THIS STUDY WAS DESIGNED TO TEST THE THEORY THAT LABORATORY GROUPS MAKING COMPLEX DECISIONS WILL DISTORT THEIR PERCEPTIONS OF EACH OTHER IN WAYS PREDICTABLE FROM NEWCOMB'S A-B-X MODEL OF PERCEPTUAL DISTORTION IN WHICH "A" REPRESENTS THE PERCEIVING INDIVIDUAL, "B" REPRESENTS ANOTHER MEMBER OF THE GROUP, AND "X" THE ISSUES UNDER DISCUSSION. FOUR HYPOTHESES DEALING WITH PERCEPTUAL DISTORTION WERE TESTED IN SIMULATED SCHOOL BOARDS, EACH COMPOSED OF FOUR SUBJECTS AND A GRADUATE STUDENT, THE LATTER PLAYING THE ROLE OF A NONDIRECTIVE SUPERINTENDENT OF SCHOOLS. EACH SUBJECT RECEIVED COMMUNICATIONS FROM, AND REPRESENTED, AN INTEREST GROUP. EACH BOARD MET FOUR TIMES, AND, AT EACH MEETING, TWO DIFFERENT BUT INTERRELATED ISSUES WERE DISCUSSED AND DECIDED. SUBJECTS WERE ASKED TO RATE EACH OTHER, THE SUPERINTENDENT, AND THEMSELVES AT THE START, MIDDLE, AND END OF EACH MEETING IN RELATION TO A HYPOTHETICAL NORMAL DISTRIBUTION OF A REFERENCE POPULATION. THE TWO TYPES OF RATINGS MADE WERE (1) POSITION ON ISSUE AND (2) SOUNDNESS OF JUDGMENT. MIXED RESULTS WERE OBTAINED. IN FACT, ONE HYPOTHESIS WAS SIGNIFICANT IN THE OPPOSITE DIRECTION TO THAT PREDICTED, SO THAT THE STATIC CORRELATION BETWEEN PERCEIVED POSITION DISCREPANCY AND EXPERTISE IS HIGH AND NEGATIVE. FURTHER SEARCH FOR THE CONDITIONS WHICH LED TO THESE RESULTS WAS RECOMMENDED. (GD)