A study examined whether illustrations that must be attended to in order to make sense of a text hinder decoding by encouraging dependence on such illustrations for interpretation of text, both for decoding and for answering questions. Subjects--167 first graders--were asked to read four stories, two of which were taken from basal readers and two of which were children's trade books. One trade book and one basal story each had illustrations that were crucial to story comprehension, and one of each had nonessential, or redundant, illustrations. The stories were presented to each child in one of four versions: text only, illustrations only, text plus illustrations, and revised text. After the students read the story aloud, they were given an oral comprehension test and asked to read a list of five words taken from the story. Decoding errors and reading time were recorded. Results showed that first graders who read poorly answered comprehension questions better when the texts they read were illustrated, and about equally well regardless of whether the illustrations were redundant or crucial; good readers answered more questions correctly on texts that did not depend on crucial illustrations, regardless of whether or not the illustrations were actually present. (HOD)