The literature implies a trend towards an increased use of interactive multimedia technologies for instruction. The increased availability of moderate cost, good quality, digital computer cards and computers with built-in capability have enabled trainers and educators to realize the potential of random access for computer-based training (CBT) and other multimedia applications. In this study, the researcher investigated the effectiveness of adding digital to CBT. The experiment was conducted at the University of Central Florida in the Spring of 1991 with students (N = 60) enrolled in the Applications of Technology in Education undergraduate course. After randomly selecting 3 out of 10 intact course sections, students were randomly assigned to one of three treatment groups. Each treatment group completed a CBT program with identical content and one of the following delivery modes: (1) text-only delivery, (2) full text and totally redundant delivery, or (3) partial text (bulleted) with full delivery. Students in all treatment modes were administered a 30-item criterion-based achievement pretest and posttest in multiple-choice format on the subject content of the CBT program. In addition, subjects were administered a modality strength test, and a post-experiment perception questionnaire was completed by all subjects to provide descriptive data. Additional tracked variables included time required to complete the CBT program, number of times a student reviewed (backed up), and number of times a student in an treatment selected the repeat audio option. Data were analyzed using descriptive statistics, one-way analyses of variance, chi-square, t-tests, multivariate analyses of variance, and the Least Significant Difference multiple comparison procedure. Major findings of this study were: (1) There were no significant differences among the three treatment groups in achievement gain. (2) There was a significant difference in the mean completion times across the three treatments, with the text-only version requiring the least time on task. (3) There was no significant relationship between student modality strengths and achievement levels. (4) Student perceptions indicated a high degree of program acceptance across all levels of treatment.