The purpose of the study was to train participants on the nutritional values (carbohydrates) of various food items and evaluate the effect over three weeks. Twenty-seven participants were assigned to three different conditions. Initially, participants pre-sorted cards labeled with food items and categories, and any incorrectly sorted items were identified and customized for each participant. Tailored stimuli were then used for many-to-one conditional discrimination training. Conditions 1 and 2 aimed to establish three-member classes, while condition 3 aimed for five-member classes. Participants in all conditions sorted food items into categories of “more than 40,” “20–40,” and “less than 20” grams of carbohydrates. Additionally, participants in conditions 2 and 3 could sort items into a “don’t know” category. Maintenance of the learned categories was tested after one week. Participants who did not meet the mastery criterion received additional training and testing. A follow-up test was conducted two weeks after the maintenance test. During both the maintenance and follow-up phases, tests for emergent relations and sorting were conducted. The results showed that most participants responded correctly in both the emergent relations and sorting tests. The majority of participants did not require additional training or testing, and the equivalence classes were maintained across all conditions over the three-week period.