examples of the former are tokens that can only be exchanged for one specific backup reinforcer such as only jelly beans or only LEGO" bricks, whereas examples of the latter are tokens that can be exchanged for more than one type of backup reinforcer such as either jelly beans or LEGO’ bricks. Research indicates that tokens that are generalized conditioned reinforcers are more effective than tokens that are simple conditioned reinforcers—probably because, atleast in part, generalized conditioned reinforcers are not dependent on just one MO (eg, see Russel, Ingvarsson, Haggar, & Jessel, 2018; Hack- enberg, 2018). Questions for Learning 6. What are tokens? Explain in two or three sentences what a token economy is. Is money a token? Justify your answer Give two examples of stimuli that are conditioned reinforcers but not tokens. Explain why they are conditioned reinforcers, and why they are not tokens. 10, Explain what a conditioned punisher is. Give and explain two examples. 11, Distinguish between a simple conditioned reinforcer and a generalized conditioned reinforcer. Explain why a generalized conditioned reinforcer is more effective than a simple conditioned reinforcer. 12, Isa subway token a simple or generalized conditioned reinforcer? Explain your answer. 13. Isa grocery store coupon a simple or generalized conditioned reinforcer? Explain your answer. 14, Is praise a generalized conditioned reinforcer? Defend your answer. 15, Were the points in Erin’ program a generalized conditioned reinforcer? Defend your answer. TABLE 7.1 Examples of Conditioned and Uncondltioned Reinforcers ‘Examples of Generalized Conditioned Examples of Unconditioned Examples of Simple Conditioned Reinforcers Reintorcers Reinforcers Air miles Money Food Being told in a restaurant, “A waiters coming! Praise Water take your order.” A gift certificate for food and beverages. Sex A subway token, ‘ata restaurant Physical comfort ‘A coupon for a free hamburger Sleep NoveltyChapter 7 Conditioned Reinforcement 85 1B Factors Influencing the Effectiveness of Conditioned Reinforcement 1. The Strength of Backup Reinforcers ‘The reinforcing power of a conditioned reinforcer depends in part on the reinforcing power of the backup reinforcer(s) ‘on which itis based. For example, because social media was a strong backup reinforcer for Erin, the points functioned as effective conditioned reinforcers, 2. The Variety of Backup Reinforcers The reinforcing power of a conditioned reinforcer depends in part on the number of different backup reinforcers with which it has been paired. Money is a powerful generalized reinforcer for us because of its pairings with many backup reinforcers such as food, clothing, shelter, transportation, entertainment, and other reinforcers (see Figure 7.1). This factor is related to the preceding one in that, if many different backup reinforcers are available, then at any given time, at Ieast one of them will probably be strong enough to maintain the conditioned reinforcer at a high strength for an individual in the program. 3. The Number of Pairings With a Backup Reinforcer A conditioned reinforcer is likely to be stronger if its paired with a backup reinforcer many times, For example, the expression "good gir!” or"good boy” spoken to a very young child immediately following a desirable behavior is likely to bea stronger conditioned reinforcer if that expression has been paired with a hug from a parent many times as opposed to having been paired with a hug from a parent just once, assuming that other backup reinforcers were not involved, 4. Loss of Value of a Conditioned Reinforcer For a conditioned reinforcer to remain effective, it must, at least occasionally, continue to be paired with a suitable backup reinforcer. In the example of the token economy described previously, in which the children could earn stamped FIGURE 7.1 Why is money a generalized conditioned reinforcer?86 Part II Basic Behavioral Principles and Procedures happy faces from their teacher, if the teacher discontinued the backup reinforcers, the children may have eventually stopped engaging in the behavior for which they received the stamped happy faces. IB Pitfalls of Conditioned Reinforcement In Chapter 6, we introduced four types of pitfalls that work against those who have little or no knowledge of behavioral principles, and we showed how those pitfalls apply to the principle of positive reinforcement. Here we consider two of those pitfall types with regard to conditioned reinforcement, Unaware-Misapplication Pitfall People who are unfamiliar with the principle of conditioned reinforcement may unknowingly misapply it in various ‘ways. One misapplication isthe unknown pairing of positive