This study is an extension of recent research, which examined the possibility that false memories in the Deese-Roediger-McDermott (DRM) paradigm predict the occurrence of false memories in misinformation paradigm. The purpose was to determine in which extent an imaging instruction reduces false memories in DRM and Misinformation paradigms. A sample of young adults was assigned to the DRM or the misinformation tasks, either in control conditions or in conditions including an imaging instruction. Findings confirm that an imaging instruction decreases false memories in DRM whereas it is not possible to conclude about such a reduction in the misinformation task. Overall, this pilot study suggests that the nature of the stimuli in each paradigm gives rise to quality differences in encoding processes, which in turn have consequences on the monitoring process at retrieval, leading to a weaker misinformation effect than DRM false recognition. In conclusion, while one has argued that the monitoringprocess is common to both paradigms, false memories in the DRM paradigm would be based on semantic association of words that is, on activation processes in semantic memory, whereas misinformation would rather rely on recollection process in episodic memory. Nevertheless, this hypothesis should be specifically tested in further experiments.