IN THIS STUDY, THIRTY ADULT RESIDENTS OF TWO MEDIUM SECURITY PENITENTIARIES WERE ASSIGNED RANDOMLY TO ONE OF TWO TYPES OF STRUCTURED GROUP COUNSELING. ONE SET OF GROUPS WAS COMPOSED OF A NON-RESIDENT LEADER AND RESIDENT-ONLY PARTICIPANTS. THE SECOND SET OF GROUPS INCLUDED A NON-RESIDENT LEADER WITH RESIDENTS AND CITIZEN VOLUNTEERS AS CO-PARTICIPANTS. IT WAS FOUND THAT THE PRESENCE OF VOLUNTEERS SIGNIFICANTLY IMPROVED CONTINUITY OF ATTENDANCE, SIGNIFICANTLY REDUCED THE LEVEL OF PROCRIMINAL ARGUMENTS WITHIN GROUPS AND, WITHIN ONE INSTITUTION, SIGNIFICANTLY INCREASED RESIDENT RATINGS OF THE OPENNESS AND GENUINENESS OF PARTICIPANT COMMUNICATION. THE EFFECT ON INTERPERSONAL OPENNESS RATINGS WAS IN THE OPPOSITE DIRECTION WITHIN THE OTHER INSTITUTION. HOWEVER, WITHIN THE RESIDENT GROUPS AT BOTH INSTITUTIONS, RESIDENT OPINIONS OF GROUP PROCESS WERE NEGATIVELY RELATED TO PROSOCIAL ATTITUDE CHANGE. WITHIN THE CO-PARTICIPANT GROUPS, OPENNESS RATINGS WERE POSITIVELY RELATED TO PROSOCIAL ATTITUDE CHANGE. THE RESULTS ARE DISCUSSED IN TERMS OF RELATIONSHIP-BASED COUNSELING THEORY, DIFFERENTIAL ASSOCIATION THEORY AND THOSE GUIDELINES FOR CORRECTIONAL INTERVENTION WHICH EMPHASIZE GROUP RELATIONS OVER THE EXPLICIT INTRODUCTION OF PROSOCIAL AND ANTICRIMINAL CONTENT ON A PROGRAMMED BASIS. A NUMBER OF OTHER FINDINGS AS WELL AS THE LIMITATIONS OF THE STUDY ARE DISCUSSED. (AUTHOR ABSTRACT)