SPECIFIC FORMS OF PUNISHMENT

This section describes specific forms of punishment. The principles that apply to punishment in general apply to each of these as well. In addition, each of these has unique features worth considering. After reading this section, you should be able to: 1. Define and give examples of each of the following specific types of punishment: (a) reprimands, (b)time-out, (c)response cost, (d) negative practice, (e) overcorrection, (f) covert behavior control, and (g) corporal punishment. 2. Describe guidelines for applying each of the types of punishment listed in the previous objective.

Reprimands One of the most common forms of punishment is the verbal reprimand. A corrective statement becomes a verbal reprimand (a form of Type I punishment) if it makes the learner feel at least mildly unpleasant. (If it results in no negative emotion, then the statement is not a form of punishment. Some recipients consider apparent reprimands to be Type I reinforcement. On other occasions corrective comments - like praise - serve some other function, such as irrelevant teacher activity.) The three main rules with reprimands are (1) that they should be brief, (2) that they should not attack the personality of the person being reprimanded, and (3) that they should convey information regarding an appropriate course of action. In addition, Van Houten et al. (1982) found that the effectiveness of verbal reprimands could be enhanced by combining them with nonverbal cues, such as eye contact or a firm grasp; that reprimands work better if they are delivered from nearby rather than far away; and that effective reprimands can reduce the misbehavior of nearby peers as well as the student being reprimanded. Abramowitz, O'Leary, & Futtersask (1988) have found that brief reprimands tend to result in fewer off-task behaviors, whereas longer reprimands often lead to talking back and other off-task behaviors. By identifying the target behavior and suggesting

alternatives (Kounin, 1977), the teacher can make it easier for students to solve similar problems in the future.

Systematic Exclusion (Time-out) Time-out is an example of Type II punishment. It involves removing the person from a source of reinforcement for a designated period of time. The assumption behind time-out is that the child finds the stimulus setting in which he is functioning to be a pleasant one and would like to continue being there. When removed from that setting, therefore, the child regards the removal as an aversive turn of events. Hence, this is an example of Type II punishment. However, if the situation is an unpleasant one or the situation into which the person will be placed is regarded as more pleasant than the current one, then the attempt at time-out would be reinforcement rather than Type II punishment. For this reason, requiring a disobedient child to sit in the corner won't work if the child wants to be left alone anyway. Likewise, sending a child to a room if it is full of toys or to the principal's office where the child can talk to all the visitors who come in will often be reinforcing rather than punishing. In each of these cases, the child is deriving a benefit, not suffering a loss, by being forced to leave the current stimulus setting.

Time-out

Vernelle constantly fights with her brothers. Her mother sees Vernelle fighting and tells Vernelle to come into the house, commenting, "If you can't play right with your brothers, you can't play with them at all." About an hour later, Vernelle is allowed to go back outside, with the warning that if there is any more fighting, she'll have to come in again. Chuck whispers to his friends in class. Mrs. Van Til sends him to a part of the room where she can keep an eye on him, but where he cannot see or be seen by the rest of the class. She lets him return to his seat after he has worked quietly for twenty minutes.

There are three types of time-out:

1. With exclusionary time-out, the student is excluded from the activity that leads to reinforcement. For example, a player may taken out of a basketball game or thrown off the team, or a student who has misbehaved in reading class may be required to leave the room. 2. With nonexclusionary time-out, the student is not removed from the instructional setting; but the teacher denies the student access to reinforcers through temporary manipulation of the environment. For example, if students misbehave during free time, the teacher may turn off the record player; or if children misbehave on the school bus, the driver may turn off the radio until they quiet down. If a teacher uses a point card to administer a token reinforcement program, confiscating the point card for a period of time would constitute nonexclusionary time-out. Another example of nonexclusionary time-out is called contingent observation. With this method the student is moved to the edge of the regular activity and is able to continue observing the activity (and to gain vicarious reinforcement) but is unable to gain direct reinforcement. 3. Seclusionary time-out goes further than exclusionary time-out: the student is removed from the regular activities and goes to an area designated for social isolation, often referred to as the time-out room. The time-out room should be a neutral (rather than unpleasant) area.

