How does speaking another language reduce the risk of dementia?

Abstract

Recently the question of whether or not bilingualism may protect against the development of Alzheimer’s disease (AD) has become a topic of great interest. Previous studies suggest that being bilingual slows the decline in executive control associated with the aging process. Spurred by these positive findings in normal subjects, investigators have tried to determine if being bilingual may be associated with delayed onset of AD. A number of studies collectively suggest that being bilingual may lead to a delay in the diagnosis of AD by as much as 5 years, even when taking into account confounders. Although a recent landmark study provides physical evidence confirming this advantage in AD patients, further work needs to be done to clarify its’ neuroanatomical basis.

Keywords::

• Alzheimer’s disease
• bilingualism
• cognitive reserve
• dementia
• neuroprotective

In the past decade, there has been increasing interest in determining if knowing another language protects against the development of dementia. As early as the 1990s, studies began to question whether knowing a second language has beneficial effects beyond the obvious. The underlying neural mechanism, though not entirely clear, is thought to relate to the suppression of one language and the recruitment of a second language, the so-called ‘switching’. How this early cognitive maneuvering exactly confers an advantage much later in life is, to date, unclear. The vast majority of research in this area has focused on three main areas. First, comparing neuropsychological performance in bilinguals and monolinguals; second, looking at outpatient memory clinic databases to compare subjective age of onset of dementia in monolingual versus bilingual patients; third, understanding the neuroantomical basis using functional neuroimaging. While there are limitations to each of these approaches, combined they provide evidence to suggest that bilingualism may be neuroprotective.

Evidence of the benefits of bilingualism on specific cognitive functions from neuropsychological testing is overwhelming across the entire age spectrum, from young to old. Bialystok and colleagues [1] reported the results of three experiments in which monolingual subjects were compared with bilingual subjects on the Simon task (a classic test of selective attention), demonstrating a robust advantage for bilingual subjects, suggesting that bilingual subjects experience less decline in age-related executive functioning. Subsequently, this finding has been replicated using numerous other tests of attention, including the Stroop test [2] and the flanker task [3]. Although this research suggests a theoretical advantage for bilinguals, it remains unclear to what extent preserved executive control impacts the onset of dementia.

While there are numerous studies showing the potential benefits of bilingualism cognitively, the impact of bilingualism on the development of dementia has recently become a subject of interest across the globe. Many studies to date have demonstrated that non-cognitive factors such as education, exercise, art and social engagement may all enhance ‘cognitive reserve’, thereby delaying the emergence of dementia [4]. It is unclear whether bilingualism may serve a similar role. The first study to examine the question of whether bilingualism delays the emergence of dementia symptoms was completed by Bialystok and colleagues [5]. Records from 184 cases of dementia, half monolingual, half bilingual, were reviewed. Patients who were bilingual were noted to have an average onset of dementia 4 years later than patients who were monolingual. Furthermore, decline in Mini Mental Status Examination score over time was equal in both groups, suggesting no change in the rate of disease progression. In a subsequent study [6] case records from 211 subjects with probable Alzheimer’s disease (AD) were reviewed. Monolingual patients were compared to bilingual patients in terms of age of onset of dementia. Patients who were bilingual were diagnosed 4.3 years later than monolingual patients and had onset of disease approximately 5.1 years later. Potential confounding effects, including education, immigration status and occupation, were controlled for, although it was noted in both studies that the majority of patients were immigrants. In a subsequent study by Chertkow and colleagues [7], attempts were made to mitigate the effects of immigration by focusing on a sample of patients who were bilingual but were non-immigrants and had lived most of their lives in Canada. The study looked at 632 patients from a memory clinic based in Montreal and demonstrated no advantage of bilingualism in terms of age of onset of dementia in patients who were bilingual, although patients who knew more than two languages had a slight advantage over patients who spoke one or two languages. In addition, there was a distinct advantage in the immigrant group of approximately 5 years in terms of delay of diagnosis, replicating the findings of Bialystok et al. and suggesting perhaps that immigration may be driving the observed changes. Gollan et al. [8] attempted to further quantify the effect of bilingualism using objective measures (Boston naming test) in a sample of 44 Spanish–English bilinguals. While higher degrees of bilingualism were associated with later onset of symptoms, this was thought to be driven primarily by education and to correlate with objective rather than subjective measures. A recent study based in India [9] examined this question in a sample of 648 non-immigrant multilingual patients, many of whom were functionally illiterate. It was demonstrated that patients who were bilingual developed dementia 4.5 years later than patients who were monolingual. Furthermore, a bilingual advantage was shown for different subtypes of dementia, including AD, fronto-temporal dementia and vascular dementia. Confounding factors, including immigration, education, sex, occupation and urban versus rural setting, were controlled for, and the study also included patients who were functionally illiterate, confirming that this was a robust finding. Further advances in the area of bilingualism and AD have focused on states that precede the development of AD, including mild cognitive impairment (MCI). In a recent paper, Ossher et al. [10] compared the age of onset for both single and multidomain amnestic MCI. An advantage was shown only for patients with the single-domain amnestic subtype of MCI, the subtype of MCI most likely to convert to AD, but no advantage for multidomain MCI. This suggests that the protective effects of bilingualism may be disease specific as the single-domain subtype is most likely to progress to AD. Although many studies to date have shown a potential advantage of bilingualism in patients with AD, there have been some limitations that should be emphasized. Many of the studies gathered limited data on bilingualism, have relied on subjective measures of disease onset and looked exclusively at memory clinic samples. Furthermore, a number of the studies did not control for confounders such as education and immigration. In addition, most of the studies looking at neuropsychological correlates have focused on healthy older adults, and few have examined patients with dementia. Nevertheless, collectively they suggest bilingualism may delay the onset of dementia by as much as 5 years, a benefit unparalleled by current drug treatments.

