MALE AND FEMALE BYSTANDERS WITNESSED A VIOLENT THEFT IN THE COURSE OF A BOGUS EXTRA SENSORY PERCEPTION EXPERIMENT. IN ADDITION TO THE SUBJECTS' SEX, TWO EXPERIMENTAL MANIPULATIONS WERE ORTHOGONALLY CROSSED. THE SUBJECT WAS THE ONLY PERSON TO WITNESS THE EMERGENCY, OR WITNESSED IT WITH AN ADDITIONAL PARTICIPANT (DIFFUSION OF RESPONSIBILITY), AND THE SUBJECT'S PRESENCE WAS EITHER KNOWN OR UNKNOWN TO THE OTHER PARTICIPANTS (EVALUATION APPREHENSION). BOTH MANIPULATIONS AFFECTED THE BYSTANDERS' LATENT RESPONSES; PARTICIPANTS WHOSE PRESENCE WAS KNOWN RESPONDED MORE QUICKLY THAN THOSE WHOSE PRESENCE WAS UNKNOWN, AND LONE BYSTANDERS RESPONDED MORE QUICKLY THAN THOSE WITNESSING THE THEFT WITH ANOTHER PARTICIPANT. FURTHER, BYSTANDERS WHOSE PRESENCE WAS UNKNOWN AND WHO WITNESSED THE EMERGENCY WITH ANOTHER PARTICIPANT WERE SIGNIFICANTLY SLOWER TO RESPOND THAN SUBJECTS IN THE OTHER THREE CATEGORIES. SINCE A THIRD OF THE BYSTANDERS' REACTIONS DID NOT LEAD TO IMMEDIATE HELPING ATTEMPTS, LATENT RESPONSES FOR EFFECTIVE HELPING WERE ALSO ANALYZED. THERE WERE NO INDEPENDENT EFFECTS OF THE TWO EXPERIMENTAL MANIPULATIONS ON HELPING LATENCIES. ONLY THE STATISTICAL INTERACTION WAS REPRODUCED, THAT IS, UNKNOWN BYSTANDERS WHO WITNESSED THE THEFT WITH ANOTHER PARTICIPANT WERE ALSO SLOWER TO HELP THAN THOSE IN THE OTHER EXPERIMENTAL SETTINGS. THE IMPLICATIONS OF THE FINDINGS ARE DISCUSSED IN RELATION TO THE UNDERSTANDING OF BYSTANDERS' DECISIONMAKING IN EMERGENCY SITUATIONS. THE ETHICS OF THE TEST PROCEDURE ARE ALSO DISCUSSED. TABULAR AND GRAPHIC DATA ARE PROVIDED ALONG WITH REFERENCES. (AUTHOR ABSTRACT MODIFIED--KBL)