the boundaries of that word are encountered. Then, when that word is read, it appears extremely and excitingly fitting, and some of the emotional accompaniments to solving a TOT may be engaged. Reading clearly involves a constructive process, highly influenced by systematic context effects (Hochberg, 1970;Rayner, 1975;Tulving & Gold, 1963). Brown (1970), in a largely ignored discussion of the causes of the TOT phenomenon, notes that reading makes use of a noticing order, that is, a regular order of processing the features of a word. Poems, I am arguing, through their constraints on meaning, rhyme, rhythm, and first letters, are capable of putting a reader part way through the noticing order of a word that is not yet read. When the word is read, it has already been constructed to such an extent that a special recognition experience takes place.One other critical parallel between key words in poems and the TOT phenomenon should be mentioned. They each concern problems of word production. If either type of word could be produced easily, the situation, by definition, would not be a TOT, and, in the case of poetry, would not be interesting or original (Kammann, 1966).If TOT and these poetic processes are equivalent, then people who differ in poetic processes should differ in TOT processes. Poets (poetry lovers) and nonpoets (poetry haters) should differ in the number of TOT experiences they have or in some other aspect of the TOT phenomenon.2 Insofar as poetic constraints involve TOT-like processes which in turn affect the appreciation of poetic passages, it seemed reasonable that, when in the TOT state, poets may be more susceptible than nonpoets to being helped by cues, especially those cues normally supplied in poetry.Frances (1958) showed that musically trained subjects were better than untrained subjects at making discriminations of rhythmic alterations in simple melodies. In addition, James' discussion of the TOT (which Thanks go to K. Bryant, M. Savage, and J. Sommers for their assistance in Experiment I; to B. J. Buckley, K. Wold, and M. Schulman for their assistance in Experiment 2; to 1. Cutting, in part for suggesting the use of the auditory filter; and to Kathleen MacHan Kozlowski. This research was supported by a grant from Wesleyan University. Reprints are available from the author at the Department of Psychology, Wesleyan University. Middletown. Connecticut 06457.Auditory cues (the target word distorted by a low-pass filter) and rhyming cues act as.retrieval aids for people who have a word on the tip of their tongue. Parallels between the tlp-of-thetongue phenomenon (TOT) and the perception of well-put passages of poetry are also discussed. It is argued that the effects of these poetic passages derive in part from the engagement of TOT-like processes. In support of this hypothesis, poets (poetry appreciators) are shown to be more aware of being helped by TOT retrieval cues than are nonpoets (poetry nonappreciators); however, the retrieval cues do not differentially influence successful recall of TOT ...