Given the amount of reading and writing that I do today, it may come as something of a surprise to know that I couldn’t read until I was in grade 3. My parents read brilliantly exciting and interesting books to me that made “See Dick run, run Dick, run,” pale by comparison. I simply memorized whatever the teacher read out loud and parroted it back; it wasn’t until part way through grade 2 that anyone noticed that I wasn’t actually reading. Frankly, I was uninterested and couldn’t be bothered. This all changed for me in grade 3 when my teacher put a world map (with each ocean and sea clearly marked) on the wall and said we were starting a new reading game. Each student in the class got to cut out and colour a medieval-style sailing ship. For every book I read and wrote a one-page book report on, my ship would move one ocean “space” around the world. When my ship circumnavigated the world, I would be rewarded with the materials to make a felt mouse bookmark with googly eyes. This was a revelation for me. I get to read what I want? And a prize for doing it? Sweet! I think my ship circumnavigated the world two-and-a-half times that year; I haven’t stopped reading since. Although I didn’t know it at the time, my reading behaviour had been successfully modified by what behavioural psychologists call a token economy.
Token economies were popularized in the mid twentieth century in clinical and therapeutic settings, but they were actually used in educational settings at least as early as 1859 (Rodriguez, Montesinos, & Preciado, 2005), and are still used for this purpose today. For example, Breyer & Allen (1975) used one to improve the behaviour of 15 first graders with academic and social behaviour problems that were preventing their progress to grade two; more recently, Boniecki & Moore (2003) successfully applied a token economy to an introductory psychology course to increase student class participation. Besides education and therapy, token economies have also been successfully applied in business and industrial settings.
In their most standard form, token economies provide participants with tokens for appropriate behaviours. These tokens (called “conditioned reinforcers”) can then be traded in for desired back-up reinforcers (a reinforcer is anything that produces a future increase in the behaviour that led to it). In and of themselves, tokens have no actual value. Commonly used tokens are poker chips, points, checkmarks in a record book, and the like. In the case of my grade 3 reading game each new ocean space my ship moved to was a token. Remember, tokens are simply markers indicating that a desired behaviour has been performed. In a clinical application Paul & Lentz (1977), for example, gave tokens to patients in return for maintaining their personal hygiene, attending therapy sessions, keeping their living space clean, and engaging in social interaction. Back-up reinforcers can be anything at all that serves to motivate participants to gain tokens. In my round-the-world reading voyage, the back-up reinforcer was the mouse bookmark. Paul & Lentz offered a more diverse range of back-up reinforcers to their patients, including candy, clothing, cigarettes, additional furniture for patients’ rooms, radio and TV use, and sleeping late; all served to maintain the desired behaviours.
Token economies are advantageous and effective behaviour modification techniques for a number of reasons. Winkler (1971) showed tokens to be more resistant to satiation than most primary reinforcers (e.g., food or water); a participant might fill up on snacks, and lose motivation, but can’t fill up on tokens. In addition, different people are motivated by different rewards. When working with groups of people it can be difficult for a psychologist to find a single effective reinforcer that is easily deliverable and preferred by all. But because tokens are cashed in for back-up reinforcers, by making a range of back-up reinforcers available each individuals can find something that is personally motivating to work for. Similarly, an individual’s preference for an item may shift over time. Again, by having a range of back-up reinforcers available, if a participant gets tired of receiving one type of reinforcer they can select another to trade their tokens in for. Consequently, token systems often develop behaviours at a higher level than other conditioned reinforcers, such as praise, approval, or feedback (Kazdin 1989). Unlike some reinforcers that must be delivered in an all-or-none manner (for example, consider rewarding a child for cleaning her room by taking her to a movie), tokens can be delivered in a gradual, incremental manner. Instead of one behaviour for one big reward, multiple behaviours, each receiving a reinforcing token, can be required until eventually enough tokens are gained to get the big reward. For example, a child may receive one token for each instance of cleaning her room, brushing her teeth, putting her clothes away, etc. and ten tokens are needed to be taken to a movie. Generally speaking, the more individual behaviours that are reinforced, the stronger the learning. Finally, the task requirements to receive tokens can be gradually increased and new back-up reinforcers of greater value can be introduced as the program progresses. By starting requirements for token delivery at low levels participants rapidly receive reinforcement and are engaged by the program. However, once established, the costs can be increased until the ultimately desired levels of behavioural performance are reached; the introduction of novel and higher-value back-up reinforcers can help ameliorate negative consequences of increased behavioural cost for token delivery.
