Tip-of-the-Tongue Phenomenon: Knowing a Word You Cannot Say

Producing a word takes two things that usually arrive together: its meaning and its sound. You think of the concept, and the spoken form follows so quickly that the two feel like one act. A tip-of-the-tongue state is what it feels like when they come apart. The meaning is fully present, you know exactly which word you want and could define it, but the link to its sound has not fired, so the form will not come. The sense that recall is imminent is the system reporting that it has found the right entry and is still trying to read out its pronunciation.

That two-stage picture explains why partial knowledge survives. The sound of a word is not stored as a single lump but as parts, and some parts come back before others. The first letter and the number of syllables are among the most accessible, which is why people in the state can report them above chance while the whole word stays out of reach. Brown and McNeill called this generic recall, and noted that the beginnings and endings of words are recalled better than their middles, the same pattern that shows up when people learn lists. A word is retrieved from the edges inward.

The most common account of the block itself is a transmission deficit: the connection from meaning to sound is weak, often because the word is used rarely or has not been used lately, so not enough activation reaches the sound to produce it. This predicts the words that cause the most states, proper names and low-frequency words, and it predicts that the state is more common with age, as connections that are used less often weaken. A competing account holds that a similar-sounding word is blocking the target, but the blocking idea has less support, since the related words that come to mind usually arrive after the block, not before it, and often help rather than hinder.

Tip-of-the-tongue states are also a case of metacognition, the mind judging its own contents. The feeling is a signal about memory rather than memory itself, and it is usually accurate: words that produce the feeling are more likely to be recognized and eventually recalled than words that produce a flat blank. People act on the signal, keeping up the search for a word that feels close and giving up on one that does not, which is mostly a good bet.

Roger Brown and David McNeill published "The tip of the tongue phenomenon" in the Journal of Verbal Learning and Verbal Behavior in 1966. Earlier observers had described the state, but it was hard to study because it happens unpredictably. Brown and McNeill's method was to bring it on deliberately. They read 49 definitions of uncommon English words, such as sampan, nepotism and sextant, to 56 Harvard and Radcliffe students and asked them to supply the word. In the pretest interviews, where the experimenter watched a single student, the signs of the state were unmistakable, something like the brink of a sneeze. In the main experiment the students sat in groups, raised a hand when a definition produced the state, and before the answer was read out wrote down the word's first letter, its number of syllables, and any words that were coming to mind.

The readings produced 360 of these states. Over the positive cases, where the word the student was reaching for turned out to be the intended target, students guessed the first letter correctly 57 percent of the time by the figure Brown and McNeill gave in their text; Gregory Jones noted in 1989 that the paper's own detailed counts imply about 50 percent. By either figure, for words that often began with uncommon letters, the rate was well above any plausible chance level. Their explicit guesses at the number of syllables were correct about as often, and the rank order of their syllable guesses matched the rank order of the real counts almost perfectly. The words that came to mind while they searched split into two kinds: words that sounded like the target and words that meant something like it. The sound-alikes matched the target's first letter and syllable count far more often than the meaning-alikes did, which told Brown and McNeill that a word in this state is held by its form, not only by its meaning.

They drew a picture of the mental dictionary from the result. A word's features are stored separately and retrieved separately, with the accessible features, the ones at the beginning and end and the overall length, coming first. Later work filled in the account. Diary and laboratory studies, reviewed by Alan Brown in 1991, established that young adults have a tip-of-the-tongue state roughly once a week and that older adults have them more often, with less partial information available, which fits the idea of connections that weaken with disuse. Bennett Schwartz's work, including his 2002 book on the subject, developed the view that the feeling is a metacognitive signal rather than a direct readout of the blocked word.

The state is a common part of word retrieval at every age. The rows below are illustrations, not findings from the studies cited on this page.

A tip-of-the-tongue state usually resolves on its own, and the main thing to counter is the urge to force it, which can prolong the block. The methods that help work with the partial knowledge the state leaves available.

• Say the parts you have. The first letter, the rough length and the stress often come before the word; voicing them gives the search something to build on.
• Stop trying for a moment. Many states resolve after attention moves away, when the word pops up unbidden. Deliberate searching can keep a wrong candidate active and in the way.
• Run the alphabet or the sounds, not the meanings. Because the block is on the form, cycling through first letters or syllable shapes is more likely to release the word than thinking harder about what it means.
• Describe it to someone else. Their guesses supply sounds from the outside, and a sound close to the target frequently triggers it.
• Treat the feeling as reliable information. A word that feels close usually is, and is worth a little more search; a flat blank usually is not, and is better looked up.
• Keep occasional lapses in proportion. Tip-of-the-tongue states become more frequent with age in the healthy adults these studies describe. Word-finding trouble that is new, persistent or getting worse, or that worries you or the people around you, is worth discussing with a qualified professional.

The tip-of-the-tongue state is a specific kind of retrieval failure, with a feeling of nearness and partial access. Several neighbors are ordinary forgetting of other kinds.

• It is not cue-dependent forgetting, where a memory is inaccessible until the right cue appears and there is no sense that it is about to come. The tip-of-the-tongue state carries that sense, and leaves parts of the word available without any cue.
• It is not absent-mindedness, the failure to encode or retrieve something because attention was elsewhere. In a tip-of-the-tongue state the word was learned and is known; the block is specifically on reading out its sound.
• It is not source confusion, mixing up where a memory came from. The tip-of-the-tongue state is not an error about a word's origin; it is a gap between knowing a word and producing it.