Similarities and differences between extinction and punishment
What are the similarities and differences between extinction and punishment as behaviour-reductive strategies? Illustrate your answer with basic and applied research examples.
Applied Behaviour Analysis is the scientific appliance of behavioural principles to issues of social significance; helping psychologists to understand, and subsequently improve human behaviour. Behaviours can be measured within dimensions, frequency, duration or intensity; being observed and recorded by either the person engaging in the behaviour or others surrounding them. Consequently, this subject has helped individuals with developmental difficulties, mental illnesses, child behavioural problems and so on, proving to be a valid area in psychological research (pp. 101-137; cf Miltenberg. 2008).
One of the main basic behavioural principles, Reinforcement, was first studied by Thorndike (1911), illustrating that reinforcement is the process in which behaviour is increased and intensified by the immediate consequence that follows its episode. Using animals, such as cats, he found that once behaviour is strengthened, it is more likely to occur often, becoming a repetitive action. To portray this belief, Thorndike placed a food-deprived cat into a cage, however placing food outside the cage. After trying unsuccessfully to escape, the cat accidently hit a lever, opening the cage door and allowing access to the food. After repeating the experiment several more times, Thorndike discovered that every time the cat was placed into the cage, the amount of time it took to hit the lever was decreased, eventually leading to the automatic pressing of the lever and access to the food. He explained this phenomenon using the term ‘Law of Effect', illustrating how the cat learned, through operant conditioning, to hit the lever due to the immediate consequence of the food (Thorndike, 1911).
However, whereas reinforcement is responsible for the acquisition of behaviour, the phenomenon called Extinction is a process which weakens this learned and operant behaviour. Extinction has been described as when a behaviour has been previously reinforced (however is no longer followed with the reinforcing stimuli), the behaviour consequence does not occur in the future. This consequently means that the behaviour has been ‘extinguished'. Skinner and Ferster (1957) were two of the first psychologists to research into extinction, carrying out studies on laboratory animals, including pigeons and rats. They placed pigeons in an experimental chamber, which used operant conditioning, reinforcing them to peck a key for food. However, they found that when the bird stopped receiving food, its key-pecking behaviour stopped; illustrating extinction of the behaviour. In a separate study, they illustrated that when a rat no longer received food pellets after pressing a lever, the lever-pressing behaviour became extinguished, resulting in the rat no longer pressing the lever as it had learned the action will no longer acquire access to food (Skinner and Ferster, 1957).
As well as extinction being tested on animals, many scientists have used humans to study this behavioural principle; Williams (1959) even found that extinction could help a young boy with a behavioural problem of night-time tantrums (resistance to bed-time). After observing the child and his parent's behaviour using a reversal (A-B-A-B) design, Williams determined that the child's tantrums were being reinforced by attention from his parents. Therefore, the extinction process involved the parents simply ignoring their son when he acted disruptively at night, resulting in the extinction of his tantrums (pp. 101-137; cf Miltenberg. 2008).
Simarly, Hasazi and Hasazi (1972) undertook a study which led to extinction being used to correct a young child's arithmetic problems. Observing the eight-year-old boy's errors, they found that whenever he carried out a sum of addition with two-digit answers, he would reverse the digits (for example; write 41, instead of a 14) which resulted in extra help and subsequently attention from his teachers, reinforcing this behaviour. Therefore, the teacher was asked to refrain from providing attention after incorrect answers, yet give attention for correct answers (differential reinforcement), resulting in addition mistakes dramatically decreasing. This study is particularly interesting as many professionals would have diagnosed the child as having a learning disability, whereas Hasazi and Hasazi revealed that his behaviour was simply due to reinforcement via attention (pp. 101-137; cf Miltenberg. 2008).
Lovaas and Simmons in 1969 even found that the social consequence of attention was reinforcing self-harming behaviour. By removing the adult's attention whenever the child hit himself, dramatically reduced the frequency from 2500 hits in a one-hour session, to zero (pp. 101-137; cf Miltenberg. 2008).
