This paper reports the results of a study aiming to investigate the effects of interactionist and interventionist dynamic assessment (DA) on learners’ writing performance. To do so, 63 upper-intermediate EFL learners were selected as the study participants. A writing pretest was administered to ensure the homogeneity of the participants and determine their level of proficiency prior to treatment. In the online interactionist DA group, the researchers implemented the treatment based on the ‘Dynamic Mediation Process’ proposed by Elliott (2002) and Xiaoxiao and Yan (2010). The framework comprised topic-choice, idea-generation, structuring and macro-revising stages. In doing so, the pre-task phase was conducted, followed by providing the mediation and performing post-task. In the interventionist DA group, Lantolf and Poehner’s (2011) framework was followed. Based on this framework, the learners’ erroneous language chunks were treated through the teacher’s underlining the error, putting a question mark, circling it, or posing a written question concerning the problematic part. Alternatively, the teacher would point out the incorrect part and offer two options for the learners to choose from. At the end, a writing posttest was given to all groups to evaluate their writing performance after treatment. As the study findings helped reveal, both interactionist and interventionist DA types improved learners’ writing performance. In addition, the comparison of the posttest scores indicated that the interactionist DA group participants even outperformed the interventionist ones on the writing posttest. The implications of the findings are discussed throughout the paper.