Advantages of Time-out Time-out can be relatively free from negative side effects. It has the additional advantage that is often easy to combine with Type II reinforcement at its termination. Mr. Davis's treatment of Walter (page xxx) is a good example of such a combination. Removal of recess and suspension from school can be effective time-out procedures, if the child really wants to be engaged in the activity from which he or she is being removed. Of course, if the child dislikes the activity from which he or she is being removed, the child would experience Type II reinforcement rather than time-out.

Guidelines for Using Time-out

In addition to the general guidelines for punishment,the following guidelines are useful when implementing time-out procedures:

Remove all the important reinforcers. If a child is allowed to take his favorite comic book with her, if she can watch the varsity basketball team practice, or if the school secretary tells her jokes, she is being placed in a reinforcing (not punishing) situation. It may sometimes be a good idea for a teacher to talk sympathetically about the problems of his home life with a student kept after school. However, if the child enjoys this conversation, this procedure is not an example of time-out. The time-out environment should be a neutral (not aversive) environment, free from either pleasant or unpleasant stimuli. Putting a young child in a dark closet that frightens him is not time-out, but rather a harsh form of Type I punishment - the presentation of an aversive stimulus. Putting the child in the hall at school is often ineffective, since the child is likely to receive visitors, who can provide Type I reinforcement. Putting a child in an empty office or lighted cloak room is more likely to represent time-out - unless the child can derive pleasure from going through drawers or coat pockets. Many teachers have found that a good technique is to place a portable blackboard in front of the room, blocking the child's view so that the teacher can see the child but the child cannot interact with any of the other children in the room. This way the teacher can remove any reinforcers the child happens to find. Keep the time-out as short as possible. The learner is absent from the learning environment while in the time-out area. Therefore, she is missing opportunities to practice desirable behaviors as well as the undesirable one. Since missing these opportunities to practice positive behaviors could be harmful, it is desirable to get the learner back in action as soon as possible. Therefore, the time-out should be long enough to be as aversive as the situation requires, and short enough not to waste her time. Don't give "time-out" from aversive situations. If you do, you will be reinforcing whatever behavior the person performed to get the "time-out." The result will be an increase rather than a decrease in the undesirable behavior. A student who hates school and can most easily get suspended by smoking is likely to become a chain smoker at a very young age.

School Suspension

Suspending players from baseball games when they argue with umpires is often an effective form of punishment, because players normally like to be in the lineup. (Note that if the fans cheer wildly for suspended players, these suspensions become less effective.) Suspending students from school, on the other hand, is often ineffective, because the students who are suspended may actually dislike being in school; and so suspension is actually a form of Type II reinforcement rather than time-out. In addition, even if it is unpleasant to the suspended student, suspension has the disadvantage of removing students from the very activities that are supposed to be beneficial. Therefore, it is often better to use a strategy other than suspension to control undesirable behavior. A useful alternative to ordinary suspension is in-school suspension. The simplest form of this technique requires students to report to a different classroom than their unsuspended peers. They remain in this room all day, doing homework or other activities assigned by their teachers under the supervision of a businesslike adult. More sophisticated forms include activities designed to support adaptive behavior, such as those described by Barth (1986). While away from their regular academic activities, the students engage in these supportive activities, and there may be an exit interview during which they discuss plans for improved behavior before returning to the regular classroom.

Response Cost Response cost (Kazdin, 1972; Iwata and Bailey, 1974) is an example of Type III punishment. It involves the removal of a specific number or amount of reinforcers which have already been gained. The term often refers to the loss tokens, but it can also refer to the loss of any reinforcer that has already been gained. For example, a student who has gained additional free time may lose some of it, or a basketball player who has been promoted to first string may be demoted because of poor sportsmanship. Response cost has been found to be a powerful classroom strategy that is often free of many of the negative side effects of punishment (Kazdin, 1972; Bandura, 1973).

