Jigsaw is a peer learning procedure based on the assumption that making "children treat each other as resources" (Aronson & Patnoe, 2011, p. 8) stimulates cooperation among equals. Using a short-term, longitudinal experimental design in 14 sections of an undergraduate human anatomy laboratory, we contrasted this perspective with the idea that Jigsaw's two-group composition actually elicits a mix of opposing social-psychological processes and outcomes. Supporting this view, students' perceptions of social interdependence and sociocognitive conflict regulation covaried over time with each other and with motivation and achievement. Likewise, rather than solely elicit cooperation, Jigsaw students initially reported higher levels of competition and individualistic efforts than students in a business-as-usual (BAU) control, and lower levels of epistemic regulation. These trends then reversed over time, but the magnitude of increasing cooperation and decreasing competition and individualistic efforts among Jigsaw students never exceeded that of BAU students. In fact, at the end of the semester, the only significant differences between Jigsaw and BAU were for relational regulation and academic achievement. Overall, these findings provide a much more complete but complicated portrait of Jigsaw social psychology and its effects over time, and so doing indicate that theory development is needed to account for mixed-motive situations that simultaneously elicit cooperative, competitive, and individualistic goals and different forms of sociocognitive conflict regulation. For practice, findings also suggest that simply distributing resources among jigsaw group members does not result in optimal outcomes, and consequently Jigsaw must change in ways that strengthen positive interdependence and decrease social comparison.