Meta-analytic review of the effects of suggestion on children’s memory: Implications for testimony

Abstract

Background/Objective: The implantation of misleading information in memory is a highly relevant topic due to its implications for the evaluation of witness memory. Suggestion has been the most studied means of implantation. Therefore, a meta-analytical study was designed to estimate the effect size of suggestion in the implantation of misleading information in testimonies and memories, and the moderators of such effect. Method: A total of 17 studies designed to evaluate the implantation of suggested memories or the mere acceptance of misleading information were selected. Bare-bones meta-analyses of experiments were performed. Results: The results showed a significantly positive, generalizable across studies, and more than large overall mean effect size (d = 1.68 95% CI [1.47, 1.89]) for the implantation of misleading information due to suggestion. When converted to probabilities, the estimated likelihood of implanting any externally suggested information was 31.3%. Likewise, the results evidenced a significant, positive, generalizable among studies, and more than large average effect size for the moderators: degree of suggestion (d = 1.60, 95% CI [1.47, 1.89]), likelihood of suggestion (d = 1.68, 95% CI [1.48, 1.88]), direct participation in the event (d = 1.31, 95%CI [1.13, 1.49]), indirect participation in the event (d = 2.00, 95% CI [1.77, 2.23]), suggestion through the individual differences paradigm (d = 1.44, 95% CI [1.23, 1.65]), and suggestion through the misinformation paradigm (d = 2.03, 95% CI[1.82, 2.24]). Conclusions: The implications of the results for the evaluation of the reliability of child testimony are discussed.