In two experiments, subjects read a transcript of a trial in which testimony revealed that defendants had confessed either on their own accord (no constraint), in response to a threat of punishment (negative constraint), or not at all (control group). In Experiment 1, subjects discounted the negatively induced confession. However, their estimates of the probability that the defendant committed the crime were increased not only by the unconstrained confession, but also by the positively constrained one. Experiment 2 essentially replicated this pattern for probability-of-commission estimates and verdicts despite the additional finding that the positively constrained confession was in fact perceived as involuntary. As a safeguard, the State courts could require a stringent standard of proof (e.g., proof beyond a reasonable doubt) by which to determine the voluntariness and hence admissibility of a prior confession. In addition, judicial instructions could correct jurors' biases. In a short form of instruction, the judge simply tells jurors to discount a coerced confession. A longer form would additionally define coercion as either a promise or a threat, thereby effectively sensitizing jurors to the dangers of relying on a positively induced confession. Because this study examined the judgments of individual, nondeliberating subjects, it is conceivable that questions about the reliability of evidence would arise during group discussions and serve to decrease jurors' use of a positively-induced confession. Twenty references are appended.