However, whilst Lerman and Iwata (1995) researched into extinction they found that when a reinforcer is removed after a behaviour, an individual's action increases briefly in duration, intensity or frequency. For example, if an individual turns a door knob, however the door fails to open, the individual may try turning the door knob harder (increasing intensity) and more often (increasing frequency) than they normally would have done, before ultimately stopping. This theory, called an ‘extinction burst' was evident in 24% of participants in previously published literature on extinction, demonstrating very significant levels.
On the other hand, prior research has proven that although extinguished behaviours were believed to have been eliminated permanently, occasionally they may arise again, especially in environments where the behaviour occurred before (Chance, 1988). This spontaneous recovery will not last very long if the reinforcement is still not in place, however if the spontaneous recovery continues and is followed by the previous reinforcement, the effect of extinction will be lost as the problem behaviour will significantly increase (pp. 101-137; cf Miltenberg. 2008).
Extinction can be separated into two categories; positive extinction and negative extinction, in which they differ in procedural application. This means that whether a behaviour is maintained via positive reinforcement or negative reinforcement, they are both able to get extinguished. However, there practical methods vary because, if behaviour has been positively reinforced, to get extinguished the researcher has to withhold the reinforcement that had previously been presented after the behaviour. On the other hand, if a behaviour had been negatively reinforced, the extinction of the behaviour involves the avoidance of the aversive incentive which was in place before.
Comparable to extinction, another process which weakens operant behaviour is ‘punishment', which has been described as if an individual behaves in a certain way it may result in an immediate consequence which will make it less likely for them to act that way in the future. For example, if a young girl reached over a cage barrier to pet a dog, leading to the dog biting her hand, the girl would be less likely to reach over the cage again (pp. 101-137; cf Miltenberg. 2008). A punisher (which may also be referred to as an aversive stimulus) is therefore a consequence of an action and behaviour which decreases the probability of the behaviour being carried out again. However, unlike extinction, a punisher is only defined as the effect the consequence has on an individual's behaviour in the future, and is only described as a punisher if it decreases, or even eliminates, the same behaviour to happen again. For example, if a mother scolds her son every time he hits his twin brother, it is only a punishing act if he stops hitting his brother in the future. If not, the scolding punishment actually exhibits positive reinforcement as his disruptive act is repeated. This is one area where the behavioural applications extinction and punishment differ as punishment cannot be defined by what punishing act is applied, only by what behaviour follows. Therefore, psychologists define punishment by whether the problem behaviour increases, or decreases (as it is meant to). As well as this, because the child's behaviour may be temporarily stopped due to punishment, the adult's behaviour may, in fact, become negatively reinforced as they believe their punishing actions are effective and therefore continue every time the child misbehaves, which does not happen with extinction (pp. 101-137; cf Miltenberg. 2008).
On the other hand, punishment and extinction can be comparable in that there are many misconceptions about the both applications. For example, extinction may be seen as simply ignoring the problem behaviour, however this is not the case. Ignoring in this way functions as extinction only if attention is the reinforcer of this problem behaviour. Moreover, there are many misconceptions about punishment as even though the term punishment is only used in applied behaviour analysis as a consequence which decreases or eliminates certain behaviours, the general public opinion is very different. During every-day language, the term ‘punishment' is used, not only used to reduce crimes, but also to apply pain and retaliation to the criminal. Punishment, therefore, may be given in a form of fine, prison-time, or even a death-sentence; far from the definition used in applied behaviour analysis. Because of this misconception, adults are more wary when psychologists use this application to help childhood problems than they would be when told about ‘extinction' (Cooper et al, 1987). Perhaps this is one reason why extinction is a preferable application choice, as generally a punishment method is only carried out if other, nonaversive interventions (one of which being extinction), have been considered.
Like extinction, there are two variations of punishment; positive punishment and negative punishment. Positive punishment has been described as being when an occurring incident of a behaviour, which is followed by an application of an aversive stimulus, results in the likelihood of the behaviour not happening again. On the other hand, negative punishment has been described as being when a repeating behaviour's reinforcement gets removed ultimately resulting in the decrease of the problem behaviour. However, there is much confusion with these two definitions and positive and negative reinforcement as the definitions parallel each other, the important difference is; reinforcement strengthens behaviour yet punishment weakens it (Cooper et al, 1987).
