Statement of Research Significance
Research Question(s) or Topic(s): This study examines how bilingualism and aging influence episodic autobiographical memory (EAM) specificity among Hispanic adults, a group underrepresented in cognitive aging research. Main Findings: Young bilingual Hispanic adults generated more episodic details than monolingual peers, suggesting a bilingual advantage in memory specificity during early adulthood. This advantage was not observed in middle-aged/older adults, indicating that age-related changes may reduce bilingual benefits. Older adults recalled fewer episodic details than younger adults, while non-episodic detail retrieval remained stable across age groups. Study Contributions: Extends research on autobiographical memory and bilingualism in Hispanic populations. Highlights age-dependent effects of bilingualism on memory specificity. Emphasizes the need for culturally and linguistically tailored cognitive assessments for Hispanic older adults.
Introduction
Hispanic individuals are the largest racial/ethnic minority population in the United States (U.S.) and represent one of the most rapidly expanding segments within the aging population, a trajectory projected to persist in the foreseeable future Funk & Lopez, 2022). Nonetheless, there are significant gaps in our understanding of how cognition changes, or not, with older age in U.S. Hispanic adults. One unknown pertains to how typical aging influences the way Hispanic individuals learn and remember personal, autobiographical memories. This becomes particularly relevant considering the prevalence of bilingualism among the Hispanic population (Pew Research Center, 2004), which may shape the way autobiographical memories are orally recalled and shared with others in daily life (Libersky et al., Reference Libersky, Crespo, Reppe and Kaushanskaya2023), and may delay or alter signs of age differences in this type of memory (Anderson et al., Reference Anderson, Hawrylewicz and Grundy2020; Bialystok et al., Reference Bialystok, Poarch, Luo and Craik2014; Chertkow et al., Reference Chertkow, Whitehead, Phillips, Wolfson, Atherton and Bergman2010; Kavé et al., Reference Kavé, Eyal, Shorek and Cohen-Mansfield2008). In the present study, we sought to determine how a well-established finding in cognitive aging, namely that older adults tend to recall autobiographical life events during structured, laboratory-based interviews in less specific detail relative to young adults (Levine et al., Reference Levine, Svoboda, Hay, Winocur and Moscovitch2002; Simpson & Sheldon, Reference Simpson and Sheldon2020), applies in a sample of bilingual and monolingual U.S. Hispanic adults.
Episodic autobiographical memory (EAM) retrieval is the mental capacity that allows individuals to recall specific events from their own lives. It encompasses the retrieval of specific details, emotions, and other contextual information related to singular moments in time (Conway, Reference Conway2005; Prebble et al., Reference Prebble, Addis and Tippett2013), contributing to one’s sense of identity, decision making, and social interactions (Alea & Bluck, Reference Alea and Bluck2003, Conway et al., Reference Conway, Singer and Tagini2004,). Recalling memories in high episodic detail has been shown to aid in decision-making by providing rich, detailed past experiences that guide future actions (Levine et al., Reference Levine, Svoboda, Hay, Winocur and Moscovitch2002; Holland & Kensinger, Reference Holland and Kensinger2010; Moscovitch et al., Reference Moscovitch, Cabeza, Winocur and Nadel2016). Additionally, episodic detail is believed to shape a person’s sense of self, influencing how and what individuals integrate from personal experiences into their identity (Conway & Pleydell-Pearce, Reference Conway and Pleydell-Pearce2000; Vanderveren et al., Reference Vanderveren, Bijttebier and Hermans2017). The ability to retrieve specific emotional memories is also associated with emotional regulation and psychological well-being (Williams et al., Reference Williams, Barnhofer, Crane, Hermans, Raes, Watkins and Dalgleish2007). Tracking episodic detail generation in autobiographical event memory therefore has broad implications for understanding the neuropsychology of everyday cognition.
Prior research has established that young and older adults differ in the way they orally retrieve and narrate EAMs (Addis et al., Reference Addis, Wong and Schacter2008; Schacter et al., Reference Schacter, Gaesser and Addis2013). Arguably the most prominent age difference in EAM retrieval pertains to episodic specificity, such that older adults, relative to young adults, narrate their EAMs with fewer details that capture the uniqueness of a singular moment and instead producing more details describing the background knowledge and meaning of the experience (Levine et al., Reference Levine, Svoboda, Hay, Winocur and Moscovitch2002; Schacter et al., Reference Schacter, Gaesser and Addis2013). Although why young and older adults differ in episodic specificity is a subject of ongoing research, this age effect is believed to reflect a combination of cognitive, neurological, and psychosocial factors associated with typical aging. For instance, cognitive decline within executive function may reduce the capacity for recalling detailed episodic memories in older adults (Park & Festini, Reference Park and Festini2017; Memel et al., Reference Memel, Woolverton, Bourassa and Glisky2018). Moreover, hippocampal structural integrity is associated with age- and disease-related changes (Setton et al., Reference Setton, Sheldon, Turner and Spreng2022), and hippocampal integrity among older adults has also been linked to the retrieval of episodic details from autobiographical memory (Memel et al., Reference Memel, Wank, Ryan and Grilli2020). Psychosocial shifts, such as changes in life priorities and a focus on broader life themes rather than specific events, may further contribute to reduced episodic detail in older adults (Grilli & Sheldon, Reference Grilli and Sheldon2022; Acevedo-Molina et al., Reference Acevedo-Molina, Matijevic and Grilli2020).
