This study investigates the predictive ability of the Implicit Association Test (IAT) and compares it with Linear Mixed-Effects Models (LMMs) that provide Rasch-like parametrization of accuracy and time responses. The IAT is a widely used measure for assessing implicit attitudes, but its typical scoring method, the D score, has been found to have poor predictive ability for behavioral outcomes. The study suggests that LMMs can address this issue by accounting for across-trial variability in IAT data and providing more accurate predictions. The IAT measures the strength of associations between targets and evaluative dimensions by considering the speed and accuracy with which participants assign prototypical exemplars of two targets to their own category in two contrasting conditions. The D score, which is derived from the difference in performance between the two conditions, is often used to express the IAT effect. However, the D score is sensitive to across-trial variability and may provide biased estimates of the construct. The study compares the predictive abilities of D scores and LMM estimates using data from a Chocolate IAT. The results show that LMM estimates have better predictive ability than the D score and allow for in-depth analyses at the stimulus level that help reduce across-trial variability. The study also finds that the IAT effect is mostly driven by Good and Milk exemplars, and that the performance in each associative condition is mostly influenced by the associations between one of the two chocolates and positive attributes. The study introduces a modeling framework that can improve the predictive ability of the IAT while providing information on respondent performance and stimulus functioning. This information can be used to reduce across-trial variability due to stimulus heterogeneity, leading to better functioning, more informative, and potentially briefer IATs. The study also finds that the D scores computed on the most informative data set explain the highest proportion of variance and provide better predictions than the D scores computed on the full-length data set. The study highlights the importance of stimulus representativeness and the sensitivity of the D score to across-trial variability. It also suggests that the proposed modeling framework can be used to identify malfunctioning stimuli and those that mostly contribute to the IAT effect, especially when target categories are more heterogeneous. The study concludes that the IAT effect is mostly driven by a milk chocolate preference, but the performance in each condition is mostly influenced by the associations of positive attributes with one of the two chocolates. The results have implications for both marketing and applied social psychology, as they can help in designing more effective marketing campaigns and understanding the role of in-group preferences in implicit attitudes.This study investigates the predictive ability of the Implicit Association Test (IAT) and compares it with Linear Mixed-Effects Models (LMMs) that provide Rasch-like parametrization of accuracy and time responses. The IAT is a widely used measure for assessing implicit attitudes, but its typical scoring method, the D score, has been found to have poor predictive ability for behavioral outcomes. The study suggests that LMMs can address this issue by accounting for across-trial variability in IAT data and providing more accurate predictions. The IAT measures the strength of associations between targets and evaluative dimensions by considering the speed and accuracy with which participants assign prototypical exemplars of two targets to their own category in two contrasting conditions. The D score, which is derived from the difference in performance between the two conditions, is often used to express the IAT effect. However, the D score is sensitive to across-trial variability and may provide biased estimates of the construct. The study compares the predictive abilities of D scores and LMM estimates using data from a Chocolate IAT. The results show that LMM estimates have better predictive ability than the D score and allow for in-depth analyses at the stimulus level that help reduce across-trial variability. The study also finds that the IAT effect is mostly driven by Good and Milk exemplars, and that the performance in each associative condition is mostly influenced by the associations between one of the two chocolates and positive attributes. The study introduces a modeling framework that can improve the predictive ability of the IAT while providing information on respondent performance and stimulus functioning. This information can be used to reduce across-trial variability due to stimulus heterogeneity, leading to better functioning, more informative, and potentially briefer IATs. The study also finds that the D scores computed on the most informative data set explain the highest proportion of variance and provide better predictions than the D scores computed on the full-length data set. The study highlights the importance of stimulus representativeness and the sensitivity of the D score to across-trial variability. It also suggests that the proposed modeling framework can be used to identify malfunctioning stimuli and those that mostly contribute to the IAT effect, especially when target categories are more heterogeneous. The study concludes that the IAT effect is mostly driven by a milk chocolate preference, but the performance in each condition is mostly influenced by the associations of positive attributes with one of the two chocolates. The results have implications for both marketing and applied social psychology, as they can help in designing more effective marketing campaigns and understanding the role of in-group preferences in implicit attitudes.