Raymond G. Miltenberger, PhDBryon G. Miller, MSHeather H. Zerger, MSMarissa A. Novotny, MS
Department of Child and Family Studies, University of South Florida
Shaping is the differential reinforcement of successive approximations of a target behavior. That definition relies on a handful of basic behavioral principles. Reinforcement refers to an increase in the future probability of a given class of behavior under similar conditions due to the relatively immediate occurrence of a consequence. Reinforcement, used for the acquisition and maintenance of a behavior, is a component of most applied behavior analysis procedures. The behavioral principle of extinction is the reduction and eventual near elimination of a behavior; extinction has occurred when a behavior no longer produces a reinforcing consequence. The combination of reinforcement and extinction is referred to as differential reinforcement, defined as the reinforcement of a specific response, while other response forms are placed on extinction (i.e., reinforcement is withheld). The outcome of differential reinforcement is the increased probability of the reinforced response and a reduction in all other nonreinforced responses. Successive approximations are the steps in response forms that lead incrementally to the target behavior. When successive approximations are differentially reinforced, response forms probabilistically change in the direction of the target. Shaping is a training procedure that can be used to generate novel behavior, to reinstate a previously exhibited behavior, or to change a dimension of an existing behavior; these applications are discussed in detail below.
Shaping can be conceptualized as both an explicit training procedure and a behavioral phenomenon that can occur naturally or unintentionally. As a training procedure, a simple yet illustrative example of shaping is teaching a pigeon to make a complete clockwise turn (Chance, 2014). At first, any turn in either direction (i.e., the starting behavior) results in reinforcement (i.e., typically a conditioned reinforcer, such as an auditory stimulus, paired periodically with an unconditioned reinforcer, such as grain). After this response occurs reliably, only turns in a clockwise direction are reinforced, whereas counterclockwise turns are placed on extinction. The next several steps involve reinforcing closer and closer approximations of a complete clockwise turn (e.g., quarter-, half-, and three-quarter-clockwise turns), with all previous approximations placed on extinction. In this example, the pigeon is specifically trained to engage in a selected target behavior. However, shaping often occurs naturally or unintentionally as a result of the prevailing contingencies of reinforcement (both social and nonsocial) and extinction.
The intensity of problem behavior such as tantrums or self-injury can be shaped unintentionally, where new and often disruptive or dangerous topographies of behavior emerge (e.g., Rasey & Iversen, 1993; Schaefer, 1970). For example, parents may reinforce a child’s tantrum by removing their demands, such that engaging in tantrums typically results in the child not having to comply with the parents’ instructions. Initially, the problem behavior consists of the child stating an emphatic “No!” when instructed to complete a task, which results in the parents removing the demand (i.e., giving in). In an attempt to increase compliance, the child’s parents begin to follow through with their instructions by not removing the demand when the child protests (i.e., extinction). In this context, extinction is often associated with an extinction burst, which can consist of a temporary increase in the severity of the problem behavior, the occurrence of novel behavior, or emotional responding. When faced with an extinction burst consisting of more-severe problem behavior (e.g., vocal protest and yelling at the parents), the parents might give in again, thus reinforcing a successive approximation to what will ultimately emerge as tantrum behavior. This process is then repeated as the parents begin to inadvertently reinforce more and more severe topographies of their child’s tantrums. This can result in the problem behavior being shaped, from a low-severity vocal protest to a severe tantrum, such as yelling, crying, throwing objects, and engaging in aggressive behavior.
It’s important for therapists to understand the inadvertent use of shaping so they can make sure that caregivers do not succumb to this practice. However, the rest of this chapter discusses shaping as a training procedure and reviews the steps involved in using shaping consistently and correctly. It presents illustrative examples of shaping from the literature and discusses them in further detail.
To implement shaping, the starting behavior is reinforced until the individual consistently engages in that response. Once this occurs, the next approximation is reinforced and the previous approximation is not reinforced (extinction). Once the individual consistently exhibits the second approximation, it is placed on extinction as the third approximation is now reinforced. The first and second approximations should stop occurring, as reinforcement is provided only for the subsequent approximation. This use of differential reinforcement is implemented for each successive approximation until the individual consistently engages in the target behavior. Although the number of approximations within a specific application of shaping might vary due to an individual’s ability or the complexity of the target response, in general, the following steps should ensure that shaping is implemented correctly (Miltenberger, 2016).
