Slane, C. R., & Dodson, C. S. (2022). Eye-witness confidence and Mock juror decisions of guilt: A meta-analytic review. Law and Human Behavior, 46(1), 45-66. https://doi.org/10.1037/lhb0000481
In this journal article, Slane and Dodson (2022) investigate the impact of eye-witness confidence on different dependent variables, including (i) jurors’ perception of accuracy, (ii) judgment of guilt, and (iii) guilt or non-guilt verdict. The primary hypothesis (hypothesis) was that highly confident eye-witnesses are more persuasive to jurors than those with lower confidence. The second hypothesis was that eye-witness confidence at trial is more persuasive to the mock juror than the identification. Finally, the third hypothesis (hypothesis 3) is the numerical expression of confidence is more persuasive than verbal confidence.
To test the hypothesis, Slane and Dodson (2022) adopted a meta-analysis of 35 studies (20 published papers, seven theses or seven dissertations) to investigate the effect of confidence on juror judgment and the influence of the primary moderator variables. The latter included the format of the confidence expression and the time of the confidence. The researcher adopted the comprehensive metanalysis software (CMA) to run four primary analyses. The primary moderating variables include the confidence timing and the confidence expression type. The secondary moderating variables in the study included the level of confidence and the safeguards. The most significant safeguards include the expert testimony or the juror instructions.
The meta-analysis shows eye-witness confidence influences juror decisions. When the eye witness was highly confident, reaching the guilty verdict was much easier than when the eye witness was less confident. However, the eyewitness is unlikely to express low confidence before the juror. The moderator analysis reveals that the timing of the confidence statement is not a significant factor for the juror’s decisions of guilt or accuracy. The results of the moderator analysis show that numerical or verbal expressions do not moderate the influence of eye-witness confidence. Moreover, the result shows including the safeguards does not influence the effect of eye-witness confidence across the studies.
Slane and Dodson (2022) explain that eye-witness confidence persuades the jurors, although the impact is modest. Thus, increased confidence affects the juror’s decision for guilt and the suspect’s identification. In addition, the persuasive effect of the verbal and numerical expression of confidence is similar. Finally, the study results show the jurors are not sensitive to the differences in the eye-witness confidence at trial from the initial identification.
Quigley-McBride, A., Crozier, W., Dodson, C., Teitcher, J., & Garrett, B. L. (2021). Face value? How jurors evaluate eye-witness face recognition ability. SSRN Electronic Journal. https://doi.org/10.2139/ssrn.4032866
In this study, Quigley-McBride et al. (2021) examine the significance of face memory ability tests in determining eye-witness accuracy. Post-identification confidence is a common determinant of eye-witness accuracy, but it is not often reliable. Despite this shortcoming, the jurors rely on the eye-witness confidence to measure accuracy. When an eye-witness testifies, the jury examines information such as how well the perpetrator and the police administration of the lineup. This study emphasises the significance of face memory ability tests to access objective information about an individual ability to recognise a face as a key determinant for accuracy.
Quigley-McBride et al. (2021) ran three online studies ((N₁ = 3,143, N₂ = 1,156, N₃ = 3,180) to determine how the jury and lay persons evaluate the face memory ability. Scientific tools such as the Cambridge Face Memory Test (CFMT) are significant in measuring face recognition ability. In experiment 1, the participant read a mock court transcript of an eyewitness in specific versions to determine how this information was interpreted.
The study results indicate that jury-eligible participants can interpret the information as objectively good or poor facial memory. The eye-witness face memory ability and post-identification confidence significantly influenced the verdict decisions. The classic effect in the jury decision-making literature is that a high confidence eye-witness influences high rates of guilty verdicts. In addition, the jury-eligible participants can integrate the face memory information into the results of post-identification confidence. The incidences of good face memory scores and low confidence present a unique challenge to the mock jurors. The result of this study shows the jury intentionally dismisses the low post-identification confidence and relies on the high face memory score.
