OBJECTIVE Over the past 4 decades, discrepant research findings have emerged in the juror-confession literature, prompting the need for a systematic review and meta-analysis that assesses the effect of confession evidence (coerced or noncoerced) on conviction rates and the efficacy of trial safeguards. HYPOTHESES We did not predict any directional hypotheses. Some studies show increased convictions when a confession is present (vs. not), regardless of whether that confession was coerced; other studies demonstrate that jurors are able to discount coerced confessions. Studies have also demonstrated sensitivity effects (safeguards aided jurors in making appropriate decisions), skepticism effects (safeguards led jurors to indiscriminately disregard confession evidence), or null effects with regard to expert testimony and jury instructions. METHOD We identified 83 independent samples (N = 24,860) that met our meta-analytic inclusion criteria. Using extracted Hedges' g effect sizes, we conducted both network meta-analysis and metaregression to address key research questions. RESULTS Coerced and noncoerced confessions (vs. no confession) increased convictions (network gs = 0.34 and 0.70, respectively), yet coerced (vs. noncoerced) confessions reduced convictions (network g = -0.36). When jury instructions were employed (vs. not), convictions in coerced confession cases were reduced (this difference did not emerge for noncoerced confessions; a sensitivity effect). Expert testimony, however, reduced conviction likelihood regardless of whether a confession was coerced (a skepticism effect). CONCLUSION Confession evidence is persuasive, and although jurors appear to recognize the detrimental effect of coercive interrogation methods on confession reliability, they do not fully discount unreliable confessions. Educational safeguards are therefore needed, but more research is encouraged to identify the most effective forms of jury instructions and expert testimony. One potential reform could be in the interrogation room itself, as science-based interviewing approaches could provide jurors with more reliable defendant statement evidence that assists them in reaching appropriate verdict decisions. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

中文翻译：

---

供述证据对陪审员判决的影响：系统评价和荟萃分析。


目的 在过去 4 十年中，陪审员供述文献中出现了不一致的研究结果，促使需要进行系统回顾和荟萃分析，以评估供述证据（胁迫或非胁迫）对定罪率和审判效率的影响保障措施。假设我们没有预测任何方向性假设。一些研究表明，当有供述时（相对于没有供述时），定罪率会增加，无论该供述是否是被胁迫的；其他研究表明陪审员能够忽视强迫认罪。研究还证明了敏感性效应（保障措施帮助陪审员做出适当的决定）、怀疑效应（保障措施导致陪审员不加区别地忽视供述证据）或专家证词和陪审团指示的无效效应。方法 我们确定了 83 个符合荟萃分析纳入标准的独立样本 (N = 24,860)。使用提取的对冲 g 效应大小，我们进行了网络荟萃分析和元回归来解决关键研究问题。结果 胁迫和非胁迫供述（与不供述相比）增加了定罪（网络 gs 分别 = 0.34 和 0.70），但胁迫（与非胁迫）供述减少了定罪（网络 g = -0.36）。当采用陪审团指示（与不采用陪审团指示）时，胁迫认罪案件中的定罪量减少（这种差异对于非胁迫认罪没有出现；敏感性效应）。然而，无论认罪是否是胁迫的，专家证词都会降低定罪的可能性（怀疑效应）。 结论 口供证据具有说服力，尽管陪审员似乎认识到强制审讯方法对口供可靠性的不利影响，但他们并没有完全忽视不可靠的口供。因此，需要教育保障措施，但鼓励进行更多研究，以确定陪审团指示和专家证词的最有效形式。一项潜在的改革可能是在审讯室本身，因为基于科学的访谈方法可以为陪审员提供更可靠的被告陈述证据，帮助他们做出适当的判决决定。 （PsycInfo 数据库记录 (c) 2024 APA，保留所有权利）。