TABLE 1.
The studies associating bilingualism to reduced incidence of dementia.
| The age of L2 exposure | Subjects | Languages | Major findings | References | ||||
| Age (Mono/Bi) | Number (Mono/Bi) | L1 | L2 | |||||
| Neurological mechanisms underlying bilingualism | – | 7.2/6.9 | 20/20 | Seven different languages | English | Bilinguals performed better than monolinguals in memory and attention tasks | Bialystok and Feng, 2009 | |
| <10 | 21.9/21.1 | 54/55 | Nineteen different languages | English | ||||
| <11 | 70.5 ± 3.0 | 14/14 | English | Different languages | The connectivity of frontoparietal control and default mode networks were increased in bilinguals | Grady et al., 2015 | ||
| 0 | 8.5 ± 0.5/8.5 ± 0.5 | 15/15 | English | Thirteen different languages | Bilingual children responded faster in inhibitory control and cognitive flexibility than monolinguals | Bialystok and Viswanathan, 2009 | ||
| 8.5 ± 0.5/8.6 ± 0.7 | 15/30 | Tamil Telugu | English | |||||
| 0 | 11.1 ± 0.8/11.4 ± 0.9 | 10/14 | French Dutch | Roman German | The MFA value of lIFOF were significantly higher in simultaneous bilinguals than those in monolinguals, and the highest degree change in the MFA value of LIFOF was sequential in bilinguals | Mohades et al., 2015 | ||
| 3 | 11.1 ± 0.8/11.3 ± 1.0 | 10/16 | ||||||
| <1.5 | –/20.3 ± 3.7 | –/19 | Catalan | Spanish | Language switching revealed activation in the left caudate nucleus, pre-SMA, and ACC in bilinguals. Additionally, the left caudate nucleus was involved in forward switching, and pre-SMA and ACC were involved in backward switching | Garbin et al., 2011 | ||
| <5 | 23.5 ± 3.2/23.5 ± 3.2 | –/6 | Spanish | English | The dorsolateral prefrontal cortex was involved in language switching. The reaction time in the dorsolateral prefrontal cortex was decreased in bilinguals as compared with that of mixed language individuals. Additionally, the activation of the dorsolateral prefrontal cortex was increased in bilinguals as compared with that in mixed language individuals | Hernandez et al., 2000 | ||
| 5 ∼ 11a,b | 23.5 ± 4.5/23.5 ± 4.5 | 14/14 | German | Italian English | The activity of the left putamen was increased in bilinguals when they learned a non-proficient language | Abutalebi et al., 2013 | ||
| <5 5 ∼ 13 13< | 80.8 ± 4.3/80.8 ± 4.3 | 18/18 | Thirteen different languages | English | Extensive practice controlling both languages helped older adults remember episodic memories | Schroeder and Marian, 2012 | ||
| <6 | 43.0/43.0 | 10/10 | Tamil | English | Bilinguals were more effective at controlling processing than monolinguals. Bilingualism helped delay age-related losses in certain executive processes | Bialystok et al., 2004 | ||
| 71.9/71.9 | 10/10 | |||||||
| <6 | 25.6/23.9 | 24/24 | Different languages | English | Bilinguals resolved different types of response faster than monolinguals and these patterns increased with age | Bialystok et al., 2006 | ||
| <12 | 66.9/64.5 | 24/24 | ||||||
| 10.3 ± 7.3 | 23.0 ± 4.1/21.8 ± 2.4 | 23/15 | English | Chinese | There were no differences in the Stroop interference effect between English speakers who learned Chinese and monolinguals for both languages. | There was no difference in lexical access speed between bilinguals and monolinguals, but the lexical access speed of L2 in bilinguals was delayed as compared with that of L1 in bilinguals | Coderre et al., 2013 | |
| 11.0 ± 2.7 | 23.0 ± 4.1/21.0 ± 1.6 | 23/22 | Chinese | English | The stroop interference effect was reduced for both languages when Chinese speakers learned English as compared with monolinguals | |||
| 11.0 ± 7.0 | 25.9 ± 6.4/26.8 ± 6.6 | 66/67 | English | Spanish | Selective attention skills of bilinguals were improved at low working memory requirements | Lee Salvatierra and Rosselli, 2010 | ||
| 19.7 ± 5.7 | 63.4 ± 8.4/64.8 ± 7.3 | 42/58 | ||||||
| 12 | 21.4/21.4 | –/24 | Chinese | English | The dorsal anterior cingulate cortex and supplementary motor area were involved in local inhibition. The dorsal left frontal gyrus and parietal cortex were involved in global inhibition | Guo et al., 2011 | ||
| 19 | 49.0 ± 16.0/49.0 ± 16.0 | 13/14 | Chinese | Sign-language | Switching of sign language and spoken language exhibited high functional activation of the left caudate nucleus region. | Zou et al., 2012 | ||
| Structural changes in the brain by bilingualism | 10.2 ± 4.2 | 28.2 ± 5.3/31.9 ± 8.1 | 25/20 | Different languages | English | Tract-Based Spatial Statistics analysis indicated that fractional anisotropy values for bilinguals in several WM tracts were higher than those in monolinguals | Pliatsikas et al., 2015 | |
| <11 | 70.6 ± 3.0/70.4 ± 3.7 | 14/14 | English | Different languages | The temporal lobe cortical thickness was decreased in elderly monolinguals. This was not observed in bilinguals, and the frontal lobe WM integrity was higher in bilinguals than that in monolinguals | Olsen et al., 2015 | ||
| <11 | 70.5 ± 3.0/70.5 ± 3.0 | 14/14 | English | Different languages | The WM integrity in the anterior to posterior functional connectivity was higher in elderly bilinguals than that in elderly monolinguals | Luk et al., 2011a | ||
| 5 ∼ 7a | 26.6 ± 4.2/23.4 ± 4.6 | 14/17 | German | Italian | The activity of GM in the dorsal anterior cingulate cortex was increased in bilinguals | Abutalebi et al., 2012 | ||
| <5 | - | 25/25 | English | European languages | The density of GM was increased in early and late bilinguals | Mechelli et al., 2004 | ||
| 10 ∼ 15 | 25/33 | |||||||
| <7 | 23.1 ± 4.8/21.5 ± 2.7 | 22/22 | Catalan | Spanish | Heschl’s gyri were larger in bilinguals than those in monolinguals | Ressel et al., 2012 | ||
| 11.6 ± 1.2 | 25.4 ± 4.3/25.4 ± 4.3 | 12/12 | German | French | The activities of the left caudate and anterior cingulate cortical areas were increased in bilinguals | Abutalebi et al., 2008 | ||
| – | 6.92 ± 6.80/62.17 ± 5.36 | 23/23 | Chinese | English Cantonese Mandarin | The GM volumes in left temporal pole were increased in the aged bilingual brain | Abutalebi et al., 2014 | ||
| – | 61.85 ± 6.71/63.2 ± 5.86 | 30/30 | Cantonese | English Mandarin | Bilinguals were increased GM along the ACC | Abutalebi et al., 2015b | ||
| – | 71.42 ± 4.88/77.13 ± 4.52 | 40/45 | German Italian | Italian German | The bilingual individuals were increased ECN and DMN metabolic connectivity | Perani et al., 2017 | ||
aKindergarten age. bElementary school age. ACC, anterior cingulate cortex; Bi, bilingual; DMN, default mode network; ECN, executive control network; GM, gray matter; lIFOF, left-inferior-occipitofrontal fasciculus; MFA, mean fractional-anisotropy; Mono, monolingual; L1, native language; L2, secondary language; SMA, supplementary motor area; WM, white matter.