Most of our knowledge of how EAM retrieval differs between young and older adults comes from studies with samples primarily composed of non-Hispanic White adults or with populations in which race/ethnic background is not reported. As such, we know virtually nothing about the influence of older age on EAM retrieval in Hispanic adults. One study examined episodic memory, future thinking, and counterfactual thinking in young and older Colombian adults (De Brigard, Rodriguez, & Montañes, Reference De Brigard, Rodriguez and Montañés2017). The study found that younger adults generated more episodic details across all types of mental simulations than older adults. In contrast, older adults produced more non-episodic details. These findings suggest that some Hispanic groups may show typical age-related episodic specificity differences in autobiographical memory narration.
Although the work by De Brigard and colleagues (2017) addresses EAM in a Hispanic population, it is unclear if bilingualism influences the effect of age on EAM. Hispanic adults in the U.S. are commonly bilingual (Pew Research Center, 2004). Although controversial, findings from several studies have suggested that bilingualism may confer certain cognitive benefits, particularly in the domain of executive functions (Bialystok et al., Reference Bialystok, Craik and Luk2012; Goriot et al., Reference Goriot, van den Noort, Schelvis and Bosch2021; Lehtonen et al., Reference Lehtonen, Soveri, Laine, Järvenpää, de Bruin and Antfolk2018). Executive functions, in turn, have been associated with higher episodic specificity in studies of EAM retrieval in young and older adults (Addis et al., Reference Addis, Wong and Schacter2008; Piolino et al., Reference Piolino, Desgranges, Hubert, Bernard, Matuszewski, Chételat, Baron and Eustache2008). This relationship raises the possibility that practice switching languages leads to improved attentional control and memory monitoring, both of which are essential for maintaining the accuracy and detail of EAMs (Green & Abutalebi, Reference Green and Abutalebi2013), especially when the retrieval goals prioritize episodic specificity. Nonetheless, it is important to mention that a “bilingual advantage” for cognition has been heavily debated and recent evidence suggests that it is dependent upon a variety of factors, including task design and participant characteristics (Antoniou, Reference Antoniou2019; Lehtonen et al., Reference Lehtonen, Soveri, Laine, Järvenpää, de Bruin and Antfolk2018; van den Noort et al., Reference van den Noort, Bosch, Haverkort and Hugdahl2019; Marsullo et al., Reference Masullo, Casado and Leivada2024). Therefore, cognitive benefits observed in bilingual individuals may be influenced by factors beyond bilingualism itself, including cultural experiences and sociolinguistic environments.
Beyond executive functions, bilingualism allows a person to draw from two distinct sets of vocabulary. Two sets of vocabulary could enhance the ability to encode and articulate episodic memories more precisely, leading to richer and more vivid autobiographical recollections (Kroll et al., Reference Kroll, Dussias, Bogulski and Valdes Kroff2012). Also, bilingual adults often navigate different cultural frameworks, which may enhance their ability to encode and retrieve contextually rich and socially relevant memories (Marian & Neisser, Reference Marian and Neisser2000). Research by Marian & Kaushanskaya (2003) indicates that bilinguals’ narratives can shift between individualistic and collectivist themes depending on the language in use. This suggests that bilinguals may access a broader range of cognitive and cultural schemas during memory retrieval, which could further enhance the episodic specificity of their EAMs.
The overarching goal of the present study was to advance our understanding of aging and EAM retrieval in U.S. Hispanic adults and determine whether bilingualism influences EAM episodic specificity in this population. Given that age-related changes in memory performance can vary depending on the type of task used (Mair et al., Reference Mair, Poirier and Conway2021), it was important to test the roles of age and bilingualism among Hispanic adults using a measure that has consistently shown age differences. We therefore employed the Autobiographical Interview (Levine et al., Reference Levine, Svoboda, Hay, Winocur and Moscovitch2002), a well-established measure that captures both event-related episodic and non-episodic (semantic and other linguistic features, as well as unrelated events) details of autobiographical memories and consistently reveals robust age group differences (Simpson et al., 2023). We hypothesized that, consistent with findings in predominantly non-Hispanic White samples, older Hispanic adults who are cognitively unimpaired would recall EAMs with less specificity than their younger counterparts. Additionally, we posited that bilingual Hispanic adults—regardless of age—would demonstrate a greater ability to recall event-specific details compared to their monolingual peers. This would suggest that bilingualism might enhance EAM specificity across the adult lifespan, and therefore promote the benefits of recalling the autobiographical past in high episodic detail.