Eddie (Response Cost)

Loses

his

Tokens

Eddie has been working on a token reinforcement system. He gets one token for every five

arithmetic problems he solves correctly. Eddie has discovered that if he works quickly but carelessly, he can turn in ten items, get half of them wrong, and still get a token for the five correct ones. His teacher notices Eddie's behavior and changes the rules so that Eddie gets one token for every correct answer, but loses two tokens for every incorrect answer. Since he realizes he will be penalized for careless work, Eddie slows down and performs his work with near-perfect accuracy. Used consistently and fairly, response cost is a relatively powerful and convenient strategy for bringing about rapid reduction of behaviors. In addition to the general guidelines for punishment, the following guidelines will be helpful in employing response cost: 1. Allow the person to accumulate a sufficient supply of reinforcers before removing any. With a token system, if students can collect several tokens and even trade some of them for back-up reinforcers, they will place a higher value on the tokens and consequently be more concerned about losing them. 2. Penalize sparingly and fairly. If you remove all the tokens or points, you can no longer use response cost. (What else can you remove?) If students view the whole situation as hopeless, they will stop looking for the reinforcers which could have been earned and seek rewards from some other source, perhaps through undesirable behavior. 3. Allow opportunities to regain lost reinforcers. By doing this, you can combine reinforcement for desirable behaviors with punishment for the undesirable behavior. In addition, if you don't provide opportunities to regain lost reinforcers, students may lose interest in the reinforcement program and seek reinforcers elsewhere. However, if you give the reinforcers back too easily, then the punishment loses its effectiveness because the cost for misbehaving becomes almost insignificant. 4. If you plan to use response cost, use reinforcers that can be easily withdrawn. For example, with a token reinforcement system, if you use play money to reward good behavior, you may have trouble repossessing it when you decide to use response cost. It may be better to use a "bank book" or point system over which you can exercise more direct control. Likewise, if you use candy as a reinforcer, the child is likely to wolf down the candy rapidly to keep you from repossessing it. 5. Make the response cost system as natural as possible. To the extent that students perceive that they are being bribed to be good, they are likely to behave appropriately only as long as the response cost system is in effect. Any system that emphasizes fines is likely to be perceived as arbitrary and artificial. Try to focus on the natural benefits of engaging in

productive behavior and to use the response cost system as a way for students to selfmonitor their own behavior. If students view the removal of reinforcers as the loss of a bonus they could have gained by remaining productive, they are less likely to experience negative side effects than if they feel that they have been arbitrarily fined.

Negative Practice Negative practice is a form of punishment which requires a person to repeat an originally pleasant behavior to the point where it becomes aversive. This technique works especially well with uncontrolled, automatic behaviors which a person would like to eliminate but cannot.

Negative Practice

Rocco wanted to quit smoking but couldn't. One day he bought a carton of cigarettes and forced himself to smoke the whole carton in one evening. He hasn't touched a cigarette in nineteen years. His problem now is that he feels sick when other people smoke around him. Crystal bit her fingernails. She wanted to stop but couldn't. One day she bit them until they hurt. The next day, when she brought her hand to her mouth, the pain was still there; and so she consciously withdrew her hand. Even after the actual pain was gone, its memory remained. Crystal no longer unconsciously bites her fingernails. Steve was throwing Paper Wads across the room one day. The teacher asked him to stay after class and handed Steve a stack of 200 sheets of scrap paper and instructed him to wad them up and throw them into the basket. If he missed, he was required to pick that piece up and try again. When Steve finished, the teacher dumped the wastebasket in front of Steve and told him to try it again. Steve did this five times. He no longer enjoys throwing Paper Wads. (Note that a problem similar to Steve's will be treated by satiation later in this chapter. The recipient of negative practice feels bad, whereas the recipient of satiation feels bored.)

Note that this technique could border on cruelty. It should be employed with care. It is often effective when the recipient voluntarily requests help eliminating a maladaptive behavior. Negative practice will be successful only if the behavior is actually repeated to the point where it becomes thoroughly unpleasant; otherwise negative practice could actually result in a strengthening of the behavior through practice. In addition, in the case of voluntary treatment (such as Rocco and Crystal - the examples just given), the person must be strongly motivated to change the behavior. Since negative practice is a rather harsh technique, it is likely to be accompanied by negative side effects except, of course, when it is undertaken voluntarily. Negative practice often can and does occur by accident. For example, parents and teachers sometimes find that a behavior is so rewarding to a child that they insist that the child perform the behavior as often as possible. The behavior may not only reach a satiation point, but it may actually become aversive to the child. A familiar example is the child who likes to play a musical instrument, and the parents (often because of a financial investment) insist that the child keep at it. Often such children grow to hate music.