The result of the study shows that lay persons understand the significance of face memory skills and place more consideration on individuals with good face memory over those with poor face memory. The mock juror values the low post-identification confidence when the eye-witness achieves a significantly high CMFT score. However, the result of the study maintains that the mock juror considers the high-confidence eye-witness to be significantly more reliable than the low-confidence eye-witness. The study is significant since it explains the need to effectively communicate the integration between post-identification confidence and objective face memory information.
Dodson, C. S., Garrett, B. L., Kafadar, K., & Yaffe, J. (2021). Eye-witness identification speed: Slow identifications from highly confident eye-witnesses hurt perceptions of their testimony. Journal of Applied Research in Memory and Cognition, 10(2), 259-267. https://doi.org/10.1016/j.jarmac.2020.08.015
In this article, Dodson et al. (2021) evaluate whether highly confident eye-witnesses are more persuasive. Primarily, the study tests if a highly confident eye-witness is very persuasive to the extent their testimony overshadows countervailing evidence. The literature informed this research, which explains that highly confident eye-witnesses are perceived as more accurate. In contrast, the less confident eye-witness is perceived as less accurate in identifying the suspect. Dodson et al. (2021) identify a research gap since, in the previous studies, the participant was not provided with information about the eye-witness level of confidence, and the absence of the information could matter since individuals hold implicit assumptions about other people. In other words, people assume that faster identification indicates faster identification while slower identification reflects less confident identification. The purpose of this research was to close the gap in the literature to account for the effect of a control condition to be certain if the effect of faster identification is consistent with high accuracy.
Dodson et al. (2021) evaluated highly confident witness identifications in three experiments (1, 2, and 3) to fulfil the research purpose and answer the research question. Particularly in experiment 1 and experiment 2, the authors examined the effect of different identification speeds on the participant’s perceptions of the accuracy of the eye-witness lineup identifications and the likelihood of the guilt of the accused. These experiments are significant in the study’s contribution to the literature since they explain the significance of a control condition. In experiment 3, Dodson et al. (2021) asked the participants to translate an eye-witness verbal expression into a number. Primarily, the experiment was established in the expression that manipulating the identification speed affects the perceived numeric confidence of the eye-witness.
Dodson et al. (2021) (1 and 2) show the participants either quickly identified a suspect, slowly identified the suspect, or did not learn anything about the eye-witness identification time. When a highly confident eye-witness makes a slower identification, they are perceived as less confident, and the suspect is considered less guilty than when the eye-witness makes a fast identification or when information about identification speed is unavailable. These results are consistent with the preregistered predictions. The study findings raise a question on people’s interpretation of eye-witness testimony for inference on the likely accuracy of the eye-witness identifications. Dodson et al.’s (2021) findings are consistent with previous findings showing that people are sensitive to the eye-witness confidence level.
Iida, R., Itsukusima, Y., & Mah, E. Y. (2020). How do we judge our confidence? Differential effects of meta‐memory feedback on eye-witness accuracy and confidence. Applied Cognitive Psychology, 34(2), 397-408. https://doi.org/10.1002/acp.3625
In this article, Iida et al. (2020) test confidence in memory based on the self-credibility cues among the eye-witness testimony. Primarily, the study is based on the premise that people consider eyewitness testimony more reliable when made with high confidence. For mock juror studies, the participants view the highly confident eyewitnesses as more reliable. Some studies suggest eye-witness confidence is not a good predictor of accuracy, but others show a strong relationship between confidence and accuracy. This research identifies that poor confidence accuracy in literature is due to the inconsistencies reflecting the memory accuracy-confidence judgments due to the subjective evaluation of memory accuracy. Confidence judgment is influenced by different processes, including (1) familiarity elicited by stimulus and (2) determination that the memory trace is accurate. Thus, in this study, the author determined that the participants’ memory is accurate and produces confident judgements. Primarily, the research objective is to have an in-depth examination of the processes that influence confidence judgements.
Based on the cue-belief model, the researcher performed two experiments to determine the influence of the self-credibility cues that had not previously been studied. The experiments involved manipulating the self-credibility cues and then determining the effect of the intervention on the eye-witness testimonial confidence and accuracy.
