Journal of Experimental Child Psychology 72, 235–270 (1999) Article ID jecp.1999.2491, available online at http://www.idealibrary.com on Emotion and Memory: Children’s Long-Term Remembering, Forgetting, and Suggestibility Jodi A. Quas University of California, Berkeley Gail S. Goodman University of California, Davis Sue Bidrose, Margaret-Ellen Pipe, and Susan Craw University of Otago, Dunedin, New Zealand and Deborah S. Ablin University of California, Davis Children’s memories for an experienced and a never-experienced medical procedure were examined. Three- to 13-year-olds were questioned about a voiding cystourethrogram fluoros- copy (VCUG) they endured between 2 and 6 years of age. Children 4 years or older at VCUG were more accurate than children younger than 4 at VCUG. Longer delays were associated with providing fewer units of correct information but not with more inaccuracies. Parental avoidant attachment style was related to increased errors in children’s VCUG memory. Children were more likely to assent to the false medical procedure when it was alluded to briefly than when described in detail, and false assents were related to fewer “do-not-know” responses about the VCUG. Results have implications for childhood amnesia, stress and memory, individual differences, and eyewitness testimony. © 1999 Academic Press Key Words: children’s memory; infantile amnesia; emotion; attachment; suggestibility; eyewitness testimony. That emotions play a role in influencing the memorability of personal expe- riences has been recognized since, if not before, the early writings of Freud This study was funded by grants to Gail S. Goodman from the Society for the Psychological Study of Social Issues (Division 9 of the American Psychological Association) and the University of California, Davis Faculty Research Grant Program, and a grant to Margaret-Ellen Pipe from the New Zealand Health Research Council. We thank Deidre Brown, Jan Egerton, Judy Gabore, Megan Gollop, Jenna Luenberger, 235 0022-0965/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved. 236 QUAS ET AL. (Freud, 1915/1957, 1938; see also Christianson, 1992; Rapaport, 1942). Accord- ing to Freud, experiences occurring prior to the age of approximately 6 years are “covered in a veil of amnesia.” These memories, he believed, are repressed because of their emotional nature, specifically their underlying sexual and ag- gressive nature (1915/1957), and typically remain inaccessible to conscious recall into the adult years. Since Freud called attention to this phenomenon, a considerable amount of research has been devoted to the study of adults’ earliest childhood memories. This research has resulted in a number of revisions to Freud’s original ideas, including reduction in the age of the childhood-amnesia “barrier” to 2 or 3 years (Kihlstrom & Harackiewicz, 1982; Usher & Neisser, 1993) and the introduction of numerous other theoretical accounts explaining the inaccessibility of early memories (see Howe & Courage, 1993; 1997; Pillemer & White, 1989, for reviews). Less well documented, however, is the role of trauma, or negative emotion more generally, in the selective forgetting of early childhood events as predicted by Freud’s theory. The reasons for this dearth are largely methodological: In general, it has been difficult to examine empirically the accuracy of children’s and adults’ memories for highly stressful or traumatic events, including those of particular relevance to Freudian theory (e.g., events involving genital touch). Most of what is known about the recall of experiences with a strong emotional component has been based on studies of adults’ retrospective accounts of their early experiences and on case studies or anecdotal reports of children’s memory for traumatic events (Hewitt, 1994; Terr, 1988; Waldfogel, 1948). Recently, however, the study of children’s memory for stressful medical procedures has opened the door to empirical investigations of emotion and memory in childhood (Bugental, Blue, Cortez, Fleck, & Rodriguez, 1992; Goodman, Hirschman, Hepps, & Rudy, 1991; Goodman, Quas, Batterman-Faunce, Riddlesberger, & Kuhn, 1994; Merritt, Ornstein, & Spicker, 1994; Peterson & Bell, 1996; Steward & Steward, 1997). What remains unknown, however, is how children’s memo- ries of emotional events are retained over long durations (but see Peterson & Parsons, 1996). Understanding children’s remembering and forgetting of stress- ful events, including those involving genital contact, over long delays may provide theoretical insight into memory for traumatic experiences, for instance, into whether special memory mechanisms are involved. Knowledge about children’s long-term memory for stressful events can also provide valuable information about children’s