🤖 AI Summary
This study investigates the key mechanisms underlying scientific team success, focusing on the interplay between team persistence (long-term collaboration stability) and freshness (formation of novel collaborations atop existing expertise). Method: Leveraging temporal coauthorship data from 25.2 million papers and 31.8 million authors, we construct a dynamic scholarly collaboration network and integrate network science with large-scale bibliometric analysis. Contribution/Results: High-impact outputs predominantly emerge early in team lifecycles; actively incorporating fresh collaborators—rather than relying solely on persistent ties—significantly enhances research quality; teams achieve sustained success through structural reconfiguration and strategic infusion of freshness. Critically, this work challenges the conventional assumption that persistent collaboration inherently yields superior performance, providing the first empirical evidence that collaboration freshness constitutes a core network mechanism driving breakthrough scientific outcomes.
📝 Abstract
Team science dominates scientific knowledge production, but what makes academic teams successful? Using temporal data on 25.2 million publications and 31.8 million authors, we propose a novel network-driven approach to identify and study the success of persistent teams. Challenging the idea that persistence alone drives success, we find that team freshness - new collaborations built on prior experience - is key to success. High impact research tends to emerge early in a team's lifespan. Analyzing complex team overlap, we find that teams open to new collaborative ties consistently produce better science. Specifically, team re-combinations that introduce new freshness impulses sustain success, while persistence impulses from experienced teams are linked to earlier impact. Together, freshness and persistence shape team success across collaboration stages.