Methods
Participants
Table 1 presents demographic information for the entire sample. Data from 100 cognitively unimpaired young and middle-aged/older participants are reported in this study (50 bilingual Hispanic adults [30 young, 20 middle-aged/older]; 50 monolingual English-speaking Hispanic adults [30 young, 20 middle-aged/older]). Participants classified as young adults were between the ages of 18–35 and middle-aged/older adults were 50 and older. Participants were compensated for their time with a $20 gift card upon completion of the study. Participants were recruited through a combination of word-of-mouth, flyers, community events (in the Tucson, AZ area), and online advertisements. Recruitment primarily relied on online advertisements aimed at different departments within the university, which became particularly important during the COVID-19 pandemic when in-person recruitment was limited. Young adults were predominantly recruited from the university community, while middle-aged and older adults were recruited from community events and via word of mouth. Informed consent was provided by all participants and all study procedures were in compliance with and approved by the Institutional Review Board of The University of Arizona. This research was completed in accordance with Helsinki Declaration.
Table 1. Sample characteristics by group
a Results from one-way ANOVAs between groups or Chi-square for categorical variables.
Power analysis
An a priori power analysis conducted with G Power 3.1 evaluated the minimum sample size to detect medium-to-large size (f = .33) group effects and interactions. We settled on this expected effect size in light of our previous research and power analysis calculations involving studies of cognitive aging and event specificity in EAM (Acevedo-Molina et al., Reference Acevedo-Molina, Matijevic and Grilli2020). This power analysis revealed that we would have adequate power (>.80) to pass the p < .05 significance threshold (two-tailed) if our total sample included at least 90 participants. We intentionally aimed to surpass this required sample size and our sample sizes were settled on prior to enrolling participants in the present study.
Procedures
The present study was largely run in a remote/virtual environment during the first 1.5 years of the COVID-19 Pandemic. Potential participants were phone screened to determine study eligibility, which included being Hispanic per self-report, being proficient in English, and denying a history of psychiatric or neurological disorders. We screened for bilingualism via phone using the Language Experience and Proficiency Questionnaire (LEAP-Q) (Marian et al., Reference Marian, Blumfield and Kaushanskaya2007), a self-report tool designed to assess an individual’s language history and proficiency across different languages. Participants who rated themselves at least 7 on a 1–10 scale in speaking, understanding, and reading in both English and Spanish were placed in the bilingual group. Hispanic adults who reported knowing English only or who rated themselves 6 or lower on the LEAP-Q for Spanish were considered English monolingual. See Table 2 for sample characteristics for the bilingual Hispanic individuals obtained via self-report using the LEAP-Q.
Table 2. Bilingual sample characteristics by age group reported on the Language Experience and Proficiency Questionnaire
a First language acquired refers to the language participants reported learning first in life, either English or Spanish.
During the initial phone screen, we administered the Telephone Interview for Cognitive Status (TICS), which is an objective cognitive screening tool. Participants had to score 26 or higher on the TICS to be included in the study (Manly et al., Reference Manly, Schupf, Stern, Brickman, Tang and Mayeux2011). This cutoff was selected based on work by Manly and colleagues, which showed that this cutoff maximized sensitivity and specificity between normal cognition and mild cognitive impairment in Hispanic adults and non-Hispanic White older adults (Manly et al., Reference Manly, Schupf, Stern, Brickman, Tang and Mayeux2011). The consent process was done using REDCap econsent. Following enrollment, participants received instructions for their video conference appointment. Bilingual participants were also asked to complete the rest of the LEAP-Q (we asked question #3 as part of the screen). On a separate day, all participants completed a standard Autobiographical Interview (Levine et al., Reference Levine, Svoboda, Hay, Winocur and Moscovitch2002). Twelve young bilingual Hispanic adults completed the study in person prior to halting in-person research at the start of the COVID-19 pandemic. The rest of the participants completed the study over Zoom for Health.
Autobiographical interview
The Autobiographical Interview (Levine et al., Reference Levine, Svoboda, Hay, Winocur and Moscovitch2002) is a well-established method for assessing EAM. Participants are asked to retrieve unique event memories (i.e., EAMs) from different time periods and describe them in detail. For monolingual participants, a total of 5 memories were retrieved. For bilingual participants, 10 memories were retrieved, one in English and one in Spanish for each of the 5 time periods (order of Spanish and English was counterbalanced across bilingual participants). The narratives were audio recorded and manually transcribed by trained research assistants. The transcriptions were then analyzed following the standard Autobiographical Interview protocol. Specifically, each detail was scored as an “episodic” detail, meaning specific to the target event (e.g., description of the event, sensory details specific to the event), or an “non-episodic” detail, meaning semantic or factual statements, repetitions, or other details not specific to the event. Although the scoring protocol refers to episodic details as “internal” and non-episodic details as “external” details, for clarity and consistency, we use the terms episodic and non-episodic throughout this article. No automated coding software was used; the scoring was conducted manually by trained raters to ensure accuracy and adherence to the established guidelines. Because the language in which instructions are provided may influence memory retrieval, for the bilingual group, the instructions were provided in the language with which the memory would be narrated in.
