Eighty-two undergraduate and graduate students at a major southwestern university participated in a study integrating reading research utilizing tachistoscopes with current knowledge based on analysis of reading connected texts. A tachistoscope was used to examine the nature of visual perception in relation to the ability of the subjects to apply their knowledge of written language when perceiving and reporting test stimuli. Ten lines of print were presented to the subjects for one-tenth of a second each, with time between each presentation for the subjects to write down what they perceived. The lines of print varied from a complete and meaningful English sentence, predicted to be the most accurately reported, to a line of meaningless, irregular geometric shapes. Predictions of accuracy were based on the Goodman Model of Reading. The results were analyzed along two dimensions: (1) the degree to which the reports were complete and accurate, and (2) the degree to which subjects applied knowledge of written English in perceiving and reporting each line of print. The findings indicated that the more like language or more central to language a particular line was, the more accurately and completely it was reported. The findings strongly indicate that the more individuals were able to apply their knowledge of language when perceiving written material, the more rapidly and accurately they were able to perceive the material. (Tables and a bibliography are included.) (Author/JC)