Children's online processing of written language inferences from eye movements

This thesis reports seven experiments that investigated children's online processing of written language. A variety of visual and linguistic factors, for which there are well-documented effects in adults, were manipulated in order to examine children's and adults' eye movements as they read sentences containing these manipulations. Experiment 1 investigated saccadic targeting of long and short words and showed that adults and children are generally alike in where they target their initial saccade to a word, and how they use parafoveal word length information to skip words. It also showed that while the length of a word directly influences the eye movement behaviour of both adults and children during text reading, the magnitude of these effects is greater in children. Experiments 2 and 3 showed that reliable word frequency effects are observed in adults and children when age-appropriate texts are used to index frequency counts. Experiment 3 also showed that word frequency effects occur in children, even when the age at which words were acquired is held constant. In Experiment 4 lexical processing of semantically ambiguous words was examined and the data were suggestive of there being a cost associated with processing words with more than one meaning for older children. However, the effects were not robust in adults or younger children. Experiments 5 and 6 investigated syntactic parsing and showed that children are slightly delayed relative to adults in their detection of initial syntactic misanalysis, but that they appear to have a similar sentence-parsing mechanism in place as adults. Finally, Experiment 7 investigated thematic processing of anomalous and implausible sentences, and showed that while there is no difference in the time course of thematic anomaly detection in adults and children; children are delayed in their detection of thematic implausibility as compared to adults. Overall, the data show that adults and children appear to have similar rnechanisms in place for processing written language visually, lexically, syntactically and thematically. They also show that the magnitude of disruption associated with these effects is greater in children than in adults, that the time course of children's syntactic processing is slightly delayed relative to that of adults, and further, that children are delayed in the efficiency with which they are able to integrate pragmatic and real world knowledge into the discourse representation. The thesis also makes a number of methodological points that have implications for conducting future research with developmental populations.