Detail scoring reliability
As noted, memories generated by bilingual and monolingual Hispanic participants were scored by two raters who were trained on the Autobiographical Interview materials. Each person scored half of the narratives. Inter-rater reliability for detail scoring was calculated based on a random selection of ten young adults and ten older adults (20% of the total number of Hispanic participants), which were scored by both raters. Inter-rater reliability, for total details, episodic details, and non-episodic details was good to excellent (Cronbach’s α’s range = .83 to .93).
Statistical analyses
All analyses were conducted in JASP (Version 0.16.4), an open-source statistical software developed by the University of Amsterdam. JASP has been rigorously tested and validated against traditional software, ensuring that its computations are accurate and trustworthy for use in scientific research (Wagenmakers et al., Reference Wagenmakers, Love, Marsman, Jamil, Ly, Verhagen and Morey2018). All graphs were created using GraphPad Prism version 9.4.1 for Mac OS X (GraphPad Software, La Jolla California USA, ww.graphpad.com). Separate one-way ANOVAs were used to calculate differences between the groups for the variables presented in Table 1. Significant effects were followed up with relevant post-hoc tests. To control for multiple comparisons and reduce the risk of Type I errors, we applied the Holm correction method (Holm, Reference Holm1979). The Holm correction is a stepwise procedure that adjusts p-values in a manner that is less conservative than the Bonferroni correction, thus retaining more statistical power. To determine if EAM episodic specificity is sensitive to age in cognitively typical Hispanic adults, and to reveal whether bilingualism influences EAM episodic specificity on its own or in the context of age, we ran a two (Bilingualism Status: Bilingual Hispanic vs. Monolingual Hispanic) x two (Age: Young vs. Middle-aged/Older) x two (Detail: average episodic vs. non-episodic per memory) analysis of covariance (ANCOVA). Education was added as a covariate, given that young monolingual participants had significantly less education than all groups, as reported in the Results. Additionally, prior research has shown that individuals with higher levels of education often perform better on episodic memory tasks, likely due to the cognitive reserve associated with formal education (Manly et al., Reference Manly, Jacobs, Sano, Bell, Merchant, Small and Stern2003; Stern, Reference Stern2002). By controlling for education, we aimed to account for this potential confound and isolate the effects of bilingualism and aging on episodic specificity. For all analyses, significant effects were followed up with relevant post-hoc tests using the Holm correction. Data from the bilingual participants reported in the analysis are from memories retrieved in English. A separate analysis using data from memories retrieved in Spanish by the bilingual participants yielded comparable results. These analyses are reported in the Supplemental Materials section.
Results
Sample characteristics
Sample characteristics are reported in Table 1. Not surprisingly, age significantly differed among the groups F(3,96) = 396.99, p < .001, η p2 = .92. Middle-aged/older bilingual and monolingual participants were older than the young bilingual and monolingual participants (p’s < .001). Age among young bilingual participants did not significantly differ from young monolingual participants (p = .99). Similarly, age among the middle-aged older bilingual participants did not significantly differ from middle-aged/older monolingual participants (p = .51). Education significantly differed among the groups, F(3, 96) = 7.30, p < .001, η p2 = .18, such that young monolingual participants had less years of education than young bilingual participants, as well as middle-aged/older monolingual and bilingual participants (p’s < .009). Education did not significantly differ between young bilingual participants, middle-aged/older bilingual participants, and middle-aged older monolingual participants (p’s > .95). TICS scores did not significantly differ between the groups F(3, 96) = 2.43, p = .12, η p2 = .02 . Male to female ratio did not differ between the groups χ2(2, 100) = 0.09, p = 1.00. The majority of the sample identified as Mexican, followed by Puerto Rican, Cuban, and par ticipants from South or Central American countries, in descending order of representation.
Table 2 summarizes language characteristics reported by the bilingual sample, including dominant language, first language acquired, and proficiency in English and Spanish. Details on the age of acquisition for participants’ dominant language are provided in the text below. Among the young bilingual participants, 53% reported English as their dominant language, while 75% of middle-aged/older bilingual participants reported English as their dominant language. Among the young bilingual participants, 73% acquired Spanish as their first language, while all middle-aged/older participants acquired Spanish as their first language. Among the young bilingual participants, those who reported English as their dominant language reported acquiring it at a median age of 6 years (IQR = 4–7) and Spanish at a median age of 2 years (IQR = 1–3). Those who reported Spanish as their dominant language reported acquiring it at a median age of 2 years (IQR = 1–3) and English at a median age of 5 years (IQR = 4–6). All middle-aged/older bilingual participants identified Spanish as their dominant language, which they reported acquiring at a median age of 0 years (IQR = 0–1), indicating that they were exposed to Spanish from birth or early infancy. Middle-aged/older participants reported acquiring English at a median age of 5 years (IQR = 3–7). Regarding proficiency, all age groups rated themselves as fluent in speaking, reading, and understanding Spanish and English.
