Selective effect on punished versus unpunished responding in a conflict test as the criterion for anxiogenic activity.

The punished drinking test has been used successfully for identifying and studying anxiolytic agents. By reducing the level of punishment (i.e., decreasing the intensity of shock), it has also been used as a method for measuring anxiogenic activity. Because anxiogenic behavior is a novel and important concept that is not yet fully established, we have reinvestigated the effects of two putative inverse benzodiazepine agonists and pentylenetetrazol in this conflict test. In a series of experiments, using both our version of the procedure and a replication of a previously published method, we were unable to demonstrate a selective reduction in punished responding over unpunished responding caused by CGS 8216 (3 to 40 mg/kg), FG 7142 (2 to 6 mg/kg), and pentylenetetrazol (10 to 20 mg/kg) as reported previously. A careful comparison of the details of our method and the published procedure failed to reveal the source of this discrepancy. If anxiogenic behavior is to be defined as a selective effect of a drug on punished response, the value of this test will depend on identification of its critical variables.