The purpose of this study was to examine which of the two vocabulary-learning methods non-contextualized vocabulary learning and vocabulary learning combined with semantic relations promote the learning and retention of new vocabulary items. The following research questions were asked: Which of the two above-mentioned methods best promotes the students’ vocabulary learning and word comprehension? What differences can be noted in the students’ test results regarding learning and retention of new vocabulary items between these two methods? Which method promotes the students' long-term retention of vocabulary items and word comprehension? In total, 36 students from two classes, divided into four groups, participated in the study. Each group was taught using one of the two methods. After the first round of lessons, the teaching methods were swapped such that all the students tried both methods and so that the survey could be considered reliable. The empirical material was obtained from field experiments and tests in direct connection with the lessons and retention tests four days later. The results were analyzed using Schema theory and a theory of memory processing called Executive Control Processes. These theories were used to investigate the students’ retention of vocabulary items and how students use schematic knowledge to associate words to one another. The results showed that the second method, the combined method, seemed to be the most promising not only for retention, but also for the number of learned vocabulary items and for word comprehension.