Laboratory animal drug self-administration procedures for evaluating pharmacological abuse potential have focused on performance measurements involving relative rates of drug-maintained responding, discrete-trial choice determinations, and response cost or progressive ratio values. Relative rate measures have proved historically difficult to use for reliable reinforcement strength determinations. Discrete-trial choice procedures for assessing the relative reinforcing properties of stimulus events have recently been reported to effectively discriminate between different drugs and different doses of the same drug. Additionally, response cost or progressive ratio procedures involving systematic increases in the number of responses required for successive drug reinforcements have begun to reveal orderly relationships between different reinforcing drugs at various doses and a "breaking point" measure of the relative strength of a reinforcer. Comparisons between selected doses of cocaine, methylphenidate, and secobarbital with a series of five baboons using this procedure have shown that over the same behaviorally active dose range, cocaine breaking points were higher than all of the breaking points obtained with methylphenidate. Dose-response differences were also revealed in "breaking point" comparisons between secobarbital on the one hand, and methylphenidate and cocaine, on the other.