Shared book reading, reading sessions between adults and young children, leads to better general vocabulary development. However, studies show that preliterate children, although able to remember words from books immediately, do not retain most of the book-specific words they learn. Thus, words learned from books may require additional memory supports to be retained after a delay. The current dissertation integrates memory, word learning, and book reading research to characterize both the variability and presentation timing of words within shared book reading sessions and examines how these factors lead to increased retention of words learned from books. First, this dissertation determines how variability in the questions asked during shared book reading affects the retention of novel words. Specifically, in Study 1, 38 4-year-olds were read books with words that were accompanied by irrelevant questions, the same questions, or different questions each time the word was presented. Next, I examined how two different presentation timing styles affect the retention of novel words. In Study 2, 39 4-year-olds were read books with words presented either sequentially or spaced in time. Children in both studies were tested at three different time delays (i.e., immediate, 5 minutes, and 24 hours) to determine how many words were retained in memory. This dissertation found, broadly, that neither variability nor presentation timing led to better word retention across the three different time delays. However, children in this study remembered more words than previous studies even when tested after 24 hours. Implications of these preliminary findings are discussed. Ultimately, these results inform word learning theories and serve to create materials for design-based interventions aimed to improve the quality of shared book reading sessions between caregivers and children.