ABSTRACTWe present the results of a study that compared two placements of the Retrospective Think-aloud (RTA): A Post-session RTA where the think-aloud occurs after all tasks are complete, and a Post-task RTA where the think-aloud is elicited after each task. Data from task performance and verbal measures were collected from 24 participants. The results suggest that in terms of task performance, participants in the Post-session RTA condition performed tasks faster, with fewer errors and fewer clicks than in the Post-task RTA condition. In terms of utterances, participants in the Post-task RTA condition produced significantly more utterances that explained actions, expectations and procedural descriptions than in the Post-session RTA condition.