Building constructive prison environments: the functional utility of applying behavior analysis in prisons.
By the end of 2001, there were 2.1 million people incarcerated in America's prisons (Harrison and Beck 2002). In 2000, the National Center for Policy Analysis reported, "an estimated 630,000 inmates were released from prison last year, with an estimated 160,000 of those being violent inmates" (Du Pont 2000). These statistics beg several questions: is it surprising that the United States has surpassed South Africa as Western Civilization's most imprisoned people? What is being done, during the inmates' incarceration, to address antisocial behaviors that are being reinforced in prisons today? What are the maintaining variables that reinforce a person's development of pro-social behaviors? Does the society that imprisons these people recognize that criminal behavior does not simply stop at the entrance or the exit of a correctional facility? What is being done to change the behavior of these inmates into people who will eventually become our neighbors? Over the last twenty years there has been much advancement in the technology of Applied Behavior Analysis. Have these advancements facilitated the building of constructive prison environments? The purpose of this paper is to describe a current program that applies Behavior Analysis in a maximum-security prison setting.
Although the national trend toward more stringent sentencing laws is decreasing, prisons are left to deal with a subset of the population that is considered "unteachable". Likewise, "unteachable" was also the term used in the 1950's and 1960's to describe people with severe and profound retardation. It was in the 1960's that Behavior Analysts began to research the field of Mental Retardation and Dual Diagnoses (MR/DD), and today, there are few behaviors that cannot be changed through the systematic manipulation of envirom-nental factors. Also, starting in the 60's and continuing into the early 80's, Behavior Analysts became interested in the application of the principles of behavior analysis in the correctional setting. In 1974, B.F. Skinner spoke out on the issue of how to build constructive prison enviromnents in a "letter to the editor" of the New York Times:
"It is possible for prisoners to discover positive reasons for behaving well rather than the negative reasons now inforce, to acquire some of the behavior which will give them a chance to lead more successful lives in the world to which they will return, to discover that the educational establishment has been wrong in branding them as unteachable and for the first time to enjoy some sense of achievement. But that can only be brought about through positive action."
Most of the groundwork for building Behavior Analytic based constructive prison enviromnents have been outlined in the research emanating from this era in Behavior Analysis. A specific example of this research was undertaken in 1971 by John McKee, a Behavior Analyst, regarding the issue of contingency management in a correctional institution. McKee describes a system in which 16 inmate volunteers' performance in academic areas were effectively managed through the use of contingency management. The study consisted of three phases. The first phase was a three-week baseline period in which academic productivity was established. During phase-two, the experimenter systematically controlled the "academic output" of the inmates. During phase-three, the inmates were taught to generalize what they had learned during this experiment in the form of self-management. Overall, the inmates' performances were increased by 20%. Another example is that of Levinson et al (1968). Levinson and his colleagues used negative reinforcement (an increase in any behavior that terminates decreases, removes or avoids the aversive stimulus). In this paper, he describes a system in which the subjects were 17 inmates who had been repeatedly classified to the "Segregation Unit". The negative reinforcer used was continued involvement in therapeutic group therapy. "Escape" from the unit required the individual go three months without engaging in a pre-defined, maladaptive behavior. Eleven subjects completed the program, thereby successfully escaping from group attendance. Of the 11 subjects, three did not return to segregation in three months, and six were placed on work release.
Disagreements regarding incarcerated individuals as research participants, research methodologies, technical terminology and the political environment of the prison itself resulted in many Behavior Analysts leaving the prison environment. When the empiricists left this setting, the prisons were left without a technology of behavior change.
Once incarcerated, the majority of inmates will participate in the programs and treatments that are available, and do not engage in problematic behavior. Some, however, take a different path and become what is termed a "behavior problem." Mental health providers are usually unable to effectively treat the behaviors of individuals that range from harassing and provocative to dangerous and deadly because the behaviors are not due to a documented mental illness, and are therefore resistant to typical forms of therapy and psychopharmacological approaches. The task of managing the behavior of such inmates is left to security staff who are poorly trained to deal with the extreme behavior of such inmates on a long term or on an ameliorative basis. As a result, the inmate is often placed in the most restrictive housing that a states' Department of Corrections, or the Federal Bureau of Prisons has to offer, the Supermax.