Is EAM episodic specificity sensitive to age in cognitively normal hispanic individuals and is bilingualism associated with higher episodic specificity?
Figure 1 shows episodic and non-episodic details generated by bilingual and monolingual Hispanic adults. An ANCOVA with factors of bilingualism status (i.e., bilingual Hispanic vs. monolingual Hispanic), age group (i.e., young vs. middle-aged older), and detail type (i.e., episodic vs. non-episodic), controlling for education, revealed a main effect of detail type, F(1, 95) = 13.28, p < .001, η p2 = .12, such that more episodic details were generated relative to non-episodic details, t = 16.28, p < .001, d = 2.37, 95% CI [1.96, 2.67]. There was a main effect of age, F(1, 95) = 37.01, p < .001, η p2 =.02, such that the young adults’ memories were more detailed than the older adults’ memories, t = 6.08, p < .001, d = 0.91, 95% CI [0.58, 1.23]. However, there was not a main effect of bilingualism status, F(1, 95) = 2.61, p = .10, η p2 =.02).
Figure 1. Adjusted means of episodic and non-episodic details generated by young and middle-aged/older bilingual and monolingual adults. This figure shows the adjusted means of episodic and non-episodic details generated by participants, controlling for years of education.
The main effect of age was qualified by a significant interaction between detail type and age, F(1, 95) = 55.94, p < .001, η p2 = .37. Whereas young adults generated more episodic details than middle-aged/older adults, t = 9.57, p < .001, d = 1.99, 95% CI [1.31, 2.68], young and middle-aged/older Hispanic adults did not significantly differ in the number of non-episodic details generated, t = .84, p = .40, d = .17, 95% CI [0.38, 0.73]. Although there was not a main effect of bilingualism, there was a significant detail type by bilingualism status interaction, F(1, 95) = 9.03, p = .003, η p2 = .08, indicating that bilingual Hispanic adults generated more episodic details than monolingual Hispanic participants, t = 3.25, p = .003, d = .69, 95% CI [0.10, 1.27]. Bilingual and Monolingual Hispanic adults did not differ on the number of non-episodic details generated, t = .93, p = .35, d = .19, 95% CI [0.37, 0.77]. There was not a significant interaction between bilingualism status and age, F(1, 95) = 1.57, p = .21, η p2 = .01.
Finally, there was a significant three-way interaction between bilingualism status, age, and detail type, F(1, 95) = 4.13, p = .04, η p2 = .04. Planned comparisons of interest revealed that young Hispanic bilingual adults generated more episodic details than young Hispanic monolingual adults, t = 4.19, p < .001, d = 1.17, 95% CI [0.23, 2.11], middle-aged/older Hispanic bilingual adults, t = 8.60, p < .001, d = 2.48, 95% CI [1.39, 3.57], and middle-aged/older Hispanic monolingual adults, t = 9.30, p < .001, d = 2.68, 95% CI [1.57, 3.80]. Young Hispanic monolingual adults generated more episodic details than both middle-age/older Hispanic bilingual and monolingual adults, t’s ≥ 4.19, p’s ≥ .001, d’s ≥ 1.17. Interestingly, middle-aged/older Hispanic bilingual and monolingual adults did not significantly differ on episodic details, t = .64, p =, d = .20, 95% CI [0.81, 1.21]. There were no significant differences in non-episodic details generated regardless of age (i.e., young vs. middle-aged/older) or bilingualism status (i.e., bilingual vs. monolingual), t’s ≤ .07, p’s = 1.00, d’s ≤ .37.
Discussion
General discussion
The primary goal of this study was to examine the relationship between bilingualism, age, and EAM, in a sample of U.S. Hispanic adults. The findings revealed an age-related reduction in episodic detail generation among Hispanic participants, with no significant difference in non-episodic detail between age groups. Notably, young bilingual participants demonstrated higher episodic specificity compared to their monolingual counterparts, an effect that was not observed between older bilingual and monolingual participants.
The findings of the present study demonstrate that young bilingual Hispanic adults exhibit a heightened level of episodic detail in their EAMs compared to their monolingual counterparts. This observed effect suggests potential benefits of bilingualism for enhancing personal memory storage and retrieval during early adulthood. By controlling for educational attainment, our findings indicate that the observed EAM advantage is unlikely to be solely an artifact of educational differences and instead may reflect a cognitive benefit associated with bilingual language use, though other contributing factors cannot be ruled out.