The first experiment measures if the pre-identification feedback affected the eye-witness identifications. The feedback conditions (no feedback score, bad score, good score) and the question types were measured as the independent variables, while the accuracy was the dependent variable. The two-way mixed analysis of variance (ANOVA) results show that the accuracy is lower in leading questions but not affected by feedback. In contrast, the result of the ANOVA based on confidence as the dependent variable shows that the identification of normal and lead questions was affected by feedback.
Iida et al. (2020) evaluated whether post-testimony self-credibility feedback impacts eye-witness confidence and accuracy for the second experiment. In this experiment, three-way ANOVA based on accuracy as the dependent variable shows the question type. In addition, the results show good feedback after the first questionnaire increases the accuracy of the leading questions. The results show that the feedback relating to the self-credibility cues affects eye-witness confidence judgments across the two experiments.
Jin, S., Verhaeghen, P., & Rahnev, D. (2022). Across-subject correlation between confidence and accuracy: A meta-analysis of the confidence database. Psychonomic Bulletin & Review, 29(4), 1405-1413. https://doi.org/10.3758/s13423-022-02063-7
In this article, Jin et al. (2022) examine the relationship between accuracy and confidence across different subjects. In high-stakes situations like eye-witness testimony, the relationship between confidence and accuracy is of great witness. At the individual level, confidence influences the commitment to make decisions and seek additional information and advice from others. At the interpersonal level, it influences how one perceives the information presented before them. Although increasing evidence shows the relationship between confidence and accuracy in real-world situations, such as eye-witness testimony, Jin et al. (2022) identify a research gap in the effect of overall confidence. In the first research question, the study examines the extent to which we should trust confident people vs non-confident people in situations where each person completes many trials. The second question examined the factors that moderate the confidence-accuracy relationship across subjects.
To fulfil the study purpose, Jin et al. (2022) selected 213 unique datasets from the Confidence Database to determine the across-subject relationships between the study variables. The recently published Confidence Database includes many databases that feature confidence ratings. These databases differ in many elements, such as the study domain, confidence scale, and trial-by-trial feedback.
The study found a significant correlation between average accuracy and average confidence across subjects (R=0.22). These findings are significant since, to the researcher’s knowledge, none of the datasets had been collected to determine confidence-accuracy relationships. The relationship between the variables is stronger for the perception tasks and for a confidence scale with fewer points (Jin et al., 2022). In other words, the correlation between the average accuracy and the average confidence is high for the granular confidence level. Moreover, the domain of the study is a key moderator to the strength of the correlation between the variables.
Primarily, the study is significant in showing the need to take confidence level seriously as a factor for decision-making, and many factors influence individual confidence judgments. The study results are consistent with previous meta-analyses in the confidence judgments of eye-witness testimony to show high confidence, leading to faster identification and improved accuracy. People’s interpretation of confidence is significantly different and remains consistent; thus, the confidence scale must be controlled.
References
Dodson, C. S., Garrett, B. L., Kafadar, K., & Yaffe, J. (2021). Eye-witness identification speed: Slow identifications from highly confident eye-witnesses hurt perceptions of their testimony. Journal of Applied Research in Memory and Cognition, 10(2), 259-267. https://doi.org/10.1016/j.jarmac.2020.08.015
Iida, R., Itsukusima, Y., & Mah, E. Y. (2020). How do we judge our confidence? Differential effects of meta‐memory feedback on eye-witness accuracy and confidence. Applied Cognitive Psychology, 34(2), 397-408. https://doi.org/10.1002/acp.3625
Jin, S., Verhaeghen, P., & Rahnev, D. (2022). Across-subject correlation between confidence and accuracy: A meta-analysis of the confidence database. Psychonomic Bulletin & Review, 29(4), 1405-1413. https://doi.org/10.3758/s13423-022-02063-7
Quigley-McBride, A., Crozier, W., Dodson, C., Teitcher, J., & Garrett, B. L. (2021). Face value? How jurors evaluate eye-witness face recognition ability. SSRN Electronic Journal. https://doi.org/10.2139/ssrn.4032866
Slane, C. R., & Dodson, C. S. (2022). Eye-witness confidence and Mock juror decisions of guilt: A meta-analytic review. Law and Human Behavior, 46(1), 45-66. https://doi.org/10.1037/lhb0000481
write