The primary aim of the present research was to explain the variability in findings across past studies regarding the effects postevent misinformation has on preschoolers' testimony and memory. It is argued that the appearance and disappearance of such effects is due to at least three limitations of past research. The first limitation concerns the failure to equate the degree to which original information is learned both across conditions within a study and across studies. The second concerns the failure to use analytical techniques that are sensitive to the different processes involved in retention (e.g, forgetting and reminiscence). The third limitation involves past failures to examine both the potential constructive and destructive effects that exposure to misinformation may have on testimony and memory. By addressing these limitations it was possible to determine whether exposure to postevent misinformation encourages preschoolers to report erroneous information, as well as whether, and how, misinformation affects memory for a witnessed event. It was also possible to examine the effects of providing consistent postevent information on preschoolers' testimony and memory. A recently developed model of long-term retention that eliminates the problems of differences in initial learning and analytical insensitivity is used to examine the effects of consistent and inconsistent information on testimony and memory. Preschoolers were presented with a slide sequence about a little girl anxious to attend a Halloween party. Half of the children received a single trial and the remaining half learned the material to criterion. Following acquisition, children received one of the following: (a) no postevent information; (b) correct information concerning peripheral event details three weeks after acquisition, presented in either narrative or questionnaire form; or (c) misleading information concerning peripheral event details three weeks after acquisition, presented in either questionnaire or narrative form. Four weeks following acquisition, all of the children received 4 test trials without further study opportunity. The results indicate the following: (a) exposure to misleading information encouraged preschoolers to report misinformation; (b) although the effects of misleading information on memory were rare, there were more story details unavailable for recall in one of the misled than nonmisled conditions; (c) the transient effects of misinformation on memory and testimony are likely influenced by the limitations of past studies mentioned above; (d) re-exposing preschoolers to story details that were embedded in a narrative increased recall; and (e) performance increased across test trials with the recall of original information, but it did not differ as a function of experimental manipulation. These results demonstrate that when initial learning is controlled and appropriate measurement techniques are used, the potential misinformation may have to impair memory could play a role in preschoolers' reporting of this information.