Feature

Drs. Lori James and Deborah Burke found that providing people with phonological cues helps them access difficult-to-recall words by strengthening the connections between word sounds.

A snippet of a name comes to mind, but only that much. No amount of concentration can bring the word from the tip of the tongue to where it would be more useful. Until hours later, maybe, when the forgotten word pops up, seemingly out of the blue. The puzzle is solved--but did it solve itself?

Tip-of-the-tongue experiences, or TOTs, are among the more irritating of everyday nuisances. For the young, the nagging feeling of knowing a name or word but not being able to retrieve it can be frustrating, inconvenient and occasionally embarrassing. For older adults, who tend to have them more frequently, TOTs can seem disturbingly--if mistakenly--like early signs of dementia.

In the TOT state, people can access semantic, or conceptual, information about a word, but they can't access enough phonological information to pronounce the word. Psychologists know that TOTs arise from deficits in the phonological system, the level of the language production system at which word sounds are represented. But the specific nature of the problem has remained uncertain.

In the November issue of APA's Journal of Experimental Psychology: Learning, Memory and Cognition (Vol. 26, No. 6), psychologists Lori James, PhD, of the University of California, Los Angeles, and Deborah Burke, PhD, of Pomona College, report that providing phonologically related cues helps both young and older adults avoid and overcome TOTs. Their research, supported by the National Institute on Aging, sheds greater light on memory changes during aging and bolsters psychologists' understanding of language production and memory.

Unknown origins

One long-standing notion of how TOTs arise has been that words with similar sounds block target words from coming fully to mind. Evidence against this blocking theory began to emerge several years ago, but the idea has remained influential--perhaps in part, Burke suggests, because it matches people's experience of what TOTs feel like.

"If you have an alternate word that comes to mind, the phenomenological experience is that that word is preventing you from getting the target word," Burke observes. "But that's not consistent with the data."

James and Burke propose a different explanation, which stems from a more general model of language production, called the Transmission Deficit model. That model, first proposed in 1991 by Burke and her collaborator--and spouse--Don MacKay, PhD, a University of California, Los Angeles, psychologist, holds that language production depends on the strength of connections within a network that includes conceptual and phonological levels.

The phonological level is particularly vulnerable to lapses in retrieval, according to the theory, because there are no inherent connections between word sounds, as there are between concepts. When a word has not been used frequently or recently, the already-tenuous phonological connections weaken, making it difficult to retrieve the word, even when its meaning is known.

Older adults are especially prone to TOTs, the Transmission Deficit model holds, because connections in memory become weaker as people age. That weakness is likely to be most evident where there are the fewest connections from the outset--as is the case for phonology.

Providing phonological cues

To test the Transmission Deficit model's predictions for TOTs, James and Burke asked research participants 114 general-knowledge questions designed to evoke target words that had been shown to provoke a high rate of TOTs. For example: What word means to formally renounce a throne? Target words--in this example, abdicate--included proper names and other seldom-used words.

For some trials, questions were preceded by a series of 10 prime words, half of which shared at least one phonological feature with the target word. For example, for the target word abdicate, a prime word was abstract. In other cases, all 10 words were phonologically unrelated to the target. None of the list words shared semantic meaning with the target words.

As each prime word appeared on a computer screen, participants were asked to say the word aloud. So that participants would not guess the purpose of the prime words and use them to generate target words deliberately, they were instructed that the purpose of the task was to assess the difficulty of pronouncing certain words. In line with this cover story, participants were asked to rate how difficult each word was to pronounce.

When participants had been primed with words that were phonologically related to target words, James and Burke found, they made more correct responses and had fewer TOTs than when they were primed with unrelated words. What's more, that effect emerged for both young and older participants.

In a second experiment, designed to more closely mimic real-life circumstances under which TOTs occur, the researchers reordered the tasks. Participants were first asked a general-knowledge question designed to evoke a target word. Only if they responded either that the word was on the tip of their tongues or that they didn't know the response were they given the series of words to pronounce. Then, the question was posed again.

As before, the phonological cues helped with word retrieval. Both young and older participants who had indicated that the correct response was on the tip of their tongues were able to generate the correct response more frequently after they had been primed with words that were phonologically related to the target word, compared with when the word list contained only unrelated words. That wasn't the case for participants who had indicated that they didn't know the response, indicating that the phonological primes didn't prompt a simple guessing strategy.

Implications

The finding that phonological cues helped both young and older participants refutes the popular blocking hypothesis, the authors argue. Instead, they say, their results provide support for the Transmission Deficit model, showing that strengthening weak phonological connections helps avert TOTs for people at any age.

In addition, James notes, "The results say something about this interesting feeling that we have when we're trying to resolve tip-of-the-tongue states, when it suddenly feels as though the word has just suddenly popped into mind. Our results indicate a possible way that those pop-ups happen--that we've likely recently encountered the phonology in the environment."

"It suggests that these auditory or phonetic cues are experienced in normal language discourse, and that they trigger the missing word, unblocking the road jam without our awareness," concurs psychologist Alan S. Brown, PhD, of Southern Methodist University.

Gary Dell, PhD, of the University of Illinois at Urbana­Champaign, comments, "We often think that our ability to speak is fixed early in life. This view is incorrect. I think that language learning never stops. We're constantly tuning our abilities to understand and our abilities to speak throughout our lives. This is why we're able to be so effective as speakers and as comprehenders. This experiment provides a nice demonstration of this tuning process."

Apart from their theoretical implications, the findings are also instructive from a practical standpoint, because they underscore the importance of maintaining language ability across the adult lifespan.

"People should keep using language, keep reading, keep doing crosswords," says James. "The more you use your language and encounter new words, the better your chances are going to be of maintaining those words, both in comprehension and in production, as you get older."

"Many older adults interpret the frequency and the severity of these retrieval problems as a signpost for cognitive decline," comments University of Kansas psychologist Susan Kemper, PhD. "TOT problems bolster negative stereotypes of older adults and reinforce views of older adults as cognitively impaired, as suffering from memory limitations. I think this work goes a long way toward explaining why this becomes such a problem for older adults."

Despite the many theoretical and practical issues that the research addresses, Burke points out, the study leaves some questions unanswered. She says, "We presented the [general-knowledge] question a second time, in the second experiment. What we don't understand is whether hearing or saying 'abacus,' for example, would make 'abdicate' come to mind when that's no longer the immediate attentional focus--which in many cases, in real life, it's not."

An important further step, says Willem Levelt, PhD, a psychologist at the Max Planck Institute for Psycholinguistics in Nijmegen, The Netherlands, will be to identify whether it's necessary that people actually articulate the phonological cue words, or whether simply hearing or reading them silently may be sufficient.

"James and Burke did the right experiment because they wanted to get an effect, and the best thing you can do is to include the levels of processing that may be of importance," Levelt says. "The question now is, where does this priming effect arise? It would be a nice experiment to just repeat it exactly as it was done, but with the word lists just auditorially presented to the subjects."

This article is part of the Monitor's "Science Watch" series, which reports news from APA's journals.