reinforcement with a verbal stimulus that is meant to sup- press some undesirable behavior. An example of this misapplication is when an adult reprimands a child who is moti: ‘vated for attention. The attention that accompanies such negative verbal stimuli may be a positive reinforcer, thus having. the opposite ofthe intended effect. Thus, in some situations, reprimands and other negative verbal stimuli such as “No!” might come to function as conditioned reinforcers, and the individual will behave inappropriately to obtain them. An extreme example of this isthe parent who scolds a child for misbehavior and then, feeling guilty from the ensuing cry- ing, immediately hugs the child and provides a trea, The possible outcome of this unthinking procedure is that scolding could become conditioned reinforcer that would maintain, not eliminate, the behavior it follows. (In the following and subsequent chapters, we discuss effective ways to decrease problem behavior that do not have detrimental side effects) Partial-Knowledge-Misapplication Pitfall Ceasing to pair a conditioned reinforcer with a backup reinforcer can have unfortunate results for those who are not aware that this will cause a conditioned reinforcer to lose its value, An example of ths is a teacher who awards stamped happy faces as tokens for good behavior but fails to use effective backup reinforcers. The result is thatthe stamped happy faces eventually lose whatever reinforcing power they may have had when they were first introduced. IB Guidelines for the Effective Application of Conditioned Reinforcement The following guidelines should be observed in applying conditioned reinforcement, 1. A conditioned reinforcer should be a stimulus that can be managed and administered easily in the situations in which ‘you plan to use it. For example, points were ideally suited for Erin's program. 2. ‘Asmuch as possible, use the same conditioned reinforcers thatthe individual will encounter in the natural environment. For example itis desirable in training programs to transfer control from artificial tokens to the monetary economy of the natural environment or to naturally receive praise and attention from others. 3. In the early stages of establishing a conditioned reinforcer, a backup reinforcer should be presented as quickly as possible after the presentation of the conditioned reinforcer. Later, schedule thinning should occur in which either (a) the delay between the conditioned reinforcer and the backup reinforcer is increased gradually or (b) the amount of behavior required for each token is gradually increased, 4. Use generalized conditioned reinforcers whenever possible; that is, use many different types of backup reinforcers, not {just one. In this way, at least one of the backup reinforcers will probably be strong enough at any given time to ‘maintain the power of the conditioned reinforcer. 5. When the program involves more than one individual, such as a class of students, avoid destructive competition for conditioned and backup reinforcers. Giving a person reinforcement to the detriment of another may evoke aggressive behavior in the second individual or cause that individual’ desirable behavior to extinguish. One should therefore avoid drawing attention to the fact that one individual is earning more conditioned and backup reinforcement than another. Of course, people differ in their abilities, but designing programs in which individuals earn sufficient reinforcement for performing at their own levels can minimize any difficulties that these differences might cause.Chapter 7 Conditioned Reinforcement 87 6. In addition to the preceding five rules, one should follow the same rules for conditioned reinforcers that apply to any positive reinforcer (Chapter 6). Additional details for establishing token economies are described in Chapter 24 Questions for Learning 16, List three factors that influence the effectiveness of conditioned reinforcers. 17. Explain what causes a conditioned reinforcer to lose its value 18, Describe two pitfalls of conditioned reinforcement. Give an example of each, Summary of Chapter 7 Unconditioned reinforcers, such as food for a hungry person, are stimuli that are reinforcing without prior learning. Conditioned reinforcers, such as your favorite book, are stimuli that were not originally reinforcing but have become reinforcers by being paired or associated with other reinforcers. The later are called backup reinforcers. Tokens, such as money, are conditioned reinforcers that can be accumulated and exchanged for a variety of backup reinforcers. A behav. ior modification program in which one or more individuals can earn tokens and exchange them for backup reinforcers is called a