Figure 1: Baseline choice between different doses of cocaine (0–0.1 mg/kg/injection) and food pellets in rhesus monkeys ( ) under a concurrent FR10 : FR100 schedule of cocaine injections and food availability. Abscissae: unit dose of cocaine in milligrams per kilogram per injection. Top ordinate: percent cocaine choice. Middle ordinate: rates of responding in responses per second. Bottom ordinate: number of choices completed. All points represent mean data SEM obtained during the last 3 days of saline treatment. These unpublished data demonstrate two key observations from choice procedures. First, cocaine choice increases in a monotonic function as the unit cocaine dose increases. Second, while rates of responding display the prototypic, inverted-U-shaped dose-effect function, rates of responding are not predictive of cocaine choice, nor are rates of responding predictive of the number of choices completed per component.