Studied the effects of multiple representations of information and verbal and spatial ability on foreign-language text comprehension. An interactive multimedia presentation of a German short story was read by 152 English-speaking college students. The presentation included either no annotations for key words, verbal annotations, visual annotations, or both (variation of cognitive load). Students' spatial ability was assessed with the Card Rotation Test from the Kit of Factor-Referenced Cognitive Tests (KF-RCT), and their verbal ability was measured with the Vocabulary Test from the Nelson-Denny Reading Test (NDRT). Recall of word translations was assessed with a pre- and post-test of 35 German key words, and students typed a text summary for the assessment of text comprehension. When visual annotations were provided, students with low verbal and spatial abilities recalled fewer word translations than students with high abilities. The presentation of verbal annotations, however, did not influence vocabulary recall. Furthermore, visual annotations reduced text comprehension for all learners. The results are interpreted in terms of cognitive-load theory and the generative theory of multimedia learning, in which learning processes are assumed to be impaired by the increased cognitive load exerted by multiple representations of information.