There are several potential reasons why bilingualism might contribute to an episodic specificity advantage in EAM among young adults. Bilingualism has been associated with enhanced executive functioning in some studies, which could facilitate the ability to describe personal life events in richer, more specific detail (Bialystok, Reference Bialystok2009; Bialystok et al., Reference Bialystok, Craik and Luk2012). Relatedly, access to two vocabularies may allow individuals to find more precise language when describing episodic memories, thereby potentially enhancing memory specificity. Executive functions—such as planning, cognitive flexibility, and inhibitory control—are thought to support the organization and elaboration of memories (Conway & Pleydell-Pearce, Reference Conway and Pleydell-Pearce2000; Moscovitch & Winocur, Reference Moscovitch, Winocur, Stuss and Knight2002). Additionally, higher attentional control may help individuals focus on relevant details while ignoring distractions, thus enhancing the vividness and specificity of their recollections (Gjorgieva et al., Reference Gjorgieva, Geib, Cabeza and Woldorff2023). Given that the task instructions of the Autobiographical Interview (Levine et al., Reference Levine, Svoboda, Hay, Winocur and Moscovitch2002) prompt participants to describe event-specific details, individuals with stronger attentional control might be better able to fulfill this goal, thereby generating more episodic content.
However, it is important to acknowledge that the link between bilingualism and executive functions remains a topic of debate. While some reviews suggest positive cognitive outcomes related to bilingualism, recent meta-analyses have highlighted inconsistencies in the evidence, with many studies failing to replicate a bilingualism advantage (Lehtonen et al., Reference Lehtonen, Soveri, Laine, Järvenpää, de Bruin and Antfolk2018; Paap & Greenberg, Reference Paap and Greenberg2013; de Bruin et al., Reference de Bruin, Treccani and Della Sala2015). Such variability suggests that the cognitive effects associated with bilingualism may be influenced by additional factors, including cultural context, sociolinguistic environments, and individual differences (Calvo & Bialystok, Reference Calvo and Bialystok2014; Masullo et al., Reference Masullo, Dentella and Leivada2024). As a result, while the findings of the present study support the notion that bilingualism is associated with enhance EAM specificity in young adults, these effects should be interpreted with caution and understood as part of a complex interplay of cognitive, linguistic, and environmental influences.
Another possible explanation for this finding is the impact of culture on autobiographical memory. Cultural norms significantly influence how memories are valued, recalled, and narrated, (Park & Huang, Reference Park and Huang2010; Wang & Ross, Reference Wang, Ross, Kitayama and Cohen2007; Ross & Wang, Reference Ross and Wang2010; Gallo et al., Reference Gallo, Penedo, Espinosa de los Monteros and Arguelles2009; Stone et al., Reference Stone, Johnson, Stone-Romero and Hartman2006). Individuals from collectivist cultures focus on relational and communal aspects in their narratives, whereas individuals from individualistic cultures emphasize personal achievements and experiences (Wang, Reference Wang2021, Reference Wang2004; Wang & Brockmeier, Reference Wang and Brockmeier2002; Wang & Ross, Reference Wang and Ross2005). In the context of our study, Hispanic individuals living in the U.S. navigate between the collectivist cultural norms of their heritage and the individualistic values prevalent in American society. Indeed, research suggests that bilingual individuals often achieve higher degrees of biculturalism compared to their monolingual counterparts (Grosjean, Reference Grosjean2015; Nguyen & Benet-Martínez, Reference Nguyen and Benet-Martínez2013), and bicultural individuals may develop unique cognitive skills and competencies, allowing them to effectively blend and organize their cultural identities (McCarty et al., Reference McCarty, Yim and Garcia2023; Szymanski & Kalra, Reference Szymanski and Kalra2019). These skills may facilitate the integration of diverse cultural perspectives into autobiographical narratives, enriching the detail and richness of such memories (Nguyen & Rule, Reference Nguyen and Rule2020). Therefore, bilingualism and biculturalism may significantly shape the autobiographical memories of Hispanic individuals in the U.S., adding layers of detail and depth less commonly found in monolingual narratives.
In addition to individualism and collectivism, other cultural factors, such as storytelling and oral traditions, may contribute to the differences in autobiographical memory performance observed in this study. Many Hispanic cultures place a strong emphasis on oral traditions, where personal and collective stories are passed down through generations (McDowell et al., Reference McDowell, Herrera-Sobek and Cortina1993; Morales, Reference Morales2013). These practices may reinforce the ability to recall events in rich detail, as they require individuals to frequently engage in detailed recollection and narration of life events (Mace, Reference Mace2010).). The cultural significance of oral traditions could thus be an important factor in shaping the autobiographical memory processes of Hispanic participants, contributing to the differences observed between bilingual and monolingual groups (Schrauf & Rubbin, Reference Schrauf and Rubin2003).
