Abstract/Summary

Inference generation and comprehension monitoring are essential elements of successful reading comprehension. We know that these skills improve with age and that children engage in both. However, we know much less about when and how children make inferences and detect inconsistencies as they read. Over two experiments, we monitored the eye movements of two groups of children (age 8-13 years) as they read short passages and answered questions that tapped local (experiment 1) and global (Experiment 2) inferences. Questions were presented before and after the texts in order to examine the effect of question location on inference generation. Results showed that children make local but not global inferences online and that inconsistencies are detected only when an inference has been made. In addition, also reading the question before the passage had a large effect on reading times in the passage and of the question, this did not affect global comprehension. We therefore conclude that presenting questions before reading does not benefit overall comprehension, and that children, like adults, prioritise efficiency over completeness when reading, generating inferences spontaneously only when necessary for establishing a coherent mental model.