Digital platforms enable online deliberation and collective problem-solving among a diversity of individuals distributed across locations. To facilitate the convergence of diverse perspectives into a solution, some form of centralization—such as mediators or a panel that screens participants or selects among proposed solutions—is often used. However, centralization can cause premature convergence to sub-optimal solutions and suppress participants’ motivation to engage. We develop a minimally invasive intervention that improves deliberation quality and allows participants to converge on solutions. Specifically, we introduce a Shared Attribute Space (SAS)—a set of attribute dimensions that allows participants to analyze and compare solution proposals. In two experiments using a custom-built online deliberation platform, we find that providing SAS enhances deliberation quality, improves the expected impact of proposals, and promotes their clustering within a convergent solution space. Our findings provide empirical evidence for improving collective problem-solving in online deliberation without exercising centralized authority.
