An eye tracking experiment into the division of attention between textual information and causal diagrams in academic texts

Which contributions do diagrams, causal relations and prior knowledge make to scientific texts used for students in academic environments? To answer this question, we set up an experiment with 24 students. They took part in an eyetracking experiment in which they read a text that matched their prior knowledge or did not. Causal diagrams, which summarized the relations in the text by means of bullets and arrows, were added. Text and causal diagrams were presented simultaneously on a computer screen. The number of causal relations was also varied (many-few). Results showed that the text containing more causal relations was better retained by participants than the text containing fewer causal relations F(1, 20)=39.31, p<.0001, irrespective of prior knowledge. However, a main effect of prior knowledge could also be observed F(1, 20)=12.60, p<.01, texts that matched participants’ prior knowledge were better retained. We also checked the eyetracking patterns of participants. For the text containing many causal relations, no correlations could be found between overall scores and division of attention between textual information and diagrams. However, for the text containing fewer causal relations, the more students integrated between text and diagram, the better their memory of the text (correlation 0.63). See Mason et al. (2013) for similar results. Our findings hints towards the need to elucidate course material lacking in causality, with diagrams in order to boost memory.status: publishe