The Supermax, or Intensive Management Unit (IMU), is a prison within a prison. The National Institute of Corrections (NIC), an arm of the U.S. Department of Justice, defines a "Supermax" as:
" ... a free-standing facility, or a distinct unit within a facility that provides for the management and secure control of inmates who have been officially designated as exhibiting violent or serious & disruptive behavior while incarcerated [emphasis added]. Such inmates have been determined to be a threat to the safety and security in traditional high-security facilities, and their behavior can be controlled only by separation, restricted movement, and limited direct access to staff and other inmates ... "
The same 1997 NIC report listed 32 states, the Federal Bureau of Prisons and the Correctional Service of Canada-, as having at least one Supermax. Most supermax prisons operate similarly; the inmates are housed in their cells for 23 hours a day. Termed isolation by some, it may seem anything but that upon a tour of such a facility. The Supermax is an often-noisy environment where inmates are keenly aware of what is going on within the area in which they are housed. In addition to the high security setting many states utilize a programmatic structure referred to as "the program." The program is a behavioral ladder or levels system. Maladaptive behavior is consequated through a series of "demotions", that occur quickly after the behavior. An interdisciplinary team reviews behavior at regular intervals to allow for upward progress on this ladder. The higher the level the more privileges such as increased recreational time and visits with family as well as allowances for personal property, conversely the lower the level the fewer the privileges and property. The lowest level allows only personal hygiene items, religious artifacts and legal materials. Typically there are some educational and treatment services available within this setting, though few, if any are empirically derived or based in research. Prison administrators may be compelled to respond to these "problematic inmates" that assault staff and inmates by further decreasing the privileges of these inmates and essentially "locking the problems away" by imposing longer and longer sentences to the Supermax with no objectively defined release criteria. Therein lies a functional paradox: how does a prison provide for the safety of their staff while creating an environment that is effective in increasing pro-social behaviors while decreasing maladaptive behavior? Recently, the present author has been investigating applying the principles of Behavior Analysis in a correctional facility. While Applied Behavior Analysis in corrections is not novel, an application in the 21st Century brings the research from the last 20 years into the correctional facility, and with that a new era in corrections.
In many cases, negative punishment (the removal of previously earned privileges or tangible items to decrease the rate of a behavior) is the sole tool used to manage inmates' behavior while in prison. If an inmate does his time and stays out of trouble he is ignored. If he gets into trouble, he has privileges and personal property removed in effort to reduce the rate of such behavior. The problem begins when all privileges and personal property have been removed. What does the inmate do when he has nothing to loose? The answer is that he does what he has always done when he had nothing left to loose. If an inmate came from a background of education and communication, then he would use the coping skills he had been taught and he would manage while he earned the privileges and personal property back. If the inmate came from a background of aggression and acting out, he would aggress and act out until he is given back the item or he exhausts himself. One of the ways that Behavior Analysts build constructive prison envirom-nents is to limit the sole reliance on negative punishment used to reduce the rate, magnitude or duration of maladaptive behaviors and provide alternative approaches to inmate management.
The Oregon Department of Corrections has launched an alternative approach to the management high risk or "problematic" inmates, which involves a functional change in their program management. This facility had historically relied heavily on negative punishment to reduce the rate of anti-social or problematic behavior. The changes to the IMU program have to mediate the need for lawfully mandated security with the need for behavior change. The most significant change to the IMU program will be the implementation of an empirically derived behavioral treatment model. This model is similar to McKee's Contingency Management in a Correctional Institution (McKee 1974) but rather than addressing academic behavior, this program address problematic or dangerous behaviors. The programmatic model under the present investigation also has some underpinnings of Levinsons' negative reinforcement based group therapy. The basis for the model is cooperation between security and programs staff with the understanding that the IMU is a unit where behavior change must occur to qualify for release. An interdisciplinary team made up of mental health professionals, Behavior Analysts, social workers, educational professionals, medical and security staff function as the program committee with oversight capacities for this unit. The unit offers opportunities for access to tangible rewards, social attention and escape maintained behaviors.
Admission Process to the IMU
The inmate is classified to maximum custody as a result of an observed behavior deemed as "dangerous" and "a risk to the safe and secure operation of the facility". The operational definition of a behavior that is termed "a risk to the safe and secure operation of the facility" is any class "A" violation. Class "A" violations include, among other descriptors, assault (sexual, physical, with or without a weapon), escape, and creating a disturbance or "riot". This classification is applied by staff in a section, independent from the management of the IMU. All inmates are afforded due process and have access to an appeal process. Currently there are about 128 male inmates ranging in age from 18 to 59.
