Two model antilisterial microbial communities consisting of two yeasts, two Gram positive and two Gram negative bacteria, and originating from Livarot cheese smear were previously designed. They were used in the present study to analyse the impact of microbial population dynamics on growth of Listeria monocytogenes in cheese microcosm. Specific culture media and PCR primers were developed for simultaneous culture-dependent and real-time PCR quantification of strains belonging to Marinomonas sp., Paenibacillus sp., Staphylococcus equorum, Arthrobacter arilaitensis, Pseudomonas putida, Serratia liquefaciens, Candida natalensis, and Geotrichum candidum, in cheese microcosms. All strains were enumerated after 3, 5, 8 and 14 days at 15 C. They established well at high counts in all cheese microcosms. Growth dynamics for all strains in presence of L. monocytogenes WSLC 1685 were compared to those of microbial communities obtained by omitting in turn one of the six members of the initial community. The growth of the microbial strains was neither markedly disturbed by Listeria presence nor by the removal of each strain in turn. Furthermore, these communities had a significant reducing effect on growth of L. monocytogenes independently of pH, as confirmed by mathematical modelling. A barrier effect was observed, that could be explained by specific competition for nutrients.