In this article, the authors report the results of two experiments that explored hypotheses about the relative learning advantages of role-play and scenario design. The experiments were conducted with similar student populations in Australia and Israel. Using a matched-pairs design, participants were randomly assigned to design and role-play conditions. They worked on their tasks following an hour-long lecture on three negotiation concepts: alternatives, time pressure, and negotiating power. A lecture-only control group was implemented in the Australian experiment. In both experiments, designers, working "behind the scenes," indicated better concept learning in the short run than their role-play counterparts performing "onstage," as well as in comparison with the control group. They showed better understanding of the way the concepts are related and retained the learning gains over time. Moreover, the designers were at least as motivated as role-players and controls and, for the Israel participants, showed more motivation. The results, favoring designers, spread widely across the various questions, asked immediately after the experience and 1 week later: 86% of the answers given favored designers in terms of direction; 52% of these were statistically significant. Implications are discussed for explanatory mechanisms, programmatic research, and teaching/training approaches.