Linguistic Approaches to Bilingualism - Volume 6, Issue 1-2, 2016

Healthy non-pathological aging is characterized by cognitive and neural decline, and although language is one of the more stable areas of cognition, older adults often show deficits in language production, showing word finding failures, increased slips of the tongue, and increased pauses in speech. Overall, research on language comprehension in older healthy adults show that it is more preserved than language production. Bilingualism has been shown to confer a great deal of neuroplasticity across the life span, including a number of cognitive benefits especially in executive functions such as cognitive control. Many models of bilingual language processing have been proposed to explain bilingual language processing. However, the question remains open of how such models might be modulated by age-related changes in language. Here, we discuss how current models of language processing in non-pathological aging, and models of bilingual language processing can be integrated to provide new research directions.

Among papers considering L2 performance, a subset take into account the length of residence (LOR) in the country where the L2 is spoken. In about half of these, LOR makes no difference for performance of at least one variable measured. Since those who reside in an L2 environment for many years tend to be older, the beneficial effects of longer LOR may at some point be counteracted by declines due to aging. This article draws on research in cognitive aging to consider how age could impact L2 performance. This is particularly important when investigating the effects of LOR or age of L2 acquisition since LOR, age of acquisition, and age at testing are linearly dependent variables, making conclusions based on any of these variables problematic. We argue that aging is a largely ignored confound in the literature on L2 attainment, particularly for studies that include older adults in their samples.

This study addresses a gap in the literature on executive function advantages among bilingual speakers by investigating a group of elderly, long-term, immersed bilinguals. Our participants are native Dutch speakers who emigrated to Australia as adults and have spent many years in that country. They are compared on a range of cognitive and linguistic measures to native Dutch and native English control groups. We argue that, due to the massive differences in the bilingual experience, group analyses may fall short of capturing the full picture. We argue instead for a more qualitative approach, which takes into account as detailed a picture of bilingual development, daily language habits and, in particular, code-switching habits as possible.

We investigated age-related decline of bilingual language control. Thirteen older and 13 younger bilinguals performed a verbal fluency task (completing the same letter and semantic categories in each language and switching languages after every category), and a non-linguistic flanker task. In letter fluency, bilinguals produced fewer correct responses after switching languages, suggesting inhibition of the previously-used language. However, this testing-order effect did not differ between groups and older bilinguals produced few wrong-language intrusions, implying intact ability to apply inhibition in older age. In contrast, age-related deficits in the flanker task were robust, implying dissociations between language control and domain-general executive control. In semantic fluency, there were no testing-order effects but older bilinguals produced more intrusions than younger bilinguals, and more intrusions than in letter fluency. Thus, bilinguals may flexibly modulate the degree of inhibition when they can benefit from semantic priming between languages, but less efficiently so in older age.

Recent research suggests that bilingual experience reconfigures linguistic and nonlinguistic cognitive processes. We examined the relationship between linguistic competition resolution and nonlinguistic cognitive control in younger and older adults who were either bilingual or monolingual. Participants heard words in English and identified the referent among four pictures while eye-movements were recorded. Target pictures (e.g., cab) appeared with a phonological competitor picture (e.g., cat) and two filler pictures. After each eye-tracking trial, priming probes assessed residual activation and inhibition of target and competitor words. When accounting for processing speed, results revealed that age-related changes in activation and inhibition are smaller in bilinguals than in monolinguals. Moreover, younger and older bilinguals, but not monolinguals, recruited similar inhibition mechanisms during word identification and during a nonlinguistic Stroop task. Results suggest that, during lexical access, bilinguals show more consistent competition resolution and recruitment of cognitive control across the lifespan than monolinguals.

Studies across the lifespan have revealed modifications in executive control (EC) from bilingualism, but studies of working memory (WM), a key aspect of EC, have produced varied results. Healthy older (M = 71.0 years) and younger participants (M = 21.1 years) who were monolingual or bilingual, performed working memory tasks that varied in their demands for EC. Tasks included a star counting task, a flanker task, and a nonverbal recent probe memory task. Bilinguals performed similarly to monolinguals on the star counting task after controlling for differences in vocabulary. Monolinguals were faster than bilinguals on the flanker task with only age group differences significant for the WM manipulation. Bilinguals were faster than monolinguals on the nonverbal recent probe memory task, particularly for the condition that included proactive interference. The interpretation is that better bilingual performance in nonverbal working memory tasks is linked to the need for executive control.

Increasing our understanding about neuroprotective lifestyle variables has become a practical imperative in our aging society. Cognitive reserve (CR) refers to the use brain resources in a way that allows for coping with neuropathology and maintaining cognitive functioning. A growing body of evidence suggests that bilingualism may represent a form of CR against Alzheimer’s disease (AD). The purpose of the present review is to summarize both behavioral and neuroimaging evidence for bilingualism as a reserve variable against AD. The potential influences of literacy, intelligence, immigration status are discussed. Evidence is reviewed suggesting that bilingualism may delay clinical AD symptoms by protecting against age-related declines in the brain’s executive control circuitry. It is suggested that such potential beneficial effects within executive control systems may enable bilinguals to circumvent the typical effects of AD pathology on symptom expression for several years.

Recent findings indicate that bilingualism delay the onset of dementia. Using data from the Betula longitudinal cohort study on memory, health and aging (www.betula.su.se) the issue of a possible protective effect of bilingualism was addressed.

Monolingual (n = 736) and bilingual (n = 82) participants (≥ 60 years) without dementia at inclusion were followed for incident dementia over a time-period up to 10 years. In total, 112 participants developed dementia. Analyses were performed with Cox proportional hazards regression adjusted for age, sex, and presence/absence of the Apolipoprotein E (APOE) ɛ4 allele, with dementia outcome as the dependent variable.

Results showed no delayed onset of dementia in bilinguals compared to monolinguals. However, because of the findings from a study using participants from the same population showing beneficial longitudinal effects of bilingualism on episodic memory; we argue that our results may depend on the frequency of use of the second language after retirement.

Within the current debates on cognitive reserve, cognitive ageing and dementia increasingly showing a positive effect of mental, social and physical activities on health in older age, bilingualism remains one of the most controversial issues. Some reasons for it might be social or even ideological. However, one of the most important genuine problems facing bilingualism research is the high number of potential confounding variables. Bilingual communities often differ from monolingual ones in a range of genetic and environmental variables. In addition, within the same population, bilingual individuals could be different from the outset from those who remain monolingual. We discuss the most common confounding variables in the study of bilingualism, aging and dementia, such as group heterogeneity, migration, social factors, differences in general intelligence and the related issue of reverse causality. We describe different ways in which they can be minimised by the choice of the studied populations and the collected data. In this way, the emerging picture of the interaction between bilingualism and cognitive aging becomes more complex, but also more convincing.