token system or a token economy—usually a token system implemented with more than one individual A conditioned reinforcer that is paired with a single backup reinforcer is called a simple conditioned reinforcer, and a stimulus that is paired with more than one kind of backup reinforcer is called a generalized conditioned reinforcer. Money and praise are examples of generalized conditioned reinforcers. Factors iniluencing the effectiveness of con: ditioned reinforcement include (a) the strength of the backup reinforcers (the stronger the better); (b) the variety of backup reinforcers—the greater variety the better; (c) the number of pairings with a backup reinforcer—the more the better; and (4) whether pairings of conditioned and backup reinforcers continue at least intermittently. One pitfall of conditioned reinforcement is the pairing of a conditioned reinforcer with a stimulus meant to decrease problem behav- ior such as a reprimand, This can cause the latter stimulus to be a reinforcer because it provides attention that the indi- ‘vidual may not otherwise be receiving. A second pitfall is ceasing to atleast occasionally pair a conditioned reinforcer with a backup reinforcer, which will cause the conditioned reinforcer to lose its value Application Exercises A. Exercise Involving Others ‘What is the probable reinforcer and what behavior does it strengthen in each of the following situations? Are these reinforcers unconditioned or conditioned? Justify your choice in each case. An individual walks through a park in autumn and admires the beautifully colored leaves on the trees. ‘A person finishes jogging 3 miles and experiences the runner's “high’ ‘A teenager finishes mowing the lawn and is allowed to use the family car, A thirsty child holds a glass of milk and drinks several swallows. pore B. Self-Modification Exercise Identify a behavioral deficiency of yours that you would like to overcome. Next, describe the details of a plausible token system that might be applied by a friend or a relative to help you overcome your behavioral deficiency. Notes for Further Learning 1, Howisit that infants appear to learn new words when those words are not immediately followed by an observable reinforcer? A part of the answer lies with automatic conditioned reinforcement—a reinforcing effect produced by a response due to the resemblance of that response to a conditioned reinforcer (Skinner, 1957).Suppose that a parent88 Part Il Basic Behavioral Principles and Procedures says, “Say mama’ to an infant while providing reinforcement such as tickling, touching, feeding. After several such trials the sounds “mama” will become a conditioned reinforcer. Later, when in the crib alone, the infant may begin saying “mama” because of the automatic conditioned reinforcement received from reproducing the same sound. ‘More generally, vocal responses of infants may increase in frequency because the sounds that those vocal responses produce have become conditioned reinforcers and thus automatically strengthen their production responses. Studies have clearly confirmed this role of automatic conditioned reinforcement in early language acquisition (Smith, Michael, & Sundberg, 1996; Sundberg, Michael, Partington, & Sundberg, 1996). Automatic reinforcement appears to be important in not only language acquisition but also the strengthening of a variety of practical and artistic behaviors (Skinner, 1957; Vaughan & Michael, 1982). 2. What gives cigarettes their reinforcing power isthe addictive drug nicotine. Its doubtful that anyone would smoke just to experience the smell, taste, and sensations that come from cigarette smoke in the mouth, lungs, and nose. ‘Whatis perhaps not so well known is that because these stimuli are paired with the reinforcing effects of nicotine in the bloodstream, they become strong conditioned reinforcers, and for regular smokers the conditioned reinforcing effects of the stimuli associated with nicotine appear to be comparable to the unconditioned reinforcing effects of nicotine (Juliano, Donny, Houtsmuller, & Stitzer, 2006; Shahan, Bickel, Madden, & Badger, 1999). This underscores the point that when treating smoking and other addictions, therapists need to attend to the effects that conditioned. reinforcement can have on the progress ofthe treatment. Questions for Further Learning 1. How is conditioned reinforcement involved in influencing babies to babble sounds in their native language, even ‘when no adults are around to reinforce this behavior? 2. Discuss how conditioned reinforcement is involved in an addiction such as to nicotine and in making it difficult for people to quit their addiction. Note 1, Based on a case described by Watson and Tharp (2014).