Eighty (6-person) juries heard 1 of the 5 reasonable doubt instructions in a trial that either had strong evidence for guilt or favored acquittal. None of the instructions produced acceptable criteria of self-reported reasonable doubt, although instructions that informed juries to be firmly convinced (FC) elicited the highest standards of proof. The FC instructions yielded verdicts that tended to correspond with the evidence in both versions of the trial FC juries focused proportionately more on the evidence and less on non-probative issues than juries given other instructions. Juries had lower selfreported standards of proof when the trial evidence was weak for conviction as opposed to when it was strong. The discussion addresses the surprisingly low self-reported standards of proof.

Jury nullification is a mechanism, and a defense, which allows the jury, as representatives of the community, to disregard both the law and the evidence and acquit defendants who have vioiated the letter, but not the spirit of the law. Should juries simply follow the law as articulated by the trial judge, or should they act as "conscience of the community," and neglect the strict requirements of the law when it would lead to unjust or inequitable verdicts? The present study was aimed at providing empirical data for the following question: will the jury operate in a manner which is different than its normal functioning if given explicit nullification instructions? Three nullification instructions varying in explicitness as to nullification were combined with three criminal cases to yield a 3 x 3 factorial design. Forty-five six-person juries (270 subjects), were randomly assigned to the nine experimental groups. The results showed that juries given explicit nullification instructions were more likely to vote guilty in a drunk driving case, but less likely to do so in a euthanasia case. The third case, which dealt with murder, did not show any differences due to instructions. Juries in receipt of nullification instructions spent less deliberation time on the evidence and more on defendant characteristics, attributions, and personal experiences.

An investigation of the impact of innovations in civil trial procedure manipulated trial structure (unitary vs. separated trial issues), order of decisions (liability or causation first), and number of decisions made (one to four) in a simulated toxic tort trial. Juries gave verdicts and damage awards. Recordings were made of the juries' deliberations. The results showed significantly more verdicts in favor of the plaintiffs in unitary trials (causation, liability, and damages heard together) as opposed to separated trials. However, average damage awards were higher in the separated than in the unitary trial conditions. Juries assigned more responsibility to the defendant in the unitary trial. Juries also used the totality of the evidence to decide all issues, especially general causation, which contained the most ambiguous testimony. The performance of these simulated juries in complex litigation was discussed in terms of group and individual cognitive factors.

The ability of a civil jury to render fair and rational decisions in complex trials has been questioned. However, the nature, dimensions, and effects of trial complexity on decision making have rarely been addressed. In this research, jury-eligible adults saw a videotape of a complex civil trial that varied in information load and complexity of the language of the witnesses. Information load and complexity differentially affected liability and compensatory decisions. An increase in the number of plaintiffs decreased blameworthiness assigned to the defendant despite contrary evidence and amount of probative evidence processed. Complex language did not affect memory but did affect jurors' ability to appropriately compensate differentially worthy plaintiffs. Jurors assigned compensatory awards commensurate with the plaintiffs' injuries only under low-load and less complex language conditions.

Forty-eight jury-eligible adults heard 1 of 4 versions of a tort trial. The design combined high and moderate levels of evidence technicality and the placement of substantive judicial instructions either before or after evidence presentation. Jurors given instructions before hearing the evidence for liability and before the evidence for compensation made clear distinctions among 4 differentially worthy plaintiffs, whereas jurors instructed after evidence presentation were not able to distinguish among the plaintiffs. Preinstructions enabled jurors to devise a causal model, as measured by both verbal representation of the evidence and recognition tests, that contained more probative evidence and less nonprobative and evaluative information than the models constructed by jurors who were postinstructed. Preinstructed jurors were better able than postinstructed jurors to correctly reject recognition items not part of the trial text and to correctly identify items from the trial.