IT WAS HYPOTHESIZED THAT WITNESSES INTERROGATED WITH AGGRESSIVELY WORDED QUESTIONS WOULD RECALL AN INCIDENT AS BEING NOISIER, MORE VIOLENT, AND IN GENERAL MORE AGGRESSIVE THAN WITNESSES INTERROGATED WITH PASSIVE QUESTIONS. SUBJECTS FOR THE INVESTIGATION WERE 56 UNDERGRADUATE STUDENTS AT THE UNIVERSITY OF WASHINGTON. THEY WERE SHOWN A 3-MINUTE VIDEO TAPE THAT DEPICTED THE DISRUPTION OF A CLASS BY 8 DEMONSTRATORS. THE CONFRONTATION WAS RELATIVELY NOISY AND BASICALLY NONVIOLENT. AT THE END OF THE VIDEO TAPE, SUBJECTS RECEIVED ONE OF TWO QUESTIONNAIRES. THE PASSIVE QUESTIONNAIRE CONTAINED 25 QUESTIONS, ALL PHRASED IN A RELATIVELY MILD WAY. THE ACTIVE QUESTIONNAIRE CONTAINED 25 ANALOGOUS QUESTIONS WITH MORE AGGRESSIVE TERMINOLOGY. ONE WEEK LATER, ALL SUBJECTS ANSWERED A SERIES OF QUESTIONS ABOUT THE DISRUPTION, WITHOUT REVIEWING THE VIDEO TAPE. THE RESULTS SUPPORTED THE HYPOTHESIS THAT WITNESS DESCRIPTIONS OF A COMPLEX INCIDENT CAN BE INFLUENCED BY QUESTIONS USED TO INTERROGATE THEM ABOUT THE INCIDENT. SUBJECTS ASKED QUESTIONS WORDED IN AN AGGRESSIVE MANNER REPORTED THAT THE INCIDENT THEY WITNESSED WAS NOISIER AND MORE VIOLENT, THAT PERPETRATORS OF THE INCIDENT WERE MORE BELLIGERENT, AND THAT RECIPIENTS WERE MORE ANTAGONISTIC THAN SUBJECTS ASKED PASSIVE QUESTIONS. IT IS SUGGESTED THAT THE FIRST INVESTIGATOR TO INTERROGATE A WITNESS CAN AFFECT SUBSTANTIALLY THE WAY A WITNESS SEES AND REPORTS AN INCIDENT. A PRACTICAL IMPLICATION OF THIS FINDING IS THAT COURTS SHOULD CONSIDER WHO INITIALLY INTERROGATED THE WITNESSES. (DEP)