The input enhancement hypothesis proposed by Sharwood Smith (1991, 1993) has stimulated considerable research over the last 15 years. This article reviews the research on textual enhancement of input (TE), an area where the majority of input enhancement studies have aggregated. Methodological idiosyncrasies are the norm of this body of research. Seven major issues appear to be limiting the generalizability of the findings and holding up further progress in the understanding of the efficacy of TE for learning: (1) noticing and/or acquisition; (2) TE and comprehension; (3) simultaneous or sequential processing; (4) TE and the nature of the enhanced form; (5) TE and prior knowledge; (6) TE and input flood; and (7) TE and overuse. The existing research has nonetheless offered some important insights that future research should seek to build on.