Three experiments investigated the hypothesis that, when interpreted in context, general terms are typically encoded on the basis of an instantiation. The results indicated that a particular term naming the expected instantiation of a general term was a better cue for the recall of a sentence than the general term itself, even though the general term had appeared in the sentence and the particular term had not. This could not have happened if people encode and store the core meanings of general terms. It was theorized that people instantiate in order to select, from among the indefinitely many meanings a term can have, a sense which permits a coherent overall interpretation of the message. (Author)