Gérard Arnold (Pôle comportement) a écrit un article publié dans Current Opinion in Insect Science qui questionne la méthode de reviewing de la littérature scientifique pour écrire la première version du rapport IPBES sur les pollinisateurs. Cette première version a été partiellement corrigée suite à la relecture par des experts extérieurs.
Résumé: To understand the real causes of disorders and abnormal mortalities of bees (honeybees and wild bees) in the world, the scientific method requires that each category of potential stressors be analyzed in a balanced manner. In this article, I show that the first version of the section dealing with the evaluation of the toxic effects of pesticides on bees, in the IPBES assessment report on pollinators and pollination, revealed an incomplete and biased literature review in many places, especially downplaying the risks that pesticides in general, and neonicotinoids in particular, pose for pollinating insects. Then, according to the rules of IPBES, an independent peer review by external experts of this first version allowed the published report to be more in line with the reality of scientific knowledge, which shows, for example, that sublethal effects of pesticide exposure can impair the ability of bees to provide pollination. However, some other key points remain unchanged in the published version.