According to previous research, the variability of lists in eliciting false recall in the Deese-Roediger-McDermott (DRM) paradigm is large. A list structure made by Mukai, in which people's names were used as critical lures, was used to investigate the variability of lists. The materials were composed of 6 pairs of lists, in which the critical lure and its contextually associated study item (referred to as critical presented item) were assumed to play an important role in eliciting false recall. 80 participants ( M age = 21.7 yr.; SD = 2.6) were tested. The difference of the list pair in eliciting false recall was positively correlated with list pair differences in free association rate from critical presented item to critical lure ( r = .82, p <.05) and negatively correlated with the length of critical lure ( r = −.94, p < .01). It was shown that the variability of lists in eliciting false recall can be explained by these two factors. Moreover, the length of the lure was also negatively correlated with an index of unsuccessful source monitoring ( r = −.87, p < .05). The results were discussed in terms of the activation/monitoring theory.