Children's suggestibility is typically measured using a three-stage "event-misinformation-test" procedure. We examined whether suggestibility is influenced by the time delays imposed between these stages, and in particular whether the temporal discriminability of sources (event and misinformation) predicts performance. In a novel approach, the degree of source discriminability was calculated as the relative magnitude of two intervals (the ratio of event-misinformation and misinformation-test intervals), based on an adaptation of existing "ratio-rule" accounts of memory. Five-year-olds (n = 150) watched an event, and were exposed to misinformation, before memory for source was tested. The absolute event-test delay (12 versus 24 days) and the "ratio" of event-misinformation/misinformation-test intervals (11:1, 3:1, 1:1, 1:3 and 1:11) were manipulated across participants. The temporal discriminability of sources, measured by the ratio, was indeed a strong predictor of suggestibility. Most importantly, if the ratio was constant (e.g. 18/6 versus 9/3 days), performance was remarkably similar despite variations in absolute delay (e.g. 24 versus 12 days). This intriguing finding not only extends the ratio-rule of distinctiveness to misinformation paradigms, but also serves to illustrate a new empirical means of differentiating between explanations of suggestibility based on interference between sources and disintegration of source information over time.