The advent of functional neuroimaging over the last decade has allowed scientists to try and localize what areas of the brain may be responsible for ‘switching’ between languages, and if being bilingual leads to reduced compensatory recruitment, with age, of brain areas involved in attention. There has been comparatively less exploration of this question in patients with dementia unfortunately. In a recent paper by Gold et al. [11], young and old bilingual subjects were compared on the ability to perform a perceptual task switching experiment while undergoing functional MRI. While age-related decrements in function were noted in both groups, the older bilingual group performed much better than the monolingual group and showed reduced recruitment of the left lateral frontal cortex and cingulate cortex. This finding suggests that bilingual subjects may have better cognitive reserve when it comes to attentional tasks as less brain activation is required for comparable performance. Luk et al. [12], in a meta-analysis of brain regions activated when attending to one language while suppressing another, identified six significant regions, four of which are thought to correspond to executive control regions. Recent work by Hernandez and colleagues [13] provide evidence to suggest the existence of an attentional network, possibly involving the dorso-lateral prefrontal cortex, which is invoked in bilingual subjects when switching from one language to another. It was postulated that the neural basis of bilingualism may be more broad than once was thought, involving not only traditional areas such as the dorso-lateral prefrontal cortex and superior parietal lobe but also areas involved in language processing such as the superior temporal gyrus and other areas of the brain dedicated to memory, somatosensory processing and emotion. It is therefore possible that bilingualism leads to the creation of a broader neural network that is more resistive to neurodegeneration or alternatively better able to compensate for neurodegeneration. An area that requires development is that of biomarkers. Very few studies have compared imaging or neuropathology in bilingual versus monolingual AD patients. Schweizer et al. [14] compared brain volumes of bilingual and monolingual AD patients matched on cognitive performance and level of disease. Bilingual patients showed significantly greater atrophy in regions specific to AD, including the radial width of the temporal horn, suggesting a distinct advantage for bilingualism and providing the first physical evidence that bilingualism may delay the emergence of AD. However, there have been no studies that have attempted to replicate this important finding in a larger sample of patients or compared other biomarkers including genetics, CSF and neuropathology.

In summary, evidence from neurocognitive studies in normal subjects show bilingualism can delay age-associated changes in executive control networks. This has been confirmed through functional neuroimaging. In addition, multiple studies have demonstrated a robust advantage for bilingualism in AD, conferring a delay of approximately 5 years, even when controlling for confounding factors such as immigration, occupation and education. Finally, these changes appear in both the earliest stages of the disease (amnestic MCI), and there is physical evidence to support the objective existence of this advantage. While drugs currently exist that may help to slow the progression of AD, none of them have the potential to delay the onset of symptoms in the manner in which bilingualism may [15]. Future prospective studies following large groups of patients with comprehensive clinical, imaging and neuropathological data will be required to answer the question as to what the underlying mechanism of this advantage is before these findings have any treatment implications. It is likely, based on research to date, that bilingualism may result in the establishment of a well-integrated neural network that not only may slow age-related effects on specific cognitive functions but also may protect the brain from neurodegenerative disorders, perhaps in the same way that education does.

Financial & competing interests disclosure

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending or royalties.

No writing assistance was utilized in the production of this manuscript.