Abstract
Pigeons matched red and green samples to comparisons presented at short and long retention intervals within sessions. In one set of conditions the probabilities of reinforcers for correct choices of red and green were constant and equal at the short delay and differed at the long delay. In another set, the reinforcer probabilities were varied at the short delay and were held constant at the long delay. Choice at a particular delay was directly related to the reinforcer probabilities for correct red versus green choices at that delay. Choice at the delays with the constant reinforcer ratio was not affected by variation in reinforcer probabilities at the other delay. Performance at one retention interval is therefore independent of factors biasing performance at another.