This mixed-methods study investigates writers' task representation and the factors affecting it in test-like and non-test-like conditions. Five advanced-level L2 writers wrote two argumentative essays each, one in test-like conditions and the other in non-test-like conditions where the participants were allowed to use all the time and online materials they needed. The writing was done on computers, and we recorded the writing process and keystrokes using the "Screen Capture Video" and "Inputlog" programs. We audio recorded stimulated recall interviews after each writing session, with the writers reporting and commenting on their writing strategies and their reasons for following them. The findings of this study suggest that there are several factors that play a role in task representation, such as previous education, personal beliefs, and task conditions. Although these factors were present in all participants' responses, the differences in the writers' approaches to interpret and execute the writing were marked. The results highlight various pedagogical issues and options related to teaching writing in general and to the place of task representation on writing programs in particular.