One experiment which used positive punishment to weaken a problem behaviour was conducted by Corte, Wolf and Locke (1971) who reduced self-harming behaviour in institutionalized adolescents. Each time a subject hit themselves, a brief, non-harmful electric shock was applied, resulting in the reduction of the frequency of self-harm from approximately 400 to zero per hour. Yet, this research cannot be re-tested in recent times due to the strict ethical guidelines and moral concerns of the painful electric shock that was submitted onto participants (pp. 101-137; cf Miltenberg. 2008).
Additionally, positive punishment techniques were used when Sajwaj, Hibet and Agras (1974) treated a six-month old infant with a serious condition of rumination, meaning that the baby frequently regurgitated food into the mouth and then re-swallowed. Consequences of rumination include malnutrition, dehydration and even death, illustrating how much pressure was exerted on to the researchers to provide a solution. Fortunately, they found that the positive punishment technique of squirting a small amount of lemon into the infant's mouth every time she started to regurgitate, resulted in the elimination of the behaviour problem and subsequently weight gain and hydration (Sajwaj et al, 1974).
Lastly, response-blocking can be used which involves intervening in the person's behaviour, stopping them before the reinforcement occurs. This contrasts with extinction methods, as extinction can only be used after the behaviour has been reinforced. Similarly, contingent exercise is where an individual is asked to participate in an exercise which is not directly linked to the behaviour. For example, Luce et al (1980) was able to reduce aggressive and disorderly behaviour by asking children to repetitively stand up and sit down after showing the aggressive act of hitting someone. Because the consequence participants were made to carry out is unlike the behavioural act the contingent exercise method significantly differs from any extinction technique (Luce et al, 1980).
On the other hand, negative punishment can be used to weaken behaviour and can be conducted using two procedures; ‘time-out of positive reinforcement' and ‘response cost', both of which involve the loss of a positive reinforcing stimulus to solve problematic actions and behaviours. One of the first experiments that researched into the negative punishment procedure of time-out was conducted by Clark, Rowbury and Baer (1973) who were able to decrease disruptive behaviour in an eight-year-old year girl with Down's Syndrome, simply by removing her briefly from the reinforcing environment in which she was showing aggressive tendencies. For example, every time she acted aggressively towards her fellow pupils, she was sent, alone, to a small room for three minutes. Following this behavioural strategy, her disruptive behaviours decreased dramatically, especially as this technique also involved a loss of social reinforcement (attention) from her peers and teachers (pp. 101-137; cf Miltenberg. 2008).
Research into this area has indicated that there are two types of the time-out methods that may be used; non-exclusionary time-out (in which a child would still remain in the same room, for example a class room, however away from his/her friends and any reinforcing activities) and exclusionary time-out (which Clark et al used above by taking the child into a different room where reinforcers and positive stimuli are not available (Cooper et al, 1987).
Another group of researchers who studied into the time-out strategy was Plummer et al (1977) who illustrated that the context in which the negative punishment procedure of time-out is used overall influences the effectiveness of the process. They found that because time-out involves the child being removed from a reinforcing environment, time-out will not work if the ‘time-in' area is aversive with no positive reinforcers. They analysed the effects of time-out on disorderly behaviour in a class-room when the learning environment involved negative reinforcement (such as instructions). Results found that in this type of situation time-out actually increased disruptive behaviour as it allowed the child to escape from the teacher's instructions (pp. 101-137; cf Miltenberg. 2008).
However, when using an exclusionary time-out technique, there are a number of issues which need to be met to keep this negative punishment safe and effective. Because a purpose-built room is often used it is important to keep the room safe (with no sharp objects or breakable items inside), in fact this process works best if nothing, except a chair, is in the room, as objects such as toys and so on may reduce the effectiveness due to the reinforcing activities. A one-way window may also be needed to allow surveillance of the disruptive child, without taking time out of teaching other pupils which would subsequently affect their learning. Lastly, care must be taken when escorting the child to the room, as aggressive behaviour may occur
(Cooper et al, 1987).
