Most traditional predictive models for growth of foodborne pathogenic (or spoilage) organisms focus on single species behaviour. This approach may lead to a significant discrepancy between model predictions and reality, since the (potential) influence of the background flora is neglected. In this paper, a specific type of multiple species interaction is considered, namely, a pathogenic organism in the presence of a microbial antagonist, where the latter species inhibits the pathogen's growth through lactic acid production. As an experimental case study, growth curves of Yersinia enterocolitica in mono- and coculture with Lactobacillus sakei were generated in duplicate in a modified brain heart infusion medium. A complete factorial design was applied to assess the impact of temperature (4 levels) and inoculum, ratio pathogen/antagonist (6 levels) on the Y. enterocolitica growth. Based on a set of formulated model requirements, a novel model was developed, in which the experimentally observed inhibition effect, i.e., an (early) induction of the stationary phase in the pathogen's growth curve, is explicitly related to the lactic acid production curve. Experimental data and model predictions show good agreement. (C) 2003 Elsevier Ltd. All rights reserved.