THIRTY-TWO PSYCHOPATHS AND 16 NORMAL SUBJECTS WERE SELECTED FROM AN INMATE POPULATION OF 377 ON THE BASIS OF MINNESOTA MULTIPHASIC PERSONALITY INVENTORY PROFILES. THE PSYCHOPATHS WERE BROKEN DOWN INTO TWO SUBGROUPS -PRIMARY AND SECONDARY -- ON THE BASIS OF THEIR SCORES ON THE TAYLOR MANIFEST ANXIETY SCALE. HALF OF THE PRIMARY PSYCHOPATH, SECONDARY PSYCHOPATH, AND NORMAL SUBJECTS COMPLETED A COMPETING RESPONSE, PAIRED ASSOCIATES LEARNING TASK (A TEST IN WHICH THE SUBJECT IS REQUIRED TO MATCH A STIMULUS WORD WITH THE APPROPRIATE RESPONSE WORD) UNDER NORMAL CONDITIONS. THE OTHER HALF PERFORMED THE SAME TASK UNDER A CONDITION OF NEGATIVE EVALUATION; I.E., THEY WERE TOLD THAT THEY WERE NOT DOING AS WELL AS THE OTHER PEOPLE TESTED. UNDER THE NORMAL TESTING CONDITION, THE PRIMARY PSYCHOPATHS LEARNED SIGNIFICANTLY FASTER THAN DID THE OTHER TWO GROUPS. THERE WERE NO SIGNIFICANT DIFFERENCES IN LEARNING PERFORMANCE AMONG THE THREE GROUPS UNDER THE NEGATIVE EVALUATION CONDITION. THE SECONDARY PSYCHOPATHS LEARNED SIGNIFICANTLY BETTER UNDER THE NEGATIVE EVALUATION CONDITION THAN UNDER THE NORMAL TESTING CONDITION. THE FINDINGS SUPPORT A DISTINCTION BETWEEN PRIMARY AND SECONDARY PSYCHOPATHS IN TERMS OF ANXIETY OR EMOTIONAL RESPONSIVENESS. THE LEARNING PERFORMANCE FINDINGS ARE DISCUSSED IN TERMS OF THEIR IMPLICATIONS FOR RESEARCH IN BEHAVIOR PATHOLOGY AND FOR PREVENTION AND TREATMENT. SUPPORTING DATA AND A LIST OF REFERENCES ARE INCLUDED. (LKM)