IN EXP. I (N = 88) THE MAGNITUDE OF RESPONSE SUPPRESSION WAS GREATER FOR CONTINGENT PUNISHMENT THAN FOR PUNISHMENT DELAYED BY 30 SEC., AND IT WAS POSITIVELY RELATED TO THE INTENSITY OF THE PUNISHMENT (.1, .2, .3, .5, AND 2 MA.). IN EXP. II (N = 45) A SYSTEMATIC DELAY-OF-PUNISHMENT GRADIENT (0, 2, AND 30 SEC.) WAS DESCRIBED IN A FREE-RESPONDING SITUATION. IN EXP. III (N = 30) NO DIFFERENCE IN MAGNITUDE OF RESPONSE SUPPRESSION OR IN THE FREQUENCY DISTRIBUTION OF RESPONSE-SHOCK INTERVALS WAS DETECTED BETWEEN A GROUP WITH A 30-SEC DELAY-OF-PUNISHMENT AND A GROUP WITH NONCONTINGENT SHOCK. IN EXP. IV (N = 84) A DELAY-OF-PUNISHMENT GRADIENT (0, 2, 7.5, 30 SEC., AND NONCONTINGENT SHOCK) WAS DESCRIBED IN A DISCRIMINATIVE SITUATION. ALL OF THESE RESULTS ARE CONSISTENT WITH THE EMPIRICAL GENERALIZATION THAT THE MAGNITUDE OF SUPPRESSION OF AN APPETITIVE RESPONSE IS GREATER IF THE PUNISHMENT IS CONTINGENT UPON THE RESPONSE THAN IF IT IS NOT. (15 REF.) (PsycINFO Database Record (c) 2006 APA, all rights reserved)