Shaping is used to get an individual to engage in a target behavior that he or she is not already exhibiting. In the sections that follow we describe the three applications of shaping: (1) generating novel behavior (i.e., behavior that is not in the learner’s repertoire), (2) reinstating a previously exhibited behavior, and (3) changing some dimension of an existing behavior.
Shaping can be used to promote the acquisition of a behavior that an individual has never exhibited (Miltenberger, 2016). For example, Ferguson and Rosales-Ruiz (2001) used eight shaping steps and a clicker (and occasional food) as a reinforcer to get five horses to walk into a transport trailer. Previously, aversive procedures (whips and ropes) were used to get the horses loaded into the trailer.
In a human example of developing a novel behavior, Shimizu, Yoon, and McDonough (2010) used shaping to teach preschool-aged children diagnosed with intellectual disabilities to point and click with a computer mouse. The first shaping step was moving the mouse around the computer screen. The reinforcer consisted of visual and auditory stimulation (rectangles on the screen disappeared or changed color and a bubbling sound occurred). The second shaping step was pointing the cursor to a single rectangle to produce the reinforcer. In the final shaping step, the subject was required to move the mouse, point it to a single rectangle, and press and release the mouse for the reinforcer to be delivered.
Mathews, Hodson, Crist, and LaRouche (1992) used shaping to increase children’s compliance with the use of contact lenses. Four children under the age of five who had previously demonstrated noncompliance with physician instructions during routine eye exams were chosen to participate in the study. Eight shaping steps, or variations of these steps, were used to teach contact lens wear. The shaping steps included touching the child’s face, pulling open an eyelid, having the child pull open an eyelid, placing drops in eyes, approaching the child’s eye with a finger, touching the child’s eye with a finger, touching a soft lens to the corner of the child’s eye, and touching a hard lens to the corner of the child’s eye. Compliance with each shaping step was reinforced with praise, stars, bubbles, food, or access to toys. This use of shaping increased contact lens use with three of the four children. It should be noted that this example is a variation of shaping; it did not involve successive approximations of the target behavior but rather successive changes in stimulation, to which the participants were exposed while holding an eyelid open and remaining compliant.
Shaping can be used to teach an individual to engage in a previously exhibited behavior that no longer occurs. In some cases the individual may no longer exhibit the behavior because he or she lost the ability to do so (e.g., teaching someone to talk after traumatic brain injury) or refuses to do so.
Meyer, Hagopian, and Paclawskyj (1999) used shaping to increase the number of steps a student with intellectual disability correctly performed each day. Previously, he had engaged in severe aggressive behavior when asked to get ready for school. The shaping procedure included ten steps, from brushing teeth to remaining in school each day. The reinforcers they delivered were contingent on a specific number of steps being completed each day, and the number of steps required was systematically increased. Results of the study suggest that shaping can be used successfully to increase compliance with morning hygiene skills and to increase attendance at school.
Taub and colleagues (1994) used shaping and verbal feedback/praise as a reinforcer to increase the motor movements of stroke victims who had lost movement in one of their limbs. The authors restricted the movement of the unaffected limb and used shaping to promote the use of the affected limb with a variety of tasks, including turning a Rolodex file, pushing a disc in a shuffleboard game, and rolling a ball. The researchers showed that shaping increased the number of turns of a Rolodex file and the distance a subject pushed the shuffleboard disc. Additionally, the time it took individuals to move a ball from side to side decreased. This study shows that shaping can facilitate behavioral rehabilitation in individuals who suffered neurological damage due to a stroke. Shaping has since been shown to lead to greater cortical recovery as well (Liepert, Bauder, Miltner, Taub, & Weiller, 2000).
O’Neill and Gardner (1983) used a shaping procedure to reinstate independent walking with a walker in an older adult who was noncompliant with physical therapy (PT) after hip replacement surgery. To start the shaping procedure, the therapist reinforced going to the PT room (i.e., the starting behavior). Once the subject was consistently going to the PT room, the therapist reinforced standing between two parallel bars for an increasing number of seconds, and going to the PT room was placed on extinction. This process continued through a list of successive approximations, including walking between the parallel bars for an increasing number of steps and walking the full length of the bars, until the subject walked independently with a walker.
When using shaping to reinstate a previously exhibited behavior, it is essential to first determine the reason the individual is not engaging in the behavior. For example, the presence of an aversive condition associated with the behavior might decrease an individual’s motivation to engage in the behavior, and in that case manipulating the environment in a way that removes this aversive condition might be enough to promote responding without the use of shaping. Before initiating shaping, however, it is essential to identify a powerful reinforcer to strengthen each approximation in the shaping process. The use of motivational strategies to augment the impact of reinforcers (see chapter 27) can also increase the effectiveness of shaping.
