The degree to which good and poor high school readers comprehend prose was studied in relation to their ability to (1) use passage organization to facilitate comprehension; (2) use task instructions to facilitate comprehension; and (3) answer verbatim and paraphrase items used to measure comprehension. Eleventh grade students were given the Stanford Diagnostic Reading Test, so as to define two groups (good and poor) of comprehenders. The study also employed two versions of a passage and three versions of instructions, followed by a multiple choice test. Results indicated that although students in the good group were not affected by passage structure, students in the poor group were aided by it. Task instructions did not significantly influence either group of readers. However, the poor group did react to both attribute and relation instructions while the good group reacted to only the relation instructions. To give a more detailed picture of what type of information processing occurred during reading, the results are discussed in terms of the verbatim and paraphrase items used to assess comprehension. (Author/GK)