Notably, the bilingual advantage did not manifest in the older adult group, as there were no significant differences in episodic or non-episodic detail generation between middle-aged/older bilingual and monolingual Hispanic adults. This result aligns with literature that has failed to find a consistent bilingual advantage in older age groups (Bak, Reference Bak2016; Kousaie & Phillips, Reference Kousaie and Phillips2012; Paap & Greenberg, Reference Paap and Greenberg2013; de Bruin, Treccani, & Sala, Reference de Bruin, Treccani and Della Sala2015). Although some studies support the notion that bilingualism contributes to cognitive reserve (Bialystok et al., Reference Bialystok, Craik and Freedman2007; Gold et al., Reference Gold, Johnson and Powell2013), the effects of bilingualism may not be sufficient to overcome some cognitive changes associated with typical aging (Hedden & Gabrieli, Reference Hedden and Gabrieli2004; Park & Reuter-Lorenz, Reference Park and Reuter-Lorenz2009). Furthermore, it is important to consider that older adults may have distinct narrative goals compared to younger adults, which might not emphasize episodic detail. For example, Grilli & Sheldon (2023) propose that older adults tend to focus more on the “gist” of autobiographical events rather than specific details. This shift toward gist memory is hypothesized to not merely be a consequence of cognitive decline but rather partly reflective of adaptive changes in priorities and motivational factors that come with aging, such as sharing the broader meaning and emotional tone of experiences, as well as social and identity-related goals. This narrative shift could explain the absence of a bilingual advantage in episodic specificity among older adults, as memory strategies may be geared toward general rather than specific details, negating the benefit that bilingualism might have provided.
Heterogeneity in bilingual characteristics might have also contributed to the present findings. The impact of a second language may be influenced by numerous factors that vary in older adults who are bilingual, such as immigration history, age of acquisition, frequency of language use, and sociocultural context (Birdsong, Reference Birdsong2006). Moreover, some older adults experience prolonged periods of language attrition, where either use of their first or second language decreased over time, potentially weakening the cognitive benefits associated with bilingualism (Schmid & Keijzer, Reference Schmid and Keijzer2009). Additionally, socioeconomic factors might play a role in the degree to which bilingualism can support cognitive health in older adults. For example, older adults from lower socioeconomic backgrounds might not have had the same opportunities or support systems that could reinforce the cognitive advantages of bilingualism, thus impacting the overall efficacy of bilingualism as a cognitive reserve (Calvo et al., Reference Calvo, García, Manoiloff and Ibáñez2016). Lastly, individual differences in cognitive reserve capacity and resilience to cognitive decline might interact with bilingualism in complex ways, making it challenging to isolate the specific contributions of bilingualism to cognitive health in aging populations.
Clinical implications
The findings of this study suggest that heightened episodic specificity in young bilingual adults may have significant clinical implications. EAM plays a central role in personal identity, decision-making, and emotional regulation, and disruptions to EAM are commonly observed in clinical populations such as those with traumatic brain injury (Esopenko & Levine, Reference Esopenko and Levine2017; Piolino et al., Reference Piolino, Desgranges, Manning, North, Jokic and Eustache2007; Wammes et al., Reference Wammes, Good and Fernandes2017). It is possible that heightened episodic specificity among young bilingual individuals is associated with functional benefits in these decision-making, social, and self-related targets of autobiographical memory and may buffer against deleterious effects of certain neuropsychological conditions, such as acquired brain injury.
Individuals also may be able to leverage their bilingualism in cognitive rehabilitation settings. For instance, incorporating bilingual memory exercises or language-switching strategies in therapeutic interventions may enhance the retrieval of episodic memories and foster better outcomes in patients recovering from brain injury or managing neurodegenerative conditions (Gollan et al., Reference Gollan, Montoya and Werner2011). Moreover, the social and emotional functions of EAM—such as its role in self-regulation and mood stability—are particularly relevant in the context of mental health interventions, where bilingual individuals may benefit from tailored approaches that integrate their linguistic and cultural backgrounds (Alea & Bluck, Reference Alea and Bluck2003).
These findings also raise important considerations for neuropsychological assessments and treatments. Clinicians working with bilingual populations may need to adjust their diagnostic tools and therapeutic approaches to better capture the nuances of memory in bilingual individuals. The cognitive and cultural flexibility conferred by bilingualism could be harnessed to improve interventions for mood disorders, anxiety, and post-traumatic stress disorder, where memory specificity often plays a critical role in treatment efficacy.
Limitations
In addition to highlighting the dynamic interaction between bilingualism, aging, and EAM, the present study motivates several additional areas for future investigation. Expanding these studies to include diverse ethnic and linguistic groups, and employing longitudinal designs, would help determine the generalizability of the bilingual advantage in memory and its evolution over time (Luk & Bialystok, Reference Luk and Bialystok2013). Future studies should also explore the impact of lifestyle factors (e.g., diet, alcohol intake, smoking, physical activity, etc.) on memory, as these have been shown to differ in Hispanic adults relative to monolingual Non-Hispanic White adults (Fernandez, Reference Fernandez2021) and to influence memory (Celik et al., Reference Celik, Kokje, Meyer, Frölich and Teichmann2022; Raizada & Kishiyama, Reference Raizada and Kishiyama2010).