Shaping can be used to increase or decrease some dimension of a behavior (frequency, intensity, duration, or latency of a target response) that is not present at a satisfactory level. In this application of shaping, the target is a change in the behavioral dimension, such as an increase in speaking volume or a decrease in the number of cigarettes smoked per day.
Hagopian and Thompson (1999) used shaping with an eight-year-old boy with cystic fibrosis and an intellectual disability to increase his compliance with respiratory treatments. The target behavior was having the boy keep a mask on his face that released a medication mist. Initially they required the boy to keep the mask on his face for five seconds, after which he received praise and access to preferred items. The time he had to keep the mask on his face was systematically increased in five-second increments, until a goal of forty seconds was reached. Results of the study show that the duration of compliance increased from a mean of thirteen seconds to a mean of thirty-seven seconds, and the results were maintained at a fourteen-week follow-up.
In another example, Jackson and Wallace (1974) shaped behavior along the intensity dimension by reinforcing successively louder speech in a young girl diagnosed with a mild intellectual disability. In this study a reinforcer was delivered when she spoke at successively higher levels, as measured by a decibel meter.
Hall, Maynes, and Reiss (2009) used shaping to increase the duration of eye contact for two out of three individuals with fragile X syndrome. Participants received edible reinforcers and praise if they engaged in eye contact for a specified period of time. The time they had to make eye contact increased after each trial using percentile schedules of reinforcement.
Dallery, Meredith, and Glenn (2008) used shaping to decrease the number of cigarettes eight adults smoked. Following baseline, the researchers calculated a criterion that specified the number of cigarettes participants could smoke, which they determined from measured carbon monoxide (CO) levels. If participants’ CO levels were at or below the set criterion level, they received a monetary voucher. CO levels for five of the participants had decreased to levels of abstinence by the conclusion of the study.
In a novel example of shaping, Scott, Scott, and Goldwater (1997) enhanced the performance of a track-and-field athlete. The target behavior was for a pole-vaulter to raise the pole as high above his head as possible just before planting the pole to launch himself over the bar. Scott and colleagues used auditory feedback as a reinforcer for reaching a certain height with the pole. The height that was required for reinforcement was raised in five-centimeter increments over seven shaping steps until the athlete achieved his maximum arm extension.
O’Neill and Gardner (1983) describe a situation in which a woman diagnosed with multiple sclerosis interrupted her therapy program more than once per hour for bathroom visits. Ultimately, the therapist wanted the subject to wait two hours between each bathroom visit. The starting behavior, waiting one hour between bathroom visits, was reinforced until she consistently waited this amount of time. The next approximation was to wait seventy minutes. At this point, waiting one hour was placed on extinction, whereas waiting seventy minutes was reinforced with praise and approval from the therapist. This process of reinforcing increasing latencies between bathroom visits continued until the subject consistently waited two hours between bathroom visits.
Although behavior analysts have most commonly been the ones to use shaping, the opportunities for applied psychologists to use it are all around. For example, a clinician conducting psychotherapy who is interested in shaping self-disclosure, or emotional openness, or attention to the present moment can target and change this behavior in session. Potential reinforcers, such as attention, leaning forward, adopting a posture that mirrors the client’s posture, making clinical comments, clinician self-disclosure, or praise, can be explored in session, and if they function as reinforcers the clinician can systematically use them to draw out clients or help them to venture into new areas in terms of their relationships with others. Indeed, this idea is commonly used in clinical behavior analysis and contextual forms of cognitive behavioral therapy, such as functional analytic psychotherapy, which has been shown empirically to work in part through shaping in the psychotherapy session itself (Busch et al., 2009).
Shaping is a training procedure used to develop behavior that an individual is currently not exhibiting. More specifically, shaping is used to generate novel behavior, to reinstate a previously exhibited behavior, and to change the dimension of an existing behavior. A goal of most applied behavior analysis procedures is to promote the occurrence of desirable behavior that improves the quality of life of the individual engaging in that behavior. However, reinforcement cannot be used to strengthen desirable behavior if it does not already occur at least occasionally. Shaping provides a way for individuals to acquire desirable behavior in a stepwise fashion and for it to be strengthened through the application of several basic principles of behavior. Although shaping is used as a training procedure, it can also occur accidentally (e.g., the inadvertent shaping of problem behavior). The prevailing contingencies of reinforcement can occur in such a way that a variety of target behaviors can be acquired and shaped inadvertently.