Overcorrection With overcorrection the person is required to correct an inappropriate behavior to an extreme degree and/or to overcompensate for any undesirable results which occurred because of the behavior. For example, a child who slams a door loudly when she enters a room might be required to go back and close the door gently ten times before being allowed to continue her activities. Telling the child to be more careful in the future is "correction." Requiring the child to practice the proper technique to an obnoxious degree is overdoing this correction, and therefore it is "overcorrection." Or a student who spells a word wrong on an essay might be required to write the word correctly twenty-five times. Looking up the correct spelling is "correction." Overdoing this corrective process by writing the word over and over is "overcorrection."

Wanda (Overcorrection)

Stops

Sucking

Her

Thumb

Wanda sucks her thumb. She sucks it automatically and cannot stop herself. Her mother decides that whenever Wanda sucks her thumb, Wanda will be stopped immediately and taken to the bathroom, where she will be required to wash her mouth out with a vile-tasting liquid, have her mouth cleaned with a brush, and have her lips washed very firmly and thoroughly. Such a process takes up a very unpleasant five minutes for Wanda. After this process occurs for four days, Wanda stops sucking her thumb and never resumes the habit. There are two common misunderstandings with regard to overcorrection. First, many readers think that any excessively harsh punishment is overcorrection. This is an understandable mistake; but this is not the correct definition of the term. To qualify as overcorrection, the punishment must be a logical corrective procedure which becomes unpleasant because it is overdone. Second, many readers confuse overcorrection with negative practice. They are similar in that they can both take a previously automatic, involuntary (but pleasant) activity and turn it into a conscious and unpleasant activity. The difference is that with negative practice it is the activity itself that becomes obnoxious; whereas with overcorrection it is the corrective procedure that is unpleasant. To pursue the door slamming example from the previous paragraph: it would be negative practice to have the child go back and slam the door repeatedly until this activity itself became annoying; it would be overcorrection to have the child go back and close the door gently until this activity became annoying. Both of these techniques are rarely used, partly because of the possibility of negative side effects and partly because of the restricted number of activities for which they are appropriate. Overcorrection has been found to be a very powerful technique. It also could be a very harsh technique (accompanied by numerous negative side effects) unless it is administered nonvindictively. If it is administered as a logical and automatic result of the inappropriate behavior, the negative side effects are less likely to occur. In most cases some other technique would be better. Problems to which overcorrection can sometimes be applied are thumb-sucking, smoking, and similar automatic behaviors.

Covert Behavior Control Only behaviors which can be observed can be deliberately reinforced or punished using the strategies discussed throughout this chapter. Therefore, it is impossible to reward or punish internal thoughts, right? Wrong! There is one person who can observe a thought - the person who thinks it! What this means is that although it is impossible for someone else to identify and reinforce my thoughts, I can reinforce my own thoughts, if I choose to do so.

Thoughts are susceptible to many of the same influences applicable to overt behaviors. The terms "covert conditioning" and "cognitive behavior modification" have been applied to the whole group of techniques which apply the tactics discussed in this chapter to internal thought processes. Since a very large number of our most important behaviors are controlled by our thoughts, changing thoughts often leads to longterm and important changes in behavior.

Covert Behavior Control

Eldon is a juvenile delinquent who would like to reform but compulsively steals cars. He finds himself daydreaming quite frequently about his planned thefts. His counselor tells him that every time he fantasizes a car theft in the future he should immediately shift his thoughts to something as unpleasant as possible. Thereafter, whenever Eldon realizes he is daydreaming about a car theft, he imagines a huge dog jumping out of the car at him and his girl friend laughing at him for being afraid of the dog. Eldon is being covertly punished for daydreaming about car thefts. Since he knows that Eldon daydreams a lot and will continue to do so, the counselor goes one step further. He recommends that Eldon covertly reinforce himself by daydreaming about more socially acceptable things. This examples differ from counterconditioning (discussed later) in that in counterconditioning the two thoughts are paired, whereas in covert punishment the aversive thought comes immediately after the undesirable thought - as a consequence for that thought. Marty Improves (Covert Behavior Control) his Racquetball Game

Marty was a proficient racquetball player. Yet he found himself running around the court and hitting the ball without concentrating on aiming the ball to get it past his opponent. Therefore, he made the following agreement with himself: If he did not first say to himself "Think" and form a specific strategy, he would withdraw the privilege of swinging at the ball. The situation was really quite embarrassing to catch up with the ball, get set, and then not swing. After this happened a few times, he started concentrating much better. Now he plans almost all his shots and plays a much better game.

