Does self-generating a graphic organizer while reading improve students' learning?

Abstract To understand a document, learners must select the relevant information, organize this information into a coherent representation, and integrate it with their prior knowledge. One way of facilitating these cognitive processes is to display a graphic organizer alongside the document, showing the main items of information contained in the text and the links between them. To ensure students' active engagement in these processes, they can also be asked to self-generate an organizer (generative processing). However, this kind of task can be too demanding and overload their cognitive capacity (extraneous processing). We therefore compared the learning of students who were instructed to study an illustrated text either on its own or accompanied by a readymade graphic organizer (displayed statically or step-by-step). In another group, students had to self-generate the organizer while reading. As predicted, providing a graphic organizer improved students' recall compared with that of the control group. Contrary to the generative hypothesis and consistent with the cognitive load hypothesis, the self-generated organizer group 1) performed more poorly on the recall test than the readymade organizer groups, and 2) achieved lower transfer scores.