Few memories from my first course in cognitive psychology have stuck with me over time. There was the time the time the professor was appalled when he learned how many drinks the young fraternity member who always sat in the front row downed every weekend. More relevant to the course material, I also remember thinking that serial position memory curves—the idea that, when given a list of words to remember, people perform much better on the earliest items (primacy effects) and the most recent items (recency effects)—were fascinating. And I remember when my professor went to the chalkboard and plotted age vs. accuracy for serial memory tasks, telling us to enjoy our nubile brains now, because it only goes downhill from here.

My former professor and other researchers who study memory, language, and aging recognize that the story is more complex than this—cognitive function and age do not have a simple linear relationship. Some functions, like everyday language comprehension, appear to remain relatively intact in throughout our lifespan. Other functions, like memory for recent autobiographical events, are thought to diminish with age. But new research from a computational modeling group suggests that not all age-related differences—for example, the slowing that occurs when performing certain language or memory tasks—should be considered deficits. One obvious way in which older adults differ from their younger counterparts is that they have more experience at their disposal. This means they have accrued more world knowledge—something that may contribute to certain visible slow-downs in cognitive tasks.

Language and aging: comprehension vs. production

Researchers Meredith Shafto and Lorraine Tyler recently described several ways aging affects language processing. Although many studies report only limited major differences during typical online language comprehension between younger and older adults, other studies find differences in so-called “offline” language comprehension processes—for example, in answering questions about material that has just been read. Shafto and Tyler argue that these differences aren’t related to language, per se, but are instead related to more general cognitive processes such as keeping relevant information active and ignoring irrelevant information.

More striking, however, are age-related differences in language production. In particular, older adults show more difficulty when asked to name pictures: they take longer to react and can produce incorrect responses.

Older adults are also more impaired on tasks designed to produce a “tip-of-the-tongue” (TOT) state. TOT phenomena occur when an individual has the feeling of knowing what word they want to say without being able to access the word or information required to say it. In such tasks, when people are given a picture (or other item) and asked to judge whether they recognize it, older adults are more likely to report TOT states than younger adults.


Photo credit: Shafto & Tyler, Science, 2014.

Cognitive scientist Katherine DeLong and her colleagues conducted a study investigating the role of prediction in language comprehension: in other words, whether your brain can anticipate a word before you actually read it. DeLong investigated the extent to which younger and older adults would pre-activate likely upcoming information as they read sentences word-by-word while their brain activity was monitored by EEG. The subjects read sentences where a target word (always a noun) was either highly predictable (e.g., ‘The bakery did not allow credit cards so Peter would have to write a check’) or not (e.g., ‘The bakery did not allow credit cards so Peter would have to write an apology…).

The size of the brainwaves of younger adults was modulated not only at the critical noun (i.e., check vs. apology) but also at the preceding article (a vs. an) (see also an earlier study on younger adults). Younger adults therefore showed evidence of predicting word-form information (whether the upcoming word started with a vowel or not) before they actually read the critical (predicted) word. In contrast, older adults showed the effect at the noun, but not the article. The researchers also investigated a later effect (occurring 500-900 ms after the onset of the critical noun) thought to index “mispredicting,” that is, predicting the most likely continuation (e.g., check) but actually reading something else. Both younger and older adults showed this later effect to some extent, but among older adults it was most pronounced in those who were very verbally fluent (a language production measure). These findings are consistent with previous suggestions that prediction during language comprehension might actually rely on the production system: as people age, increased language production abilities may benefit language comprehension, as well.

Knowledge and aging: vocabulary and beyond

While aging is often discussed in terms of decline of cognitive function, there’s a lot to be said for the role of knowledge and its accrual in aging (for a recent discussion see this recent episode of BBC Radio4 Frontiers). For both older and younger individuals, expertise within a domain (e.g., chess, cooking, music, medicine, and many others) leads to better performance on language and memory tasks within that domain and can eliminate age-related differences in memory. Moreover, older adults benefit from prior knowledge when learning new information within a given domain—both domain-relevant prior knowledge and overall knowledge can lead to better language comprehension along with reading speed. In other words,  knowing more seems to eliminate the otherwise-present effects of aging in these language tasks.

One area where older adults consistently outpace younger adults is in vocabulary. Psychologist Paul Verhaeghen performed a large meta-analysis of 210 papers from 1986-2001 and found that older adults have much larger vocabularies than younger adults. In a recent study, cognitive scientist Michael Ramscar and his colleagues found that slower reaction times in older adults might be a natural consequence of knowing more words. For example, when asked to name a picture, older adults might take longer not due to cognitive decline but because they know more possible answers.

Ramscar’s work could impact how society looks at and defines aging. If subtle differences in language and memory are indeed the natural outcomes of the human information processing system, an understanding of these processes may lead neurologically healthy older adults to feel more at ease and less worried about the process of aging. In fact, the authors suggest “… the myth of cognitive decline is leading to an absurd waste of human potential and human capital. It thus seems likely that an informed understanding of the cognitive costs and benefits of aging will benefit all society, not just its older members.”

Further Reading

Image Credit: Allan Ajifo via flickr

About The Author

Melissa Troyer

Melissa Troyer is a Ph.D. student at the University of California, San Diego, where she studies the cognitive neuroscience of language and memory. Melissa holds B.S. degrees in Psychology and Cognitive Science and a double B.A. in Linguistics and French from Indiana University and a master's degree in Cognitive Science from MIT. Melissa is also a contributor to the UCSD-based writing group NeuWrite San Diego.