The effect of jury instructions on verdict thresholds and guilt perception

The standard of proof beyond a reasonable doubt is central to criminal adjudication in the United States. Yet, courts continue to grapple with how best to communicate this standard to jurors. In 2022, Colorado significantly revised its model reasonable doubt jury instruction, adding language that refers to a “real possibility” the defendant is not guilty. Critics argue that this phrasing may inadvertently lower the prosecution’s burden of proof. The present study empirically examined the impact of Colorado’s 2023 model instruction on juror decision-making, comparing it to both constitutionally upheld and judicially invalidated instructions. Using a simulated mock jury paradigm and a sample of 896 U.S. participants, we examined verdicts, confidence in guilt, and perceived evidentiary burdens across five jury instruction conditions (including the Colorado 2023 model instruction). The Colorado 2023 model instruction descriptively yielded the highest proportion of guilty verdicts and significantly increased juror confidence in guilt relative to an otherwise identical instruction lacking the “real possibility” statement. Examination of hypothetical jurors’ interpretation of the “real possibility” instruction indicated that over one-third interpreted the language as assigning some burden of proof to the defense. These findings suggest that Colorado’s 2023 model instruction—particularly the inclusion of “real possibility” language—may reduce clarity of the reasonable doubt standard and potentially lower the threshold for perception of guilt and conviction. The results have immediate relevance for pending litigation before the Colorado Supreme Court and broader implications for ensuring constitutional safeguards in jury instructions nationwide.