Designing effective pedagogical tasks has been of critical interest among second‐language (L2) researchers and teachers. Accordingly, several claims about how to classify pedagogic task characteristics in terms of their effectiveness in L2 learning have been made. One such example is the involvement load hypothesis (Hulstijn & Laufer, 2001), which claims that learning new words during vocabulary‐focused tasks is dependent on the degree of cognitive processing required of an L2 learner by a given task. Building upon Hulstijn and Laufer's (2001) original research, which partially supported the hypothesis, the present study examined the hypothesis by exploring the interaction between task‐induced involvement and learners’ L2 proficiency on the initial learning and retention of target words. The study consisted of two experiments that were carried out with English‐as‐a‐second‐language learners at two different proficiency levels (i.e., matriculated undergraduate students vs. students in an Intensive English Program). Experiment 1 tested the hypothesis with three tasks imposing different levels of task‐induced involvement (n= 64), whereas Experiment 2 included two tasks hypothesized to represent the same task‐induced involvement (n= 20). In line with the predictions of the involvement load hypothesis, the findings of Experiment 1 indicated that a higher level of learner involvement during task performance promoted more effective initial learning and retention of target words. Additionally, Experiment 2 showed that different tasks with the same involvement load resulted in a similar amount of vocabulary learning. These results were generally consistent across different proficiency levels, suggesting no interaction effect between task‐induced involvement and L2 proficiency on vocabulary learning. The findings of the present study further the involvement load hypothesis by providing insights into how some individual differences—in this case L2 proficiency and cognitive involvement—might be more/less important to consider when implementing pedagogic tasks. Interestingly, the results indicated that as long as L2 learners’ cognitive and language abilities allow them to complete vocabulary tasks in a given time, a deeper level of processing of the new words, especially the evaluation component of task‐induced involvement during tasks, facilitates L2 vocabulary learning. Furthermore, the findings of Experiment 2 also provide insightful pedagogical implications demonstrating that it is possible to design different types of vocabulary tasks inducing similar amounts of involvement loads and that they can be equally beneficial for vocabulary learning. In task‐based language teaching, designing tasks involves a complex series of considerations such as understanding who the target learners are and what cognitive processes each task requires. The experiments presented in this article reexamined the involvement load hypothesis and utilized tasks designed with such considerations in mind in order to explore the effectiveness of various vocabulary learning tasks.