The extant picture-learning research does not address confusing word pairs that are not concrete (e.g., in and into). In this study, university students viewed 11 timed Web pages containing information on confusing word pairs. Each page addressed one word pair and distinguished the words with examples (example group), examples and rules (rule group), examples and pictures (picture group), or examples, rules, and pictures (picture/rule group). On a concept test, all groups outperformed the example group, and the picture group outperformed the rule group. All groups outperformed the example group on a skill test. These results were consistent with the picture hypothesis and extend the previous research to include the use of pictures with paired words that are not concrete.