Cassel, William Steven

Subjects in kindergarten, grade 2, and college were shown a videotape of a bicycle theft followed by a one month longitudinal study simulating the witness's experience during the pretrial phase of a criminal prosecutorial workup. Subjects were asked for free recall as well as for responses to nonleading and correct or incorrectly leading questions. Accuracy of free recall was high for all ages. Younger children were more susceptible to suggestive leading questions and adults were found to reject many of the "correctly" leading questions. Differences were found between answers to central and peripheral items and subjects were found to change their answers within the same interview in relation to age of the witness and centrality of the question.

Kindergartners, grade 1, 2, 3, 4, 5 children, and adult college students participated in a two-year longitudinal study of eyewitness memory, recall, and suggestibility with one- and two-year assessment periods additionally, kindergartners and adults participated in a short-term repeated trials free recall study. Delays varied from a few minutes (in the short-term repeated free recall trial) to two years in the longitudinal study. Subjects were asked free-recall (open ended), unbiased cued recall, and specifically suggestive (leading) questions about the event they had observed as well as their involvement in the experiment at the initial phase one or two years before. Longer delays resulted in lower levels of correct free recall which varied positively with age. Levels of incorrect free recall were at floor for all trials. With longer delays, unbiased cued-recall questions resulted in higher levels of incorrect responses. Differential levels of correct free recall were found for different categories of items, with the central (most legally important) items being recalled correctly by all age groups more often than the less central items. Kindergarten and second grade children were the most influenced by leading questions and were found to be most influenced by the type question (correctly or incorrectly suggestive) posed first. Responses by kindergartners to later questions about the same items were influenced by the leading nature of the initial question with kindergarten and second grade children responding incorrectly to the follow-up questions based upon their answers to the initial questions. The order of presentation effect was enhanced for questions suggesting incorrect responses. The longer delays diminished the age effects of leading questions between the older children and adults. Examination of short-term delay repeated recall trials disclosed that adults produced higher percentages of correct free recall with each subsequent trial, while the kindergarten children did not. Collapsed across grade, subjects who increased their levels of correct free recall following their unbiased cued-recall trial generally remained at that higher level for the remaining free-recall trials. Small hypermnesia effects were found for certain treatment groups. Results are discussed in terms of information processing and fuzzy-trace theories.