The aim of the study was to compare the comprehension of a text which was presented to a reader with the key-word set matched to this text and processing of the same document with pictures illustrating the text content. 34 learners of Russian as a foreign language balanced by age, gender, and education level participated in the experiment. Each participant read three different texts, and had to retell them and answer 9 questions after each text: 6 questions concerning the local content of the text, 2 questions concerning the general analysis and comprehension of the text, 1 question for estimating whether the text was difficult or easy to read and comprehend. In the preliminary experiment, we revealed a “true” key-word set for each stimulus (https://advego.com/text/seo/). According to [Murzin, Shtern, 1991], a key-word set is a minimum content of a text. The comparison of the "true" key-word set with an individual set can reveal the level of text comprehension. Our findings demonstrated significant differences in the processing of the texts matched with the different types of support. Presentation of the key-word set helps to make the retellings longer, while a set of pictures matched to the verbal component leads to the subjective difficulty of the text. There was no significant difference in answering after-the-text questions. The results support the assumption of the Cognitive Theory of Multimedia Learning (Mayer, 2009) that integration of verbal and pictural information with each other (a polycode text) helps the learners to understand and memorize the text.