eyewitness capabilities. In eye- witness situations, children are most often questioned about events that happened to them personally (Goodman, Quas, Bulkley, & Shapiro, in press), and the questioning can occur months or even years after the alleged events occurred John Palmer, Allison Redlich, Jennifer Schaaf, and Don Waters for their assistance on the project. Our gratitude is also expressed to Phillip R. Shaver for consultation on attachment measures and findings. Address reprint requests to Gail S. Goodman, Department of Psychology, University of California, One Shields Avenue, Davis, CA 95616. CHILDREN’S LONG-TERM MEMORY 237 (Gray, 1993). Past research on children’s memory and eyewitness testimony has identified a number of factors that can influence infants’ and children’s mne- monic capabilities, such as linguistic capabilities (e.g., Bauer & Wewerka, 1995; Fivush, Haden, & Adam, 1995), memory trace strength (e.g., Pezdek & Roe, 1995), source-monitoring capabilities (e.g., Ackil & Zaragoza, 1995; Parker, 1995), and interviewer biases and retrieval strategies (e.g., Goodman, Sharma, Considine, & Golden, 1995; Leichtman & Ceci, 1995; Liwag & Stein, 1995; see Newcombe’s 1995 special issues of the Journal of Experimental Child Psychol- ogy). Most of these studies have not, however, examined how social and emotional factors influence children’s long-term memory, particularly memory of a salient, stressful experience. In the present study, we examined children’s long-term memory for a docu- mented medical procedure, voiding cystourethrogram fluoroscopy (VCUG), that involves painful and stressful genital contact. Our main objectives were to: (1) identify how age and delay influence children’s long-term memory and investi- gate whether typical childhood amnesia effects would emerge; (2) examine how stress relates to children’s memory following long delays; and (3) determine whether individual-difference factors predict children’s long-term memory and suggestibility. We first review background literature relating to each of these objectives. Childhood Amnesia in Adults and Children Although research on adults’ earliest memories suggests that most adults fail to recall events unless they were at least 3 or 4 years of age when the events occurred (Dudycha & Dudycha, 1941; Sheingold & Tenney, 1982; Winograd & Killenger, 1983; however, see Usher & Neisser, 1993), research on children’s early memories reveals a different pattern: Children often can remember events they experienced in early childhood (Fivush et al., 1995; Hamond & Fivush, 1991; Howard, Osborne, & Baker-Ward, 1997). Children who are almost 3 years old show implicit memory for experiences in infancy (Myers, Clifton, & Clark- son, 1987), and 2-year-olds show explicit memory for events experienced 6 months previously (Fivush, Gray, & Fromhoff, 1987). Once a long enough delay has ensued, however, children can no longer recall formerly accessible memories (Fivush et al., 1987; Goodman et al., 1991; Pillemer, 1992; Pillemer, Picariello, & Pruett, 1994). It is therefore important to assess how well young children can remember the same event, especially a stressful event involving salient body contact, over time spans ranging from a few months to many years. Emotion and Children’s Memory Research on adults’ early childhood memories suggests that emotion may have a facilitative effect on long-term retention (e.g., Howes, Siegel, & Brown, 1993; Kihlstrom & Harackiewicz, 1982; Usher & Neisser, 1993). Similarly, in studies of children’s memory for stressful events, some researchers have observed a 238 QUAS ET AL. positive relation between stress and children’s memory (e.g., Goodman et al., 1991). However, other researchers have uncovered negative (e.g., Bruck, Ceci, Francoeur, & Barr, 1995; Merritt et al., 1994) or no significant relations (Eisen, Goodman, Qin, & Davis, 1998; Howe, Courage, & Peterson, 1995; Peterson & Bell, 1996) between stress and memory in children. Two recent studies of children’s memory for VCUG, the same medical procedure examined in the present research, have also revealed inconsistent findings. Goodman and col- leagues observed children undergo a VCUG and tested their memory between 1 and 3 weeks later (Goodman et al., 1994; Goodman, Quas, Batterman-Faunce, Riddlesberger, & Kuhn, 1997). Although preliminary analyses indicated that distress was positively related to children’s errors in response to direct questions about the procedure, when parents’ attachment styles were also considered, the negative relations between stress and memory were no longer significant (see below). Merritt et al. (1994) assessed children’s memory for VCUG immediately after the procedure and again 6 weeks later. Relations between stress and memory depended