In response to the growing prevalence of improvised explosive devices, a study was conducted that called into question the current methodology used to train explosive detection dogs. The study analyzed the effects of training these dogs using concept formation training versus standard pure training. It was hypothesized that dogs who received concept formation training would have a higher accuracy at generalizing to mixtures. The logic behind this hypothesis was that if a dog was trained elementally with pure samples, they would not be able to accurately recognize a diluted or modified version of the target Ammonium Nitrate scent. Group Elemental dogs had slightly higher accuracies during the concept testing portion of the study compared to the Group Concept dogs. Additionally, during the generalization phase of the experiment, the Group Elemental dogs performed equally well to the Group Concept dogs. Overall, concept formation training and elemental training were suggested to be equally effective according to the results of the dogs in this study. For the future, this experiment should be repeated with a larger sample size as it has great potential for giving insight on improving the training of explosive detection dogs.