Modeling the effect of temperature history on the lag phase duration of Listeria monocytogenes

Estimation of lag phase duration (LPD) remains a challenging component in predicting microbial growth, food safety, and shelf-life. Most published predictive growth models do not account for prior information about the pre-inoculation environment, resulting in inaccurate estimates of LPD. The LPD does not only depend on current growth conditions but also on the previous growth environment. Therefore, the objective of this study was to develop a dynamic predictive model that quantifies the effect of temperature history on the physiological amount of work ( h 0 ) to be done in response to environmental change. Growth data for five Listeria monocytogenes strains were generated in tryptic soy broth following transitions between pre-incubation (8–40 °C) and post-incubation temperatures (8–30 °C). For all pre-post temperature combinations, Baranyi & Robert’s model was fitted to the growth data to quantify physiological parameter ( h 0 ) and LPD ( λ ). Further, a secondary model was developed showing the dependence of h 0 on previous and current environmental temperatures. Temperature downshifts resulted in higher h 0 values compared with equivalent upshifts, indicating greater adaptive workload during downshifts, while upshifts produced gradual increases in h 0 . The developed model was validated under dynamic temperature conditions, including downshifts and upshifts. A graphical comparison of predicted and observed bacterial population growth was used to evaluate model performance, with accuracy and bias factors ranging from 1.25 to 1.32 for the time to a 1-log increase, and ≥70% of prediction errors fell within acceptable prediction-zone limits. These findings demonstrate that incorporating temperature history enables prediction of intermediate lag phases under fluctuating temperature conditions, thereby reducing the chances of overprediction of bacterial growth in food systems.