Prediction-corrected Visual Predictive Checks (pcVPC) are a diagnostic tool for evaluating mixed-effects models, particularly in population pharmacokinetics (PK) and pharmacodynamics (PD). Traditional Visual Predictive Checks (VPC) compare observed and simulated data percentiles, but they can be misleading when data exhibit large variability in dose or covariates, or when applied to data from adaptive designs. pcVPC addresses these issues by normalizing observed and simulated data based on the typical population prediction for the median independent variable in each bin. This correction removes variability due to independent variables, allowing for more accurate diagnosis of model misspecification, especially in random effects models. The pcVPC retains the visual interpretation of traditional VPCs while improving diagnostic accuracy. It is particularly useful in cases with adaptive dose adjustments, where traditional VPCs may fail to detect model misspecifications. Additionally, pcVPCs can be applied to independent variables other than time, enhancing their diagnostic utility. The method was validated using both simulated and real data examples, demonstrating its effectiveness in identifying model issues. The pvcVPC further extends pcVPC by also correcting for variability within bins, improving diagnostic power. The study highlights the importance of using pcVPCs in mixed-effects modeling, especially when dealing with complex models and adaptive designs. pcVPCs provide a more reliable and informative diagnostic tool compared to traditional VPCs, facilitating the development of more accurate and predictive models. The results suggest that pcVPCs are a valuable addition to the diagnostic toolkit for PK/PD modeling.Prediction-corrected Visual Predictive Checks (pcVPC) are a diagnostic tool for evaluating mixed-effects models, particularly in population pharmacokinetics (PK) and pharmacodynamics (PD). Traditional Visual Predictive Checks (VPC) compare observed and simulated data percentiles, but they can be misleading when data exhibit large variability in dose or covariates, or when applied to data from adaptive designs. pcVPC addresses these issues by normalizing observed and simulated data based on the typical population prediction for the median independent variable in each bin. This correction removes variability due to independent variables, allowing for more accurate diagnosis of model misspecification, especially in random effects models. The pcVPC retains the visual interpretation of traditional VPCs while improving diagnostic accuracy. It is particularly useful in cases with adaptive dose adjustments, where traditional VPCs may fail to detect model misspecifications. Additionally, pcVPCs can be applied to independent variables other than time, enhancing their diagnostic utility. The method was validated using both simulated and real data examples, demonstrating its effectiveness in identifying model issues. The pvcVPC further extends pcVPC by also correcting for variability within bins, improving diagnostic power. The study highlights the importance of using pcVPCs in mixed-effects modeling, especially when dealing with complex models and adaptive designs. pcVPCs provide a more reliable and informative diagnostic tool compared to traditional VPCs, facilitating the development of more accurate and predictive models. The results suggest that pcVPCs are a valuable addition to the diagnostic toolkit for PK/PD modeling.