Developmental Psychology 1991, Vol. 27, No. 6, 975-986 Copyright 1991 by the American Psychological Association, Inc. 0012-1649/91/J3.00 Effects of Question Repetition on the Eyewitness Testimony of Children and Adults Debra A. Poole Central Michigan University Lawrence X White Beloit College This study examined witnesses' answers to repeated questions about a novel event, both within and across interviews. Ss in 4 age groups (4-, 6-, and 8-year-olds and adults; N = 133) individually witnessed an ambiguous incident. Some Ss were interviewed immediately and 1 week later; others were interviewed only once, 1 week later. Children were as accurate as adults when responding to open-ended questions, but 4-year-olds were more likely to change responses to yes-no questions. Adults speculated more frequently than children on a specific question about which they had no information, and answers to this question became more certain with repetition. An "inoculation" procedure was successful in reducing the frequency of inappropriate speculation. When open- ended questions were used, a moderate amount of repetition primarily influenced presentation style rather than accuracy. Research into the development of basic memory processes has shed little light on a perplexing phenomenon: When young children are asked to report personal experiences (e.g., "What happened at nursery school today?"), they typically have little to say. Research consistently shows that even young children are capable of providing more information about complex events than they initially volunteer, and their event memories often show the same type of constructive organization that adults evidence (Hudson, 1990; Nelson, 1986). Children give only brief responses to open-ended questions, however, so adults frequently probe for information by repeating questions or re- sorting to more specific questions. The increased involvement of children in the criminal justice system has renewed interest in how to elicit event memories from children. Laboratory research on eyewitness testimony has repeatedly demonstrated that age per se is a poor predictor of accuracy. Rather, the quality of eyewitness testimony is better viewed as a function of interactions between characteristics of the child, the context of the witnessed event, and methods of eliciting reports (e.g., Ceci, Ross, & Toglia, 1987; King & Yuille, L987; Melton & Thompson, 1987). Several consistent findings are that adults report more accurate and inaccurate informa- tion than do children on free-recall questions (e.g., Goodman & Reed, 1986; Marin, Holmes, Guth, & Kovac, 1979; Saywitz, 1987), and distortions and suggestibility are most prevalent when the youngest children (i.e., preschool age) are tested or when questions refer to specific details of the witnessed event This research was supported by National Institute of Mental Health Grant RO3 MH44468 to Debra A. Poole. We wish to thank the children and adults who volunteered to partici- pate in this study and the members of our research team: Melanie Bossenberry, Michelle Skrocki, John CCallaghan, Molly Miles, Jennifer Rauh, Julie VanHiltmayer, Elizabeth Lossing, Elaine Stypula, and Sandy Winter. Correspondence concerning this article should be addressed to Debra A. Poole, Department of Psychology, 231 Sloan Hall, Central Michigan University, Mt. Pleasant, Michigan 48859. (e.g., Ceci et al, 1987; Dent & Stephenson, 1979; Goodman, Aman, & Hirschman, 1987; King & Yuille, 1987). Although even young children can provide useful informa- tion about personally relevant events, there are concerns about the impact of prompting, both in experimental procedures and in criminal investigations. For example, service providers testi- fying before the Attorney General's task force on family vio- lence estimated that child victims average at least a dozen inves- tigative interviews during the course of child protection pro- ceedings, criminal prosecution, and custody proceedings (Whitcomb, Shapiro, & Stellwagen, 1985). In addition to con- cerns about the stress associated with multiple interrogations, there are serious questions about the effects of repetition on the accuracy of children's testimony For example, in a child moles- tation case in Hawaii, a judge prohibited the testimony of two children when he became convinced that the children's stories had been unduly influenced by repeated questioning (McKellar v. City of Honolulu, 1989). Although young children are gener- ally very accurate in describing central events when there is no external pressure to distort reports, little is known about whether children can provide consistent testimony across re- peated interrogations. Knowledge about how children respond to repeated ques- tions comes primarily from three experimental paradigms: Piagetian-based studies of cognitive development, memory de- velopment from the verbal-learning tradition, and event memo- ries tapped by eyewitness procedures. Given that different pro- cesses are involved in these tasks, it is not surprising that repeti- tion is associated with several different response patterns. With respect to Piagetian-based studies, there is ample evi- dence that question repetition within a single session decreases accuracy when emerging cognitive skills are tested. For exam- ple, Gelman, Meek, and Merkin (1986) demonstrated that chil- dren's numerical competence is underestimated when repeated requests are made, and repetition has been cited as a reason for poor performance in some class-inclusion studies (e.g., Winer, 1980). Similarly, a higher proportion of children evidence con- 975