Based on the Involvement Load Hypothesis proposed by Laufer& Hulstijn (2001),the present experimental study aims to examine whether acquisition and retention of vocabulary acquired incidentally by non English basic class students is contingent upon the amount of task-induced involvement load.The results only partially support the Involvement Load Hypothesis: tasks with high involvement load are not inexorably more conducive to enhancing word learning than tasks with low involvement load.The output tasks are superior to input tasks in fostering the knowledge and retention of vocabulary.The fill-in task with the higher amount of involvement load did not produce lasting lexical learning effect.Due to limited time of the English class,some pedagogical implications and measures are presented in order to effectively improve the incidental vocabulary learning of basic learners.