Participants tend to falsely remember a nonpresented critical word after having studied a list of the word's primary associates. This paper presents a Swedish version of the Deese-Roediger-McDermott paradigm, which provides a tractable method of experimentally inducing and investigating such illusory memories. In Experiment 1, using 60 19-32 yr olds, it was demonstrated that the constructed stimulus material induced highly reliable false-recall and false-recognition effects, and, moreover, that veridical and false memories were associated with a similar phenomenological experience of remembering. The results from Experiment 2, using 40 19-32 yr olds, indicate that the susceptibility to false recognition can be substantially reduced when participants are explicitly required to monitor the sources of their memories. These findings are consistent with predictions derived from the source-monitoring framework.