Effects of audio headings on learning

Text-to-speech (TTS) programs often do a poor job of translating writing devices such as headings from visual into audio mode. Previous research studies have attempted to address this problem but these studies have mainly used heading detection tasks. The current study seeks to investigate (a) whether the presence of audio headings improves performance in natural learning tasks and (b) the type of heading rendering that is the most useful in natural learning tasks. Two experiments compared the effects of two types of rendering strategies on note-taking and cued recall. Results from this study revealed that a prosodic rendering strategy was most useful to the listeners in natural learning tasks.