Current laboratory paradigms used to assess unconscious plagiarism consist of three tasks. First, participants generate solutions to a puzzle task with a partner (initial generation task); second, they recall their individual contribution (recall-own task); and third, they attempt to create new solutions that were not offered previously (generate-new task). An analysis of these tasks indicated that they differ in terms of the source monitoring they require. The two generative tasks require less differentiated information (e.g., familiarity) and relatively lax decision criteria. The recall-own task, however, demands more differentiated information and more extended decision criteria. In two experiments, factors known to influence source monitoring were manipulated. Consistent with the analysis, no effects were associated with the generative tasks. Recall-own plagiarisms increased when self-and other-generated solutions were difficult to distinguish (Experiment 1) and decreased when the two sources were easier to distinguish (Experiment 2).Offering an idea or a solution to a problem under the belief that it is one's novel contribution, when that idea or solution has actually been encountered earlier, constitutes cryptomnesia, or unconscious plagiarism (Taylor, 1965). Currently, two paradigms have been used successfully to study the laboratory analogue of unconscious plagiarism. The first, developed by Brown and Murphy (1989), assessed unconscious plagiarism in a category generation paradigm that consisted of three sequential tasks. The first task (called initial generation) required participants to take turns in groups of 4, generating exemplars from four categories (e.g., musical instruments, sports, etc.). Each participant offered four exemplars per category in a round-robin fashion under the admonition to avoid repeating another person's responses. In the two subsequent tasks. participants attempted to recall the exemplars that they had offered (recall-own task) and to generate new exemplars that had not been offered earlier by themselves or by their fellow group members (generate-new task). Because of the admonition instruction, plagiarisms were counted in the initial generation and generate-new tasks whenever a participant offered an exemplar that had already been given. In the recall-own task, plagiarisms occurred whenever a participant claimed to have generated an exemplar that had been offered by another group member. In all three tasks, the proportion of plagiarized responses was substantial and varied with manipulations such as task difficulty (e.g.,