Introduction to the Virtual Collection on Language and Aging

This virtual collection assembled by the Journal of Gerontology: Psychological Sciences includes five articles focusing on language function and use in older adults. These articles examine a broad range of questions relating to language and aging, including age differences in language output, links between language and other cognitive domains, and the impact of speaker characteristics such as bilingualism. While much previous research has focused on comparisons between younger and older adults, several of these studies examine language function and use across the life span, including in middle-aged adults. The study of language is central to aging research: language is critically important in everyday life, and communication is central to quality of life. Moreover, language ability predicts cognitive health in late life (e.g., Snowdon et al., 1996). While most aspects of language function remain relatively intact in later life, certain changes are observed: older adults may have increased difficulty understanding spoken language, and word-finding difficulties may occur. It has been suggested that some of these difficulties may be related to declines in other cognitive abilities such as working memory (see Kemper & Kliegl, 2002). Other aspects of language function, such as general knowledge and vocabulary, may even improve with age. The first two articles in this virtual collection analyze naturalistic language to explore the subtle changes in language function that can occur in healthy aging, and their links with other cognitive domains. First, Luo et al. (2020) recorded conflict conversations from 364 couples aged 19–82, and investigated linear and nonlinear effects of age on language use, including grammatical complexity and utterance of filled pauses. Consistent with prior research, they found that grammatical complexity increased until midlife (age 54), and then decreased, while the utterance of filled pauses increased until old age (age = 70) and then declined slightly. These findings are important for two reasons. First, they demonstrate that relations between age and language use can be curvilinear. Second, the findings may help to shed light on the connections between language use and other cognitive domains. For instance, working memory capacity shows a similar age trajectory as grammatical complexity, suggesting a possible association between the two. Similarly, increases in filled pauses up to age 70 may indicate a mechanism to offset word-finding difficulties, which increase with age. The subsequent decrease after age 70 may reflect avoidance of words that might produce difficulties. Second, Polsinelli et al. (2020) also used natural language samples to explore the association between executive function measures and language output. Natural language was recorded from 102 cognitively healthy older adults over the course of 4 days, using an electronically activated recorder that intermittently recorded short snippets of ambient sound, resulting in a log of everyday conversations. Participants also completed a battery of executive function tests. Overall, higher executive function, particularly working memory, was associated with more analytic, complex, and specific language, as well as with other language markers such as a less positive emotional tone, and more sexual and swear words. These findings provide important new evidence that natural language use is associated with executive function. In the third article in this virtual collection, Taler et al. (2019) examined changes in language organization with age by analyzing data from a widely used clinical task, verbal fluency. Over 12,000 middle-aged and older adults completed a 1-min animal fluency task as part of the Canadian Longitudinal Study on Aging (CLSA). A computational modeling approach was used to identify the factors driving performance. Overall, with increasing age, participants produced slightly fewer items, and the items they produced were of higher frequency and greater semantic neighborhood density. The semantic model provided more unique power in predicting age from fluency performance than total number of items produced, the standard approach to scoring this task. These findings show that there are subtle changes in the way that people perform this task as they age, which may reflect an accumulation of knowledge resulting in greater reliance on environmental characteristics in performing the task. An area that has been the focus of increasing research interest in recent years is the impact of speaker characteristics, such as bilingualism, on cognition. Particular interest has focused on performance on tasks of language and executive function. Chan et al. (2020) have extended this question to examine the association between a speaker’s level of active bilingualism and their executive control. In this study, active bilingualism was defined as “the regular balanced use of two languages and language switching.” Seventy-six community-dwelling older adults completed self-report measures of active bilingualism as well as a battery of executive control tasks. Participants who reported more balanced bilingualism usage and less frequent language switching showed higher goal maintenance and conflict-monitoring abilities, suggesting that active bilingualism may assist in maintaining specific executive control abilities in older adults. Finally, Turner and Stanley (2019) examined the use of language styles and perceptions of electability in younger and older adults, with the goal of shedding light on the paradox that, while older adults are viewed as “warm but not competent,” they are nonetheless highly represented in politics. In this study, 90 young and 90 older speakers gave impromptu 5-min speeches campaigning for an important position in a club or organization. Speeches were recorded and analyzed for use of individualistic language (“I”-language) and collectivistic language (“we”-language). Raters were then asked about each speaker’s electability. Overall, younger adults used more “I”-language and less “we”-language than older adults. Moreover, use of individualistic language led to reduced electability across age groups, while use of collectivist language had no impact on perceptions of electability. Thus, older adults were rated as more electable than young adults due to their reduced use of “I”-language in their campaign speeches. The studies included here examine participants from a wide range of ages: Luo et al. (2020) included participants from across the life span, and Taler et al. (2019) included both middle-aged and older adults. Such an approach allows for a more fine-grained examination of the ways in which language use and function may change nonlinearly throughout the life span. Moreover, use of computational approaches and large scale databases is an important step forward in understanding subtle changes in language processing with age. Future work can use such approaches to identify markers of incipient cognitive decline that will be important clinically as well as theoretically. Novel approaches such as collection of natural speech samples to assess language processing also provide important avenues for future work. Taken together, the findings of these studies shed light on changes in language use across the life span, associations between language and other cognitive domains, particularly executive function, the impact of speaker characteristics such as active bilingualism, and perceptions of older adults based on their use of different language styles.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.