This paper focuses on the optimum use of representational space by words in speech and in the mental lexicon. In order to do this we use the concept of entropy in information theory to plot the information contour of words. We compare different systems of Spanish words taking into account the extent to which they accurately represent speech: a citation vs. a fast-speech transcription and a dictionary lexicon vs. the speech lexicon. We also compare the information profiles yielded by the speech corpus vs. that of the speech lexicon in order to contrast the representation of words over two representational spaces: time and storage space in the brain. We discuss the implications for the mental lexicon. Finally we apply the information profile tool to speech addressed to a child.