Generalized conditioned reinforcement with pigeons in a token economy.
Flexible tokens outperformed fixed ones with pigeons, so let clients choose from a menu of backups.
01Research in Context
What this study did
DeFulio et al. (2014) worked with pigeons in a token economy. Birds earned plastic tokens they could trade for food or water.
Some tokens bought only food, others bought only water, and some bought either. The team counted how many of each kind the birds saved.
What they found
The pigeons stacked up more flexible tokens than single-purpose ones. When the price of a food-only token rose, birds still worked for the any-choice token.
This showed the any-choice tokens acted like stronger reinforcers, even though they were just bits of plastic.
How this fits with other research
Tan et al. (2015) later ran the same setup but added demand curves. They saw birds liked food tokens best, then flexible tokens, then water tokens. The two studies agree: flexible tokens beat single non-food tokens.
Andrade et al. (2017) pushed the idea further. They raised the price of food tokens and watched birds swap to flexible tokens. This substitution effect builds directly on Anthony’s finding that flexible tokens hold more value.
Matson et al. (2008) had already shown pigeons will hoard neutral tokens. Anthony added the twist of comparing token types, turning a basic accumulation study into a test of generalized conditioned reinforcement.
Why it matters
Your learners can’t tell you their preference curve, but you can still test it. Offer tokens that trade for a menu of backups instead of one fixed item. If a client starts working harder, you’ve found a generalized reinforcer. Try it next session: give one token that buys either a snack, a toy, or music time and watch the response rate.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Swap one single-backup token for a token that buys any of three backups and track response rate.
02At a glance
03Original abstract
Six pigeons were studied in a token economy in which tokens could be produced and exchanged for food on one side of an experimental chamber and for water on the opposite side of the chamber. Responses on one key produced tokens according to a token-production fixed ratio (FR) schedule. Responses on a second key produced an exchange period during which tokens were exchanged for water or food. In Experiment 1a, food tokens could be earned and exchanged under restricted food budgets, and water tokens could be earned and exchanged under water restricted budgets. In Experiment 1b, a third (generalized) token type could be earned and exchanged for either food or water under water restricted budgets. Across Experiments 1a and 1b, the number of tokens accumulated prior to exchange increased as the exchange-production schedule was increased. In Experiment 1b, pigeons produced more generalized than specific tokens, suggesting enhanced reinforcing efficacy of generalized tokens. In Experiment 2, the FR token-production price was manipulated under water restriction and then under food restriction. Production of each token type generally declined as a function of its own price and increased as a function of the price of the alternate type, demonstrating own-price and cross-price elasticity. Production of food and water tokens often changed together, indicating complementarity. Production of specific and generalized tokens changed in opposite directions, indicating substitutability. This is the first demonstration of sustained generalized functions of tokens in nonhumans, and illustrates a promising method for exploring economic contingencies in a controlled environment.
Journal of the experimental analysis of behavior, 2014 · doi:10.1002/jeab.94