IMU Program Level-2: Program Determination
The inmate enters at IMU program level-2. When outside of his cell, the inmate is handcuffed and two staff escort the inmate to the desired location. A nylon tether is attached to the handcuffs. In the event the inmate moves to assault a staff or another inmate being moved, the attending officers physically restrain the inmate using the nylon tether to maintain physical control of the inmate. An interdisciplinary team works in conjunction to assess the inmate in global skill areas such as communication, social, academic, and coping skills. The deficient skill areas are objectively defined and prioritized. These skills and deficiencies are objectively behaviorally defined and specific behaviors targeted on a form called a Behavior Action Plan (BAP). The inmate must maintain this level with no misconduct reports (a report issued by staff of any observed behavior that violates the Rule of Prohibited Conduct) issued for 60 days. Once completed the inmate may progress to program level-3.
Behavior Action Plan Design
The BAP includes the description of the targeted behaviors and the criteria for successful performance. BAP programming maybe subdivided into two categories: macro-behavior management and micro-behavior management techniques
Macro-behavior management techniques
Macro-behavior management in the IMU consists of different "in-cell" study packets and group participation programs and the use of the consequence based IMU program level system. Macro-behaviors consist of several discrete behaviors. An example of a macro-behavior is anger management. Anger management may consist of learning one's behavior cycles, recognizing and responding to antecedent stimuli and relaxation training. In cell study packets are a set of written packets that encompass various subject matters such as: anger management, learning to do prison time, journaling, and alcohol and drug treatment materials. These materials have been divided into packets that are handed-out to inmates and picked up by correctional officers on consecutive weeks. An inmate must achieve a proficiency score of 80% to successfully pass that packet and receive the subsequent packet. Correctional staff serve as a proctor to the inmate if the inmate does not pass the packet. If the inmate fails twice, he will be reassessed for different programming. An inmate may be assigned one or more programs from a list of approximately 14 programs. Inmates are consequated for compliance with the rules of the IMU, and successful program completion with climbing levels in the IMU program level system. Consequences for climbing levels include releasing from the IMU after having completed all assigned programs, the addition of certain privileges or access to personal items, and more visits. Inmates are consequated. for non-compliance with the rules of the facility or refusing to engage in assigned programs by being demoted one or more levels in the IMU program level system. The consequence of having a level demotion may include some or all of the following consequences: spending longer in the IMU, having one's program re-evaluated for effectiveness, the restriction of privileges or access to personal items, visits or phone calls. Inmates that engage in behaviors that are classified as a "major misconduct" under the Rules of Prohibited Conduct are demoted to IMU program level-1. Major misconduct includes assault, creating a disturbance, and contraband smuggling or possession. Level- I is a restrictive level in which the inmate has access to only personal hygiene items, religious artifacts and legal materials. All inmates in the IMU are reviewed monthly at an interdisciplinary team meeting; Inmates at level one are reviewed weekly while at this level. After an absence of behaviors that can be classified as a major misconduct for 30 consecutive days, the inmate is promoted to IMU program level-2. Inmates cannot be demoted to IMU program level- I for refusing to complete assigned programs.
Micro-behavior management techniques
Micro-behavior management in the IMU consists of the application of an Individualized Behavior Plan (Webb 2001). A Micro-behavior consists of a discrete behavior such as self-injury or assault. The Individualized Behavior Plan (IBP) is a part of the BAP and a set of instructions to the correctional staff and data collection techniques that the correctional staff employ in response to certain events. The behavior often starts outside of the IMU, but persists once the inmate is transferred to the IMU.
Potentially reinforcing and punishing consequences are determined through a descriptive analysis (Vollmer, Borrero, Wright, Van Camp, and Lalli, 2001). A functional analysis (Derby and Wacker, 1992) can also be used to identify antecedent stimuli or those stimuli that set the occasion for a problematic behavior. An example of an antecedent would be the mail is late in coming to the tier and the inmates begin to yell, bang cell doors and otherwise cause the tier to become noisy. Mail not coming at a predetermined time would directly precede an increase in the rate of "acting-out." To continue this example, a training opportunity is created in the presence of this clear of an antecedent. For staff, the training opportunity would be in how an antecedent sets the occasion for certain behaviors to occur; the increase in acting-out on the part of the inmates directly followed the mail being late. The training opportunity for the inmates would be flexibility, because as a consequence for their behaviors they received a misconduct report and therefore may have extended their stay in the IMU because the mail was late.
