This study analyzes the effect of text-inserted questions and post-text-reading questions, i.e., question timing, on students' processing and learning when studying challenging texts. Seventy-six freshmen read two science texts and answered ten adjunct questions with the text available, being tested on learning 5 days afterwards. Questions were presented either after reading the whole text or inserted in the text after reading the relevant information. Online processing data were recorded while reading and searching the texts, and measures of processing strategies (i.e., paraphrases, elaborations) while answering the questions were collected. Compared to students in the post-reading condition, those in the inserted condition spent more time reading the text initially, were more efficient at searching for information in the text, and produced more accurate elaborations, all of which may explain why answering inserted questions in an available text were more efective in terms of learning than answering post-reading questions. Limitations and educational implications of these results are also discussed.