Research suggests screen reading is slower and possibly less accurate than reading from paper. Six study and test sessions over 10 months examined correct scores and retrieval responses for learning material presented via these two media. Correct scores did not differ suggesting that close matching of material can eliminate any decrement in reading speed or accuracy from screens. However, the way in which knowledge was retrieved varied between the presentational formats. These differences were time related and suggest that repeated exposure and rehearsal of computer-based information is necessary to equate knowledge application with that achievable from hard copy alternatives. It is suggested that this difference might be due to cognitive interference caused by cathode-ray tube (CRT) monitor characteristics of refresh rates, fluctuating luminance, and contrast levels.