Since the early 1980s there has been much research investigating children's susceptibility to memory distortions when interviewed with a range of techniques. Early studies using the ‘Draw and Tell’ interview have shown that drawing, when used as a recall technique, increases the amount of correct information reported during a recall interview, without decreasing accuracy (Butler et al., 1995). We examined whether drawing, as a rehearsal technique, would lead 5- and 6-year-old children to claim that implausible events that they had previously said didn't happen, had actually happened. Children were asked whether a list of events on a Life Events Inventory (LEI) had occurred. One week later, half the children completed a ‘Draw and Tell’ interview for some target events but not others. All children completed the LEI again. Children who drew were more likely to claim that all events, not just those they drew, really happened. We discuss implications for the use of the ‘Draw and Tell’ procedure in clinical and therapeutic contexts. Copyright © 2003 John Wiley & Sons, Ltd.