Tasks are as useful devices for identifying learners’ needs and thus for designing specific purpose courses. Hence, the present study aimed at examining the viable effects of three varieties of task manipulation, i.e., oral reproduction, role-play, and group discussion, on the accuracy and fluency components of Iranian EFL learners’ oral speech. For the sake of homogeneity of the participants, 60 L1 Persian EFL students were selected through Oxford Placement Test (OPT) and divided into three 20-participant comparison groups: Oral Reproduction Group (ORG), Role-play Group (RPG), and Group Discussion Group (GDG). Each of the three comparison groups was treated just through a single task. ORG, RPG, and GDG were treated by oral reproduction, role-play, and group discussion tasks, respectively. The participants were pretested and posttested by the speaking part of Preliminary English Test (PET). The three groups were exposed to ten sessions of treatment, each of which lasted one hour and thirty minutes. The results of the study, obtained through ANOVA and Post Hoc analyses, revealed that the three types of task manipulation had significant influence on the fluency and accuracy of the participants’ speech. It was also revealed that GDG, treated through group discussion type of task manipulation, outperformed the other groups in terms of accuracy and fluency achievement. Tasks are hence recommended to be used in EFL teaching and testing contexts in that they are assumed to be effective tools to enhance the accuracy and fluency of English learners’ oral speech.
