Integrating pollinators' movements into pollination models

Juliane Mailly ^1,2,3* Louise Riotte-Lambert ^2,4,5,6,7 Mathieu Lihoreau ^1,2,3

• ¹ UMR5169 Centre de Recherches sur la Cognition Animale (CRCA), Toulouse, France
• ² Centre National de la Recherche Scientifique (CNRS), Paris, France
• ³ Université Toulouse III Paul Sabatier, Toulouse, Occitanie, France
• ⁴ UMR5175 Centre d'Ecologie Fonctionnelle et Evolutive (CEFE), Montpellier, Languedoc-Roussillon, France
• ⁵ Université de Montpellier, Montpellier, Languedoc-Roussillon, France
• ⁶ École Pratique des Hautes Études, Université Paris Sciences et Lettres, Paris, France
• ⁷ Institut de Recherche Pour le Développement (IRD), Marseille, Provence-Alpes-Côte d'Azur, France

Accurate prediction of pollination processes is a key challenge for sustainable food production and the conservation of natural ecosystems. For many plants, pollen dispersal is mediated by the foraging movements of nectarivore animals. While most current models of pollination ecology assume random pollen movements, studies in animal behaviour show how pollinating insects, birds and bats rely on sensory cues, learning and memory to visit flowers, thereby producing complex movement patterns. Building upon a brief review of pollination and movement models, we argue that we need to better consider pollinators' cognition to improve predictions of animal-mediated pollination across all spatial scales, from individual flowers, to plants, habitat patches and landscapes. We propose a practical roadmap for the integration of behavioural models into pollination models and discuss how this synthesis can refine predictions regarding plant mating patterns and fitness. Such crosstalk between animal behaviour and plant ecology research will provide powerful mechanistic tools to predict and act on pollination services in the context of a looming crisis.