Blog | CRiSS-LAB

Why retracted research keeps circulating

Wed, 25 Feb 2026 00:00:00 +0000

Science has formal mechanisms for correction, but correction is not the same as forgetting. Retractions can invalidate a result, yet the paper, its claims, and its downstream influence may continue circulating through citations, reviews, datasets, and public debate.

This is the motivation behind the FONDECYT Regular project Collective Memory Decay in Science: Patterns and Determinants of Forgetting Retracted Research. The project studies how scientific communities remember and forget invalidated research by combining bibliometrics, network science, natural language processing, and models of collective memory.

The core question is practical: when correction does not change the memory of the system, misinformation can accumulate inside the scientific record. Understanding that process is a necessary step toward better science communication, editorial policy, and evidence governance.

What network science adds to education

Thu, 11 Dec 2025 00:00:00 +0000

Education is relational. Students learn with peers, teachers coordinate interventions, programs compete and complement one another, and institutions shape trajectories through rules, admissions, and information.

Network science gives us a way to make those relationships explicit. It helps distinguish isolated students from well-integrated ones, redundant collaboration from diverse information access, and fragile course pathways from robust educational ecosystems.

For CRiSS-LAB, this is not only a modeling preference. It is a practical stance: better relational evidence can support better pedagogical decisions, better student support, and better institutional design.

From classroom games to evidence for school coexistence

Mon, 01 Sep 2025 00:00:00 +0000

Many school coexistence problems are relational before they become visible incidents. Isolation, asymmetry, low reciprocity, and fragmented groups can stay hidden when schools rely only on direct reports or retrospective surveys.

Capybara translates a brief student interaction into relational evidence. By combining experimental game theory, network science, and AI-assisted reporting, the platform helps schools identify early signals, prioritize prevention, and support professional judgment with clearer data.

The goal is not to replace school teams. The goal is to give them a structured view of the classroom so they can act earlier, with better evidence and less dependence on anecdotal perception.