Additionally, a normal amount of time spent in the time-out situation can range from anything from one to ten minutes, after which the child must resume to normal activity and therefore previous positive reinforcements. However, if the child is still acting in a disruptive manner at the end of the time-out, a contingent delay may occur where the duration of time can be extended (normally from one to ten minutes) until the child is no longer acting in a bad way. This contingent delay is supported by research from Erford (1999) who found that time-out with a contingent delay is a much more effective method of negative punishment than without (Erford, 1999).
Another experiment which supports findings of the time-out procedure was conducted by Rortvedt and Miltenberg in 1994. Using exclusionary time-out with two four-year-old girls, they were able to use this negative punishment technique to reduce compliance. Both of the two girls profusely resisted their mothers requests which followed by their mother's either scolding, threatening or even pleading with their daughters. However, whenever the two girls refused to comply, Rortvedt et al told the mothers to put them into a time-out situation, away from any positive reinforcers. Additionally, when the two girls conformed to their mother's requests it was followed by praise, reinforcing their good behaviour. Contingent delay was again used in this study, as the time-out period was extended to extra ten-second intervals, until the child was quiet for at least ten seconds. Results showed that after very few time-out procedures the two girl's behaviour dramatically improved (pp. 101-137; cf Miltenberg. 2008).
However, there are also disadvantages of the time-out technique, for example; there may be problem behaviour which goes undetected whilst the child is in the separate room. As it is impossible to observe the disruptive child continuously without affecting other pupils, that child may carry on acting disorderly without the teacher's acknowledgement. Additionally, another criticism of this technique is the public perception of the time-out rule. From parents of children in education, questions have been raised over how ethical this situation is and if teachers have the right to send their child unsupervised into a small room in which they are not allowed to leave without the teacher's consent. Therefore, it is perhaps best to undergo other procedures when these ethical questions are raised (pp. 101-137; cf Miltenberg. 2008).
Another procedure of negative punishment which may be used is ‘response cost' which can be defined as a removal of a particular amount of reinforcer dependent on the problem behaviour exerted. One study which used this response cost technique was by Phillips, Phillips, Fixsen and Wolf (1971) in which ‘predelinquent' adolescents in a treatment program earned points (which could be traded in for stimuli such as money, snacks and privileges) when behaving in a correct manner, with no disruption, crime or violence. However, if the adolescents arrived late for their meals, the researchers used response cost, where some of their points were taken off them, as a negative punishment. Therefore, this study illustrated how response cost can eliminate bad behaviour as late arrivals for supper decreased and eventually over-time were eliminated permanently (pp. 101-137; cf Miltenberg. 2008).
It may be the case that these two negative punishment applications get confused with the processes involved with extinction when solving problem behaviours. However they are three separate behaviour techniques and all differ very significantly. With extinction, the reinforcing event which followed the problem behaviour is stopped. On the other hand, time-out involves access to all reinforcements removed. And lastly, response-cost is where only a specific amount of a reinforcer (however not the reinforcer of the problem behaviour) is removed (for example privileges are taken away after the individual breaks a rule) (pp. 101-137; cf Miltenberg. 2008).
Extinction can similarly be contrasted with punishment whilst looking at their side-effects. Whereas the extinction technique may result in an ‘extinction-burst', where an individual displays an increase in intensity, duration of frequency of an act, the punishment application will not create this effect. Furthermore, a person's behaviour after punishment has taken place will not spontaneously recover, like it may with extinction. However, individuals using the punishment application may in fact take part in negative reinforcement if they start to escape or avoid the aversive stimuli. Similarly, punishment may involve an. This This is contrasted with extinction, as the individual engaging in the problem behaviour is not actually able to extinguish their behaviour themselves. On the other hand, these two weakening techniques are similar in that they both may result in the participant displaying emotional or aggressive behaviours, which is not good, especially when working with small children. However, overall it seems that the application of extinction is a much safer, as well as a more socially desirable, method of reducing problem behaviours (pp. 101-137; cf Miltenberg. 2008).