There are, however, some potential issues in applying token economies. Except for things like money or grades, tokens aren’t generally available in real-world settings, so the token-modified behaviour may not persist after the token economy is ended. In business or industrial contexts, this isn’t really an issue as the token economy is pretty specific to the work-place setting, but it may be a consideration in clinical or educational settings if one is trying to modify behaviours that will persist in the outside environment. In addition, it may be possible for individuals to gain tokens in unauthorized ways, perhaps by stealing them or trading them with others in the program, however, this is easily solved by making tokens unique to each individual. Finally, as Kazdin (1989) points out, back-up reinforcers that are extraneous to the setting may be difficult to remove after the token economy has been completed (e.g. using tokens to purchase food reinforcers in a classroom may be problematic as food is usually not allowed in classroom settings). Of course, by initially selecting setting-appropriate back-up reinforcers this problem can be avoided.
Finally, let’s examine a classic example of a token economy that ran for over a decade in the industrial setting of two uranium mines (Fox, Hopkins, & Anger, 1987). Tokens were given to miners for three things. First, not having accidents or injuries for specific periods of time; this provided impetus for individual miners to be more careful. Second, no miners in an entire work group being injured for a period of one month; this motivated miners to look not only to their own safety, but also that of their co-workers. Third, suggesting safety improvements that were implemented at the mines. Tokens could be exchanged by miners and their families at local redemption stores supplied with hundreds of items of merchandise; additional items were available via mail-order.
So, did it work? The graphs below show the number of days lost due to accident and injury. Compare the data from the baseline periods, before the token economy was adopted to the data after implementation. Both mines show a clear benefit.
Similarly, the token economy resulted in a substantial reduction in time lost to injuries after the token economy was put in place, as shown in the next set of graphs.
Of course, from a business perspective, ultimately one has to ask whether or not the token economy actually led to a financial benefit. A quick look at the graphs below demonstrates the dramatic reduction in costs to the mining company due to accidents and injuries. Reduced costs equals higher profits.
The applications of token economies are practically endless: personal fitness programs, retail sales, return of library materials, and dieting can all make use of token economies to motivate behavioural change.
Have you either implemented or taken part in a token economy, or know of someone who has? Feel free to add a comment to tell me about it.
References
Boniecki, K. A. & Moore, S. (2003). Breaking the silence: Using a token economy to reinforce classroom participation. Teaching and Psychology, 30(3), 224-227.
Fox, D.K, Hopkins, B. L., & Anger, W. K. (1987). The long-term effects of a token economy on safety performance in open-pit mining. Journal of Applied Behavior Analysis, 20(3), 215-224.
Kazdin, A. E. (1989). Behavior modification in applied settings (4th ed.). Pacific Grove, CA: Brooks/Cole.
Paul, G. L., & Lentz, R. J. (1977). Psychosocial treatment of chronic mental patients: Milieu versus social learning program. Cambridge, MA: Harvard University Press.
Rodriguez, J. O., Montesinos, L., & Preciado, J. (2005). A 19th century predecessor of the token economy. Journal of Applied Behavior Analysis, 38, 427.
Winkler, R. C. (1971). The relevance of economic theory and technology of token-reinforcement systems. Behaviour Research and Therapy, 9, 81-88.
No comments:
Post a Comment