A functional analysis can also be used to determine the contingencies that are reinforcing the rate of maladaptive or problematic behavior and punishing the rate of adaptive or pro-social behavior. An example of the use of consequences derived from a functional analysis in a prison setting is in Webb (200 1). In this paper, the author removes the opportunity for the inmate to escape from his cell, which had been previously applied as an unintended consequence for maladaptive behaviors. The participant would engage in disruptive behaviors (yelling, screaming, and banging on the cell doors with his shoes) and the correctional staff would move him to a cell that was located in the hall away from the other inmates. Through a functional analysis the author noted that the rate of disruptive behaviors increased in conditions that allowed for escape from the inmates cell. The rate of disruptive behavior decreased in all other conditions that did not allow for escape from cell. The functional analysis was videotaped and used to train the correctional staff how the environment affects the rate of the participants behavior. An IBP was written that encouraged the inmate to make appropriate verbal requests to move from the noise of other inmates without engaging in disruptive behaviors. If the inmate made an appropriate verbal request without engaging in disruptive behaviors then he was allowed leave his cell to be escorted to a less noisy cell. The result was the inmate decreased the rate of disruptive behaviors (banging and screaming late at night) and an increased "pro-social" verbal behavior (verbally requesting a move). All Individualized Behavior Plans (IBP) are included in the overall Behavior Action Plan (BAP).
IMU levels 2 and 3: Implement Program
The BAP is implemented under observation by correctional staff. Observation systems include the use of direct correctional staff observation, and recorded video observation. Baseline rates of targeted behaviors are established and behavior change documented through the use of a customized database called Special Inmate Management System (SIMS[C]). The SIMS[C] database is used for recording the rate and frequency of behaviors, the number and type of assigned programs, the inmates' current programmatic level and other relevant identifying information 24 hours per day. In short, this database allows for concurrent management of thousands of individualized BAP's. The SIMS[C] database has built-in behavior graphing and analysis tools, allowing behaviors across groups or sections of the IMU to be evaluated. Using a computerized database also allows for longitudinal studies of recidivism (the return to prior criminal behaviors) and ensures that if an inmate were to return to the IMU, the failure of the previous program can be noted and addressed in a new BAP for that inmate.
Correctional staff were trained to collect data using a paper system that is transferred to the SIMS[C] database. During 50% of initial training trials, a second observer simultaneously but independently record data to assess interobserver agreement. Agreement percentages were calculated by dividing the training trails into 10-s intervals and comparing observers' recorded data on an interval-by-interval basis. The number of scoring agreements between the correctional staff and the independent observer was divided by the total number of intervals and then multiplied by 100%. The mean percentage agreement between correctional staff and the independent observer was 95% with a range of 82% to 100%.
Consequating behavior at IMU level-2
After 60 days with no misconduct reports issued, or targeted behaviors observed, the inmate is promoted to IMU program level-3. IMU program level-3 includes increased visits, personal property and an increase of the amount of money the inmate may spend at the commissary. There is no criterion set for completion of programming at IMU program level-2
IMU Level 3 to 4: evaluating the efficacy of the BAP
Interim progress was assessed by determining if the objectives of the specific needs identified in the BAP have been met. If the objective for that targeted behavior has been met, then treatment is faded. If further programming of the targeted behavior is warranted because the objective was not attained or there is a need for further programming of that target behavior the portion of the BAP is reassessed. Once all outcome behaviors have been attained for that target behavior, the next priority in the BAP is addressed. Once all targeted behaviors identified through the BAP have been met, planning for release to a less restrictive environment begins. Levinson's use of negative reinforcement is also applied at this point; the inmate must have been misconduct report free during his "program". If at any point the inmate received a misconduct report, then his BAP may be reassessed and he may loose one or more levels.
Consequating behavior at IMU level-3
After 90 days with no misconduct reports issued, or targeted behaviors observed and successful completion of the BAP, the inmate is promoted to IMU program level-4. IMU program level-4 includes further increased visits, personal property and an increase of the amount of money the inmate may spend at the commissary. The inmate is also allowed to move from his cell unescorted by security staff.
IMU Level 4: Planning for generalization of learned behaviors
Some inmates return to the IMU after being discharged, which suggests that behavior modification techniques employed in the IMU are not maintained after leaving the IMU. Generalization is the degree to which a change in behavior will transfer to another setting or situation or the degree to which a behavior change program influences behaviors other than then targeted behavior. In 1977, Stokes and Baer stated that Generalization (when contrasted with behavior change programs), " ... has been considered the natural result of failing to practice a discrimination technology adequately, and thus has remained a passive concept almost devoid of a technology." Much effort is put into targeting behaviors, but very little is done to maintain the behavior in other settings. Generalization of the behaviors targeted in the IMU to a less restrictive environment called the Close Supervision Unit (CSU). The CSU is a transitional services program unit similar to the IMU, but without the high security precautions. Communication of the plan for generalization is accomplished via the SIMS[C] database. With SIMS[C], the staff at the CSU is able to access the inmate's previous BAP, the rate and type of behaviors observed, and progression through the IMU program levels system. Macro-behaviors of compliance with staff directives and integration with staff and inmates are targeted through a transitional BAP. The two units' tandem operation will be the subject of subsequent papers.
Consequating behavior at IMU level-3
After 30 days with no misconduct reports issued, or targeted behaviors observed, the inmate is transferred to the CSU. IMU program level-4 includes further increased visits, personal property and an increase of the amount of money the inmate may spend at the commissary. The inmate is also allowed to move from his cell unescorted by security staff.
This IMU was opened in 1991. The aforementioned procedures were put in place in January of 2002. It is difficult to decisively state that the change to the DVTU has changed inmates behavior, correlated with the most recent changes, are significant decreases in the rate that misconduct (a violation of the Rules of Prohibited Conduct) and in the rate of inmate grievances (an administrative appeal route used to address prison management/inmate disputes over the interpretation or application of the rules and, in some instances, the procedures of the Department of Corrections) are reported (see figure 1).
The belief that there is "nothing that can be done to change the behavior of criminals" is as dated as the concept that behavior analysis could not affect the behavior of people with retardation. Through careful planning, communication and the application of the principles of applied behavior analysis we have, and will continue to change the behavior of even the most violent people. Further areas for investigation in the area include the effect that this type of programming has on inmates returning to prison and to this unit.
Derby, K. M., D. P. Wacker, et al. (1992). "Brief functional assessment techniques to evaluate aberrant behavior in an outpatient setting: A summary of 79 cases." Journal of Applied Behavior Analysis 25(3): 713-721.
Du Pont, P. (Author). (2000). Radio America Essay. National Center for Policy Analysis. (Available from NCPA 665 15th St N.W., Suite 375--Washington, DC 20005)
Harrison, P.M. & Beck, A.J. (2002). U.S. Department of Justice Prisoners in 2001 (Bureau Justice Statistics Bulletin NCJ Publicational No. 195189)
Levinson, R.B., Ingram, G.L., & Azcarate, E. (1968). "Aversive" group therapy--sometimes good medicine tastes bad. Crime and Deliquency, 14, 336-339.
McKee, J.M. (1971). Confingency management in a correctional institution. Education Technology, 11 (4), 51-54
Stokes, T. F., & Baer, D. M. (1977). An implicit technology of generalization. Journal Applied Behavior Analysis, 10, 349-367
Skinner, B.F. (1974, January). To build constructive prison environments [Letter to the editor]. New York Times, 36
U.S. Department of Justice, national Institute of Corrections (1997). Supermax housing: a survey of current practice. (NIC Publication No. NIC-013722). Longmont Co.
Vollmer, T. R., Borrero, J. C., Wright, C. S., VanCamp, C., and Lalli, J.S. (2001), Identifying possible contingencies during descriptive analyses of severe behavior disorder. Journal of Applied Behavior Analysis, 34, 269-287.
Webb, L.R. (2001). Addressing severe behavior problems in a "super-max" prison setting. U.S Department of Justice. National Institute of Corrections (NIC Publication No. 016869)
Lonny R. Webb
Oregon Department of Corrections
|Printer friendly Cite/link Email Feedback|
|Author:||Webb, Lonny R.|
|Publication:||The Behavior Analyst Today|
|Date:||Jan 1, 2003|
|Previous Article:||Treating obsessive-compulsive disorder with exposure and response prevention.|
|Next Article:||The molar view of behavior and its usefulness in behavior analysis.|