The naming disadvantage of the dominant language in bilingual word production has been observed in healthy individuals; however, there is currently limited to no evidence of its presence in age-related disorders. 

In this study, our primary objective was to investigate potential naming disadvantages in individuals with Mild Cognitive Impairment (MCI) based on their type of bilingualism (active or passive). To this aim, we assessed naming performance in individuals with MCI and healthy older adults (OA) with diverse bilingual profiles in Catalan and Spanish which were defined by language proficiency, acquisition, and use.

A total of 58 OA and 124 patients with MCI took part in the study. Based on their responses to a bilingualism questionnaire, participants were classified into either active or inactive bilingual categories. The active bilingual group (OA=32; MCI=41) had early proficiency in both Catalan and Spanish (before the age of 6) with a balanced use of both languages. The inactive bilingual group (OA=26; MCI=83) included individuals who were Spanish-dominant with limited usage and low proficiency in Catalan. While active bilingual participants completed the naming task in both Catalan and Spanish, their inactive bilingual counterparts only performed the task in Spanish. Additionally, participants' neuropsychological and cognitive reserve data were gathered for analysis.

We conducted analyses on name accuracy and latencies using linear mixed models, which included various linguistic variables (semantic category, word frequency, and phonemic length), bilingualism (active vs. inactive), group (OA vs. MCI), global cognition, and cognitive reserve scores as fixed effects.

A preliminary analysis indicated that there was no significant difference in naming accuracy between the dominant (Catalan) and non-dominant (Spanish) languages among active bilinguals (p= .57).

For the naming performance in the dominant language, MCI patients exhibited significantly poorer performance (accuracy= 87.5%;  naming latency= 1582 ms) compared to OA participants (accuracy= 91.3%; naming latency= 1498 ms; ps < .05).

Furthermore, all linguistic variables significantly modulated accuracy (ps < .001), while for naming latencies, only phonemic length and word frequency were significant (ps < .05). Interestingly, the impact of bilingualism was not significant on its own, but it did interact significantly with the phonemic length of words (p = .04). This interaction revealed that, while naming latencies for active bilinguals remained unaffected by phonemic length, inactive bilinguals experienced slower naming for longer words (1557 ms) compared to shorter words (1607 ms, ps < .01). 

The naming accuracy in both OA and individuals with MCI for their dominant language do not appear to be negatively impacted by their level of bilingualism. If anything, active bilinguals are not affected by word length as opposed to inactive bilingual individuals. This suggests a potential protective role in word retrieval stemming from the use of a second language.