Although this study focused on bilingualism, acculturation—a process that includes language proficiency, cultural practices, and social behaviors—may have played a role in shaping participants’ autobiographical memory recall. Previous research suggests that acculturation can affect cognitive performance, including how individuals encode and retrieve EAMs (Medina et al., Reference Medina, Henry, Torres, MacDonald and Strutt2023; Mendoza et al., Reference Mendoza, Garcia, Duara, Rosselli, Loewenstein, Greig-Custo, Barker, Dahlin and Rodriguez2022). The absence of a formal measure of acculturation is a limitation of this study, and future research should incorporate validated acculturation scales to capture this complex process and its potential impact on memory performance in Hispanic. Relatedly, only participants who identified as Hispanic were included in the study. While we recognize that ‘Hispanic’ encompasses diverse groups, we only collected information about the specific country/region that each participant primarily identified with their Hispanic background. However, due to the small sample sizes for each group, we were unable to analyze country/region as a factor. Future studies should investigate how cultural differences within Hispanic populations, along with factors such as immigration status, may impact the findings.
Additionally, future studies should use objective measures to determine bilingualism/monolingualism in their samples. For example, language dominance can significantly influence neuropsychological assessment outcomes, highlighting the value of reliable measures of language proficiency and dominance to ensure accurate evaluation and diagnosis in bilingual individuals (Suarez et al., Reference Suarez, Gollan, Heaton, Grant and Cherner2014). Although the LEAP-Q provides a standardized approach to gathering historical language exposure data, it relies on the accuracy of retrospective self-reports, which can be susceptible to recall bias (Umanath & Marsh, Reference Umanath and Marsh2014). For instance, it is important to note that age of acquisition estimates in our study are based on participants’ responses to a questionnaire, which used a scale to approximate the age at which languages were acquired. Some participants interpreted this question as knowing a language from birth, which is why some age estimates are recorded as ‘0,’ indicating that they were exposed to this language from birth to early infancy. This methodological consideration should be taken into account when interpreting the age of language acquisition as it reflects participants’ earliest linguistic memories rather than precise timelines.
Similarly, while the TICS was a practical choice for remote cognitive screening, it was originally developed for detecting dementia in older adults. Given that our study includes both younger and older participants, its applicability across this broader age range warrants consideration. Additionally, prior research (Manly et al., Reference Manly, Schupf, Stern, Brickman, Tang and Mayeux2011) has noted that while the TICS is useful for identifying severe cognitive impairment, it has limited sensitivity in distinguishing mild cognitive impairment from normal cognition. This suggests that future studies may benefit from more robust screening tools, such as the Telephone Montreal Cognitive Assessment (T-MoCA; Pendlebury et al., Reference Pendlebury, Welch, Cuthbertson, Mariz, Mehta and Rothwell2013) or face-to-face measures like the Mini-Mental State Examination (MMSE; Folstein et al., Reference Folstein, Folstein and McHugh1975), for more precise cognitive classification. Another consideration is the potential for cultural and linguistic factors to influence TICS performance. Although the TICS has been validated in diverse populations, its use in bilingual Hispanic individuals requires further study, as test performance may be influenced by language proficiency and acculturation. Future studies incorporating more comprehensive neuropsychological assessments could better account for these factors and ensure a more nuanced understanding of cognitive performance across diverse Hispanic populations.
Lastly, the high educational attainment of our participants does not fully represent the broader Hispanic population in the U.S., where education levels vary widely. Research has shown that Hispanic individuals who participate in cognitive studies are often those with higher educational attainment and greater access to healthcare systems, which impacts their awareness and willingness to participate in research (Ejiogu et al., Reference Ejiogu, Norbeck, Mason, Cromwell, Zonderman and Evans2011; González et al., Reference González, Tarraf, Gouskova, Gallo, Penedo, Davis, Lipton, Argüelles, Choca, Catellier and Mosley2015). These recruitment patterns contribute to selection bias, as individuals with lower education levels may be less likely to respond to research advertisements or be aware of research opportunities (Whitfield et al., Reference Whitfield, Allaire, Belue and Edwards2008). Although we aimed to recruit a diverse population, self-selection bias inherent in study participation may have limited the representation of all segments of the Hispanic community.
Conclusion
In conclusion, our study contributes valuable insights into the effects of bilingualism and aging on autobiographical memory among Hispanic adults in the U.S. Although our findings confirm the presence of an age-related decline in the specificity of EAM among U.S. Hispanic adults, consistent with previous research in predominantly non-Hispanic White samples, the impact of bilingualism presents a more complex picture. Our results suggest a bilingual advantage in EAM among young, but not older, U.S. Hispanic adults. This study highlights the importance of including diverse linguistic and cultural perspectives in memory research to enhance our understanding of how bilingualism and cultural factors interact with cognitive aging.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S1355617725000141.
Acknowledgements
The Arizona Department of Health Services also provided support. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The authors have no conflicts of interest to disclose.
Funding statement
Research reported in this publication was supported by the National Institute on Aging of the National Institutes of Health under Award Number AG068098. The first author (Mónica Acevedo-Molina) was supported by an NRSA F31 under Award Number AG069443.
Open access
