Language learners' cognitive engagement with the content in language classes has been advocated in the last few decades (Laufer & Hulstjin, 2001). To this end, the researcher designed authentic problem-based tasks which make use of learners' cognitive and metacognitive skills to solve real-life vocabulary tasks. Nelson vocabulary test was administered to 64 Iranian EFL learners studying at a language institute in Tehran. By considering 1 standard deviation above and below the mean score, two cohorts of participants were selected for this study, i.e. the experimental group (n = 24) and the control group (n = 23). Conventional vocabulary learning tasks were implemented in the control group classes for 10 sessions while authentic problem based vocabulary learning tasks were implemented in experimental group classes. The results of data analysis revealed that the experimental group participants outperformed the control group learners in both tests of vocabulary recall and vocabulary retention (administered after a two week interval). Pedagogical Implications are discussed.