Participants were 280 students enrolled in introductory psychology or sociology classes. A 22-item acquiescence-free version of the F scale (Byrne 1974) was administered to the students to classify participants as either high or low authoritarians. High and low authoritarians listened to a murder trial and then made judgments about guilt and punishment as individuals, as six-person juries, and again after deliberations as individuals. As predicted, both high authoritarian jurors and juries reached guilty verdicts more frequently and imposed more severe punishment than low authoritarians. Further, high authoritarians showed more prediscussion-postdiscussion verdict changes than low authoritarians. Results also supported the polarization hypothesis in a jury paradigm. Deliberations produced a shift toward greater severity of punishment for high authoritarians. Guilt verdicts shifted toward acquittal for all jurors. Implications of these findings are especially strong for trials involving capital punishment, the type of trial where juries containing all high authoritarians would most likely be empaneled. This follows from selection criteria in which capital-trial jurors must indicate their willingness to consider all potential penalties and must not be irrevocably committed before the trial has begun to vote against the death penalty. High authoritarians generally appear to be most favorable toward the death penalty. Several tables, 4 reference notes, and 20 references are provided. (Author abstract modified)