ABSTRACT

Gaffan, Easton, and Parker (2002) discuss the mechanism by which frontal and inferior temporal cortices interact in a visual object strategy task that can be considered in some ways as a complex conditional task. In this task the monkey learns a series of pairs of visual objects. Of each pair, one object requires a persistent (P) strategy to be applied to it whereas the other requires a sporadic (S) strategy. Touching a P object on four consecutive trials results in the delivery of a small food reward. Once this reward has been delivered the animal can obtain another reward by touching the S object on any trial. Once an S reward has been achieved, however, the animal must return to consecutive P choices. Therefore, the animals can attain a specified number of rewards in a variety of ways, but the most efficient strategy is to choose four P objects followed by an S object followed by four P objects, and so on. This strategy will result in an average of 2.5 responses per reward obtained, and at the end of preoperative training animals performed very close to this maximally efficient strategy.