Corporal Punishment Corporal punishment refers to the infliction of physical pain in order to create an aversive situation for the recipient of the pain. Because of ethical considerations and personal convictions, most people find difficulty in giving the topic of corporal punishment an objective evaluation. Although corporal punishment has some advantages (Vockell, 1991), it has been criticized by a large number of professional organizations (Cite, 19xx), and it has the following serious disadvantages: 1. Corporal punishment emphatically teaches only what not to do. Corporal punishment cannot readily be teamed up with Type II reinforcement to teach some desirable behavior at its termination. The punishment ceases when the pain stops hurting. Therefore, no way is easily available to see to it that the child is performing a desirable behavior when the punishment ceases. 2. It is extremely difficult to make the severity of the punishment proportionate to the seriousness of the offense. 3. Corporal punishment is less likely than other forms of punishment to bear a logical relationship to the offense. 4. Corporal punishment is very likely to lead to aggressive behavior on the part of the child and/or others watching the punishment. A child is likely to think, "If Mom and Dad solve their problems by hitting me, then that's the way I'll solve my problems too." 5. Corporal punishment is more likely than many other forms of punishment to lead to suppression of spontaneous behaviors, to avoidance behaviors, and to retaliatory behaviors on the part of the child. 6. Among older children especially, peer prestige associated with spanking often supplies a reinforcing value which actually overshadows the punishing effect of the spanking. 7. Corporal punishment is too often used by the parent/teacher while in a state of frustration, exasperation, or anger. A person who has severely spanked a child may be forced to admit that had the same offense occurred on the previous day, the spanking would not have occurred.

8. Perhaps the most important argument against corporal punishment is that it is hard to integrate it with Type II reinforcement at its termination.

In most cases, there are many other more effective control techniques which do not entail the previous disadvantages. A final point that needs to be made is that possibly all the rhetoric about banning corporal punishment misses the point. If what we are concerned about is being humane to children and punishing only in an effective manner, then corporal punishment is not necessarily the worst way to punish children. For example, public ridicule from a parent or teacher is often viewed by children as being much more cruel than a privately administered spanking. As a matter of fact, with most older children, it is the humiliation aspect of a spanking that hurts more than the actual physical impact. Likewise, depriving a child of privileges or learning opportunities for prolonged periods of time is going to hurt the child in a very real sense much more than even a rather severe bruising on his or her body. Passing laws against spanking children does not eliminate the abuse of children. Spanking is only one possible form of corporal punishment. Lifting a child up by the hair or squeezing an arm severely are sometimes harsher to the child's body than a carefully administered swat with a hand or paddle. In addition, psychological abuse (including neglect) is probably a more serious problem than physical abuse. The only effective way to insure humane treatment of children is to develop attitudes of love and concern for children.

Accidental Punishment In discussing strategies for helping learners develop desirable behaviors, we have focused on strengthening such behaviors through the various types of reinforcement. However, in some cases there is an even simpler approach: stop punishing the behavior, and it will increase on its own. One of the most significant features of punishment is that the aversive situation is often brought about by accident. When an aversive situation occurs unintentionally, the rules of punishment still apply. In order for punishment to have its impact of reducing behavior, the only prerequisites are (1) that the recipient of the punishment must perceive the situation as being aversive and (2) that the recipient must see a cause-and-effect relationship between the aversive situation and a behavior which the recipient has performed. It matters not at all whether someone imposed this aversive

situation on purpose or whether it occurred without outside planning. In either case, the outcome will be the same: the punished person will be less inclined to perform the activity which brought about the aversive situation.