Self-explaining, which asks students to generate explanations while reading a text, is a self-constructive activity and is helpful for students' learning. Studies have revealed that prompts by a human tutor promote students' self-explanations. However, most studies on self-explaining focus on spoken self-explanations. This study investigates the effects of two kinds of content-free computer supports for self-explaining: one is full-text reading and modifiable typing interface and the other is prompting. The results showed that self-explaining in a full-text reading and modifiable typing interface, particularly when prompted, made students perform better in applying target procedural knowledge to similar problems than reading did. The results also showed that typing self-explaining can make similar high-quality self-explanation proportion to spoken self-explaining. In addition, this study investigates the effects of content-free prompts by comparing with no prompts and with content-related prompts. This study uses the computer as a learning companion to provide prompts without understanding the content of students' self-explanations. The results showed that both content-free and content-related prompts made students generate more self-explanation modifications and perform better in applying target procedural knowledge to similar problems. (Contains 5 tables and 4 figures.)