To conclude, there are a few similarities between the behavioural principles extinction and punishment; for example, they are both behaviour reducing strategies. However, even though they have comparisons, they are two separate techniques, with different strategies and methods. Confusion can occur specifically with extinction and negative punishment, yet extinction involves eliminating the reinforcer that was maintaining the behaviour and negative punishment involves eliminating the positive reinforcer after the behaviour. This means that, with negative punishment, the reinforcer that is removed is what the individual has previously acquired, but is not specifically the reinforcer that had previously been maintaining the behaviour. For example, if Sam interrupts his mother whilst she is reading, his disruptive behaviour is reinforced if she gives him the attention that he craved. To eliminate this troublesome behaviour an extinction application would make Sam's mother ignore his disruptive behaviour. On the other hand, a negative punishment application would involve a privilege of his to be taken off him. Therefore, these two behaviour problem techniques may both be tried in the same situation and do have the same ending effects, but generally they are both completely separate functions in which their methods of application vary significantly (pp. 101-137; cf Miltenberg. 2008).
- Chance, P. (1988). Learning and Behaviour (2nd ed.). Belmont, CA: Wadsworth.
- Clark, H., Rowbury, T., Baer, A., & Baer, D. (1973). Time out as a punishing stimulus in continuous and intermittent schedules. Journal of Applied behaviour Analysis, 6, 443-455.
- Cooper, J, O., Heron, T, E., Hedward, W, L. (1987). Schedules of reinforcement. Applied Behaviour Analysis. (pp. 275-295). Prentice-Hall International: London.
- Corte, H., Wold, M. & Locke, B. (1971). A comparison of procedures for elimination self-injurious behaviour of retarded adolescents. Journal of Applied Behaviour analysis, 4, 201-213.
- Erford, B. T. 91999). A modified time-out procedure for children with noncompliant of defiant behaviours. Professional School Counselling, 2, 205-210.
- Ferster, C. B. & Skinner, B. F. (1957). Schedules of reinforcement. Upple Saddle River, N. J: Prentice Hall.
- Hasazi, J. E., and Hasazi, S, E. (1972). Effects of teacher attention on digit reversal behaviour in an elementary school child. Journal of Applied Behavioural Analysis, 5, 157-162.
- Lerman, D. C., & Iwata, B.. A. (1995). Prevalence of the extinction burst and its attenuation during treatment. Journal of Applied Behaviour Analysis, 28, 93-94.
- Lovaas, O. I., & Simmons, J. Q. (1969). Manipulation of self-destruction in three retarded children. Journal of Applied Behaviour Analysis, 2, 143-157.
- Luce, S., Delquadri, J., & Hall, R. V. (1980). Contingent exercise: A mild but powerful procedure for suppressing inappropriate verbal and aggressive behaviour. Journal of Applied Behaviour Analysis, 13, 583-594.
- Miltenberg, R, G. (2008). Extinction. Behaviour Modification: Principles and Procedures: Fourth Edition. (pp. 101-113). USA: Thompson Wadsworth.
- Miltenberg, R, G. (2008). Punishment. Behaviour Modification: Principles and Procedures: Fourth Edition. (pp. 119-137). USA: Thompson Wadsworth.
- Phillips, E, L., Phillips, E, A., Fixsen, D., L., & Wolf, M. M. (1971). Achievement place: Modification of the behaviours of predelinquent boys within a token economy. Journal of Applied Behavioural Analysis, 4, 45-59.
- Plummer, S., Baer, D.M., & LeBlanc, J. M. (1977). Functional considerations in the use of procedural time-out and an effective alternative. Journal of Applied Behavioural Analysis, 14, 169-176.
- Rortvedt, A. K., & Miltenberger, R. g. (1994). Analysis of a high probability instructional sequence and time-out in the treatment of child noncompliance. Journal of Applied Behavioural Analysis, 27, 327-330.
- Sajwaj, T., Libet, J., & Agras, S. (1974). Lemon juice therapy: The control of life threatening rumination in a six month old infant. Journal of Applied Behaviour Analysis, 7, 557-563.
- Thorndike, E, L. (1911). Animal intelligence: Experimental studies. New York: Macmillan.
- Williams, C. D. (, C. D. (1959). The elimination of tantrum behaviour by extinction procedures. Journal of Abnormal and Social Psychology, 59, 269.
If you are the original writer of this essay and no longer wish to have the essay published on the UK Essays website then please click on the link below to request removal: