Studies on collaborative memory have revealed an interesting phenomenon called collaborative inhibition (CI) (i.e., nominal groups recall more information than collaborative groups). However, the results of studies on false memories in collaborative memory tasks are controversial. This study aimed to understand the production of false memories in collaborative memory tasks by applying the "turn-taking" method to the DRM paradigm. Experiment 1 aimed to analyse the production of false memories in collaborative memory tasks by controlling for the backward associative strength (BAS) of the presented word lists. In Experiment 2, we intended to define the limits of the collaborative inhibition effect when the nominal recall task followed the "turn-taking" method. The results of both experiments revealed that, in addition to the existence of the collaborative inhibition effect, collaborative recall produced significantly fewer false memories than nominal recall. However, collaborative inhibition was not affected by the "turn-taking" method of retrieval.