Although shaping is a valuable training tool, it is not always the best-suited or most efficient method of teaching. Again, shaping is typically used to help an individual acquire behavior that is currently not strong or has never been established as part of the individual’s behavioral repertoire. A trainer can use differential reinforcement to increase behavior that does occur only occasionally. In addition, a trainer can deliver prompts or manipulate antecedent events to increase motivation so that the behavior is more likely to occur and contact reinforcement. Additionally, shaping is not ideal for training complex chains of behavior involving multiple topographies of behavior to be performed in sequence. To train these behaviors it is more appropriate to create a task analysis, which breaks a chain of behaviors down into individual stimulus-response components. The trainer can then use behavioral-chaining strategies that use prompting and fading to teach each stimulus-response component of the behavioral chain.
Busch, A. M., Kanter, J. W., Callaghan, G. M., Baruch, D. E., Weeks, C. E., & Berlin, K. S. (2009). A micro-process analysis of functional analytic psychotherapy’s mechanism of change. Behavior Therapy, 40(3), 280–290.
Chance, P. (2014). Learning and behavior. Belmont, CA: Wadsworth Publishing.
Dallery, J., Meredith, S., & Glenn, I. M. (2008). A deposit contract method to deliver abstinence reinforcement for cigarette smoking. Journal of Applied Behavior Analysis, 41(4), 609–615.
Ferguson, D. L., & Rosales-Ruiz, J. (2001). Loading the problem loader: The effects of target training and shaping on trailer-loading behavior of horses. Journal of Applied Behavior Analysis, 34(4), 409–424.
Hagopian, L. P., & Thompson, R. H. (1999). Reinforcement of compliance with respiratory treatment in a child with cystic fibrosis. Journal of Applied Behavior Analysis, 32(2), 233–236.
Hall, S. S., Maynes, N. P., & Reiss, A. L. (2009). Using percentile schedules to increase eye contact in children with fragile X syndrome. Journal of Applied Behavior Analysis, 42(1), 171–176.
Jackson, D. A., & Wallace, R. F. (1974). The modification and generalization of voice loudness in a fifteen-year-old retarded girl. Journal of Applied Behavior Analysis, 7(3), 461–471.
Liepert, J., Bauder, H., Miltner, W. H. R., Taub, E., & Weiller, C. (2000). Treatment-induced cortical reorganization after stroke in humans. Stroke, 31(6), 1210–1216.
Matthews, J. R., Hodson, G. D., Crist, W. B., & LaRouche, G. R. (1992). Teaching young children to use contact lenses. Journal of Applied Behavior Analysis, 25(1), 229–235.
Meyer, E. A., Hagopian, L. P., & Paclawskyj, T. R. (1999). A function-based treatment for school refusal behavior using shaping and fading. Research in Developmental Disabilities, 20(6), 401–410.
Miltenberger, R. G. (2016). Behavior modification: Principles and procedures (6th ed.). Boston: Cengage Learning.
O’Neill, G. W., & Gardner, R. (1983). Behavioral principles in medical rehabilitation: A practical guide. Springfield, IL: Charles C. Thomas.
Rasey, H. W., & Iversen, I. H. (1993). An experimental acquisition of maladaptive behavior by shaping. Journal of Behavior Therapy and Experimental Psychiatry, 24(1), 37–43.
Schaefer, H. H. (1970). Self-injurious behavior: Shaping “head banging” in monkeys. Journal of Applied Behavior Analysis, 3(2), 111–116.
Scott, D., Scott, L. M., & Goldwater, B. (1997). A performance improvement program for an international-level track and field athlete. Journal of Applied Behavior Analysis, 30(3), 573–575.
Shimizu, H., Yoon, S., & McDonough, C. S. (2010). Teaching skills to use a computer mouse in preschoolers with developmental disabilities: Shaping moving a mouse and eye-hand coordination. Research in Developmental Disabilities, 31(6), 1448–1461.
Taub, E., Crago, J. E., Burgio, L. D., Groomes, T. E., Cook, E. W., DeLuca, S. C., et al. (1994). An operant approach to rehabilitation medicine: Overcoming learned nonuse by shaping. Journal of the Experimental Analysis of Behavior, 61(2), 281–293.