A token economy is a type of behavioral modification method used in an array of health and education settings. The token economy aims to shape behavior by rewarding individuals that meet target behaviors with reinforcers that they can use to obtain some type of reward. Standard components of a token economy are target behaviors, reinforcers, and rewards. Response costs and backup reinforcers are also aspects of the token economy. An example of a modification to the token economy is the level system. Concerns related to implementation of a token economy include generalizability and response maintenance.
Keywords Backup Reinforcer; Contingent Reinforcement; Level System; Natural Contingencies; Non-Contingent Reinforcement; Reinforcer; Response Cost; Response Maintenance; Reward; Self-Reinforcement; Targeted Behavior; Token Economy
Within the educational and mental health arenas, token economies are just one of several forms of behavioral methods used to decrease or increase certain types of behavior (Filcheck, McNeil, Greco, & Bernard, 2004). Liberman (2000) dated the 1960s as the time period in which token economies were created and first implemented. Montrose Wolf has been credited with developing the initial idea and terminology for a token economy in the early 1960s (Risley, 1997). It has been argued that the concept of a token economy arose well before the 1960s; Rodriguez, Montesinos, and Preciado (2005) asserted that the history of the token economy can be traced to work described in a textbook from the middle of the 19th century.
No matter the date of origin of the token economy, it is widely accepted that a token economy entails the distribution of tokens after the display of behavior deemed appropriate, (targeted behavior) where tokens can later be used to claim various items for redemption (Zlomke, & Zlomke, 2003). Rodriguez, Montesinos, and Preciado (2005) more specifically define a token economy as, "a reinforcement system in which the occurrence of appropriate behavior (or the absence of problem behavior) produces secondary reinforcement in the form of tokens (e.g., poker chips) that can periodically be exchanged for other reinforcers (e.g., food, toys, free time)." (p. 427) Liberman (2000) noted that prompts and reinforcers were used to shape behavior in this reinforcement system and that a token economy became the designation for the system because of the ease of using tokens as reinforcers.
Key to token economy efficacy is rewarding tokens soon after the target behavior has been observed and as often as the behavior takes place (Kazdin, 2000). Positive attributes of token economies include a set structure that lends to consistent reinforcement of specific behaviors, reinforcers that can be generalized across context, and the ease of using tokens in terms of distribution and attainment (Tarbox, Ghezzi, & Wilson, 2006). Kazdin (2000) described the use of positive reinforcement to promote positive behavior and reduce inappropriate behavior as the goal of token economy.
Tokens, backup reinforcers, or rewards, and the rules regarding rewards comprise the foundation of the token economy. In defining the components of the token economy, Kazdin (1982) made the analogy that tokens, or reinforcers, are income while a backup reinforcer would be an expenditure. Behaviors that earn tokens would be work and as individuals obtain tokens they would amass savings. Another aspect of a token economy is response cost. A token economy with response cost refers to an intervention where reinforcers are taken away if individuals participating in the intervention display inappropriate behaviors (LeBlanc, Hagopian, & Maglieri, 2000).
The token economy has been implemented in a number of formats and across various contexts. Much of the research that has focused on the token economy can be divided into investigations of single subjects, classrooms, and health settings. An example of single subject research on the token economy comes from Zlomke and Zlomke (2003). The researchers conducted an investigation of the impact of a joint token economy and self-monitoring intervention for an adolescent. Teachers engaged in a point system with the student throughout the day. The student had been trained in self-monitoring and for a portion of the study the student participated in a point system and self-monitoring. Targeted behaviors were seen significantly less frequently when baseline was compared to the period when the point system was implemented. Behaviors further decreased when self-monitoring took place. Although classroom behavior was impacted by the token economy and self-monitoring, home behaviors did not change.
At the Classroom Level
Kazdin (1973) explored the impact of a token economy in six elementary school classrooms. Students in classrooms that received contingent reinforcement, or the receipt of reinforcers when target behaviors were demonstrated, evidenced significantly more appropriate behavior after the intervention was implemented. Generalizability of behaviors was more evident for students in contingent reinforcement groups as compared to students who received non-contingent reinforcement, or reinforcement that was not based on the display of targeted behaviors.
Swain and McLaughlin (1998) referred to the token economy as a token-reinforcement program during their study of middle school students in special education. Math accuracy was the target behavior of interest. Students were awarded points for a number of behaviors and response costs for other behaviors were also implemented. A bonus contingency, or the opportunity for students to earn additional points if they achieved a certain score on the assessment of math accuracy, was another aspect of the study. Math accuracy was higher during the bonus contingency portion of the token-reinforcement as compared to the baseline period.
Token economies have also been implemented at the college and university level. For instance, Boniecki and Moore (2003) used a token economy in a college classroom in order to increase student participation. Students received a token after participating during class. Tokens were then to be used for extra credit by students. Compared to a baseline period before the initiation of the token economy, significantly more students participated in class, as evidenced by the number of students responding to questions posed during class. Participation declined after the token economy intervention ended. Hodge and Nelson (1991) also employed a differential reinforcement intervention during a university level course. The purpose of their study was to promote class participation for those students who did not participate often and decrease participation for the few students who tended to dominate class participation. After the intervention was conducted, the instructor deemed classroom participation was more evenly distributed across students.
Other Educational Applications
Lazarus (1990) detailed the aspects of a cooperative home-school token economy. In a cooperative home-school token economy, students who display targeted behaviors in either the home or school setting have an opportunity to accumulate tokens or points. A "Point Balance Passbook" is given to each student who brings the passbook home for review by parents and also uses it at school with teachers. The cooperative home-school token economy promoted communication between parent and child as well as parent and teacher. Parents also played a role in developing the program and received training on token economies.
Health settings, such as psychiatric hospitals or hospitals in general, are settings in which the token economy has been consistently put into practice for quite some time. Inpatient psychiatric units are an example of a setting where token economies have been utilized to promote a functional environment and reduce negative behavior such as violence (LePage, DelBen, Pollard, McGhee, VanHorn, Murphy, et al., 2003). LePage and colleagues studied a psychiatric unit over time where periods before and after token economy implementation were examined. Injuries, between patients and toward staff, were significantly reduced during the token economy intervention.
In an investigation of patients in a psychiatric hospital by Bedell and Archer (1980), comparisons were made between token economies that were either peer managed, staff managed, or used a non-contingent reinforcement schedule. Individuals in the peer managed and staff managed token economy programs earned higher amounts of tokens than did participants in the non-contingent reinforcement group. In another study described by Bedell and Archer in the same article, a group performance incentive group and standard individual incentive group were compared. In the group performance incentive group, some of the points individuals could earn were due to how many points earned by the group of patients with which they were aligned. No statistically significant differences were found between the incentive groups in regard to how many points participants earned. Similar amounts of points were earned in both studies reviewed in the article.
The Behavioral Rehabilitation and Interpersonal Treatment Environment, or BRITE program, is another example of a token economy implemented in a psychiatric hospital (Bellus, Vergo, Kost, Stewart, & Barkstrom, 1999). As part of the BRITE program, a token economy with a response cost and level system component was put in place to affect changes in assaults and self-injuries in the hospital. Self-injury and assaults for BRITE participants were significantly lower after two years of operation of BRITE as compared with levels of self-injury and assaults in traditional wards in the psychiatric hospital.
Other health issues addressed with the token economy have been substance use and eating disorders. Vouchers have been used to promote targeted behavior such as abstinence from cocaine use for individuals with cocaine dependence (Higgins, Roll, Wong, Tidey, & Dantona, 1999). In the intervention conducted by Higgins and colleagues, individuals participating in interventions where vouchers were contingent upon abstinence from substance use were significantly more likely to remain abstinent than those who received vouchers in a noncontingent manner. Another population with whom token economies have been used were individuals with anorexia nervosa (AN) (Okamoto, Yamashita, Nagoshi, Masui, Wada, Kashima, Et Al., 2002). In one study, activity restriction therapy and token economy were integrated (TET) and then compared to two other forms of behavior therapy to promote weight gain. Therapy that integrated a token economy showed an increase in body mass index for study participants. Individuals that received TET showed greater gains in BMI than one of the other forms of therapy and somewhat fewer gains than the form of...
(The entire section is 4811 words.)