The present study analyzed the efficacy of a brief intervention aimed at scaffolding readers' "recursivity" (i.e., going back to the texts) while reading multiple texts and writing an argumentative essay. The participants were 151 university students, randomly assigned to two conditions: experimental (Recursivity-induced, RI) and active control (AC). We collected data about participants' thinking dispositions, perceived prior knowledge and perceived level of instruction in argumentative writing received, and prior beliefs. Then, students were assigned two texts about the evaluation of teachers, one pro and one against. RI students were prompted to compare the argumentation of each text with their own prior beliefs, whereas AC students were asked to write a summary of each text. Immediately after reading the texts and performing the accompanying tasks, RI and AC students were asked to write an argumentative essay to express their opinion on the topic. Process data were collected through the software Kidlogger. Results confirmed that the brief intervention improves students' analysis of the belief-inconsistent text, the overall argumentative quality of students' essay, and valid inferences made in a recall task one month after. The process analysis suggested that the intervention increases recursivity in at least a certain number of RI participants.