Sex Differences in Cognitive Performance Persist into Your 80s

doi:10.21203/rs.3.rs-4900286/v1

Download PDF

Research Article

Sex Differences in Cognitive Performance Persist into Your 80s

https://doi.org/10.21203/rs.3.rs-4900286/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

Sex differences in cognitive performance have been extensively documented. Understanding the underlying factors contributing to sex differences in older adults is imperative to promote healthy cognitive ageing. Sex hormones, estrogens, and testosterone have been suggested to be associated with cognition. Nevertheless, there is a scarcity of studies investigating the sex difference in cognitive performance and the contribution of gonadal hormones in older adults. Hence, the current study aimed to investigate sex differences in cognitive performance and elucidate the association between gonadal hormones and cognitive performance in 80+-year-olds.

Methods

Using confirmatory factor analysis in a sample of 131 older adults (aged 80 to 92 years), 17 cognitive performance measures were divided into two cognitive components: executive functioning and memory. Subsequently, mediation analyses were conducted to determine the direct effect of sex and the indirect effect mediated by gonadal hormones on executive functioning and memory.

Results

Females outperformed males in executive functioning and memory. However, gonadal hormones did not mediate the sex effect on cognitive performance. Interestingly, estrogen significantly predicted executive functioning but not memory. Testosterone was not found to predict either executive functioning or memory.

Conclusion

Our study confirms enduring sex differences in memory and executive function, even among individuals aged 80 and above. Current gonadal hormone levels do not mediate these differences. While estrogen may predict executive function, its influence does not explain the sex differences. These findings underscore the complex nature of cognitive disparities between sexes in older age, warranting further investigation into underlying mechanisms.

Cognition

older adults

executive function

hormones

memory

estrogen

sex differences

sex dimorphism

testosterone

Cognitive decline is well-documented throughout the ageing process, with a notable decrease typically observed from the mid-'50s onward [1, 2]. This decline primarily affects fluid mental abilities, including memory, executive functioning, processing speed, and reasoning [1]. While these changes are a natural aspect of ageing, there is considerable variation among older adults regarding cognitive performance and its decline rate. Besides ageing, various reasons may explain this heterogeneity, such as educational level, physical fitness, lifestyle, and non-modifiable factors such as sex [3–5].

Much research has been conducted to determine the differences in cognitive performance between males and females. The results are often equivocal due to the lack of sex differences, or when differences are apparent, they are often only of a small magnitude [6]. For example, executive functions are seen as an essential set of cognitive processes necessary to control and coordinate other cognitive abilities and behaviors. A recent meta-analysis suggests no sex differences in overall executive functioning, however, there are some task-specific differences in other cognitive domains [4]