What Is a Token Economy?
A token economy is a system of contingency management in which tokens (conditioned reinforcers) are delivered contingent on target behaviors and later exchanged for backup reinforcers. Token economies are among the most widely used and researched interventions in applied behavior analysis, with applications spanning autism services, classroom management, psychiatric facilities, and organizational behavior management.
The power of a token economy lies in its ability to bridge the temporal gap between behavior and reinforcement. Many naturally occurring reinforcers cannot be delivered immediately after every instance of a target behavior. Tokens solve this problem by providing an immediate consequence that can later be exchanged for a variety of preferred items or activities.
Components of a Token Economy
Tokens
Tokens are the conditioned reinforcers that serve as the medium of exchange. Effective tokens share several characteristics. They should be easy to deliver, easy to accumulate, and difficult to counterfeit. Common examples include poker chips, stickers on a chart, check marks, points tracked on an app, or stamps on a card.
The choice of token should match the learner's developmental level and the setting. A preschooler may respond well to stickers placed on a visual board, while a teenager might prefer a point system tracked digitally.
Target Behaviors
Clearly defined target behaviors are essential for a functioning token economy. Each behavior must be operationally defined so that all implementers can consistently identify and reinforce the correct responses. For example, rather than awarding tokens for "being good," specify observable behaviors like "completing assigned work within the allotted time" or "raising hand before speaking."
Token Exchange Schedule
The exchange schedule specifies how many tokens are needed to purchase each backup reinforcer. This schedule must be calibrated to the learner's ability and motivation. If the exchange requirement is too high, the learner may give up. If it is too low, the system may not generate enough motivation to sustain effort across time.
A well-designed exchange schedule offers a menu of backup reinforcers at varying price points. Small, low-cost items can be purchased quickly, while more desirable items require saving tokens over longer periods. This menu approach teaches delayed gratification and provides options that match different motivational states.
Backup Reinforcers
Backup reinforcers are the items and activities that tokens can be exchanged for. These must be truly reinforcing for the individual, which means they should be identified through formal preference assessments rather than assumed. The backup reinforcer menu should be updated regularly based on ongoing preference assessment, since preferences change over time.
Steps for Implementation
Step 1: Assess the Learner
Before implementing a token economy, assess the learner's current skill level. Can they understand the concept of exchange? Do they have experience with conditioned reinforcement? For learners who do not yet understand token-to-backup-reinforcer relationships, you will need to establish the tokens as conditioned reinforcers through pairing before the full system can function.
Step 2: Select and Define Target Behaviors
Choose behaviors that are socially significant and within the learner's current repertoire (or close to it). Start with one or two target behaviors and expand as the system becomes established.
Step 3: Conduct Preference Assessments
Use formal preference assessment procedures such as paired stimulus (also called forced choice), multiple stimulus without replacement (MSWO), or free operant observation to identify highly preferred items and activities.
Step 4: Establish Token Value
For learners unfamiliar with token systems, pair tokens with backup reinforcers. Start by giving a token and immediately allowing exchange, then gradually increase the exchange requirement. This conditioning process establishes the token as a conditioned reinforcer.
Step 5: Implement and Monitor
Begin the system with rich reinforcement schedules and clear contingencies. Collect data on target behaviors, token delivery, and exchange patterns. Use these data to adjust the system as needed.
Troubleshooting Common Problems
Token Satiation
If the learner has accumulated many tokens and can easily access backup reinforcers, motivation may decrease. Solutions include increasing the exchange ratio, introducing new and more preferred backup reinforcers, or implementing a "spending" requirement that reduces surplus tokens.
Lack of Interest in Tokens
If tokens are not functioning as conditioned reinforcers, re-evaluate whether the pairing process was sufficient and whether the backup reinforcers are truly preferred. Sometimes the issue is that backup reinforcers are available outside the token system for free, which undermines the system.
Inconsistent Implementation
Token economies require consistent delivery by all implementers. If some staff members deliver tokens for non-target behaviors or fail to deliver tokens for target behaviors, the contingency breaks down. Training, performance feedback, and implementation fidelity checklists help address this issue.
Problem Behavior Around Exchanges
Some learners exhibit problem behavior when they cannot immediately exchange tokens or when they do not have enough tokens for a desired item. Building tolerance through gradual increases in exchange requirements and teaching appropriate waiting behaviors can prevent these issues.
Fading the Token Economy
The ultimate goal is for behavior to come under the control of natural contingencies rather than remain dependent on a contrived token system. Fading strategies include gradually increasing the exchange ratio, shifting to intermittent token delivery, replacing tokens with social reinforcement, transferring control to self-monitoring systems, and systematically removing components of the system over time.
Successful fading requires careful data monitoring to ensure that target behaviors maintain as the system is thinned. If behavior decreases during fading, the clinician should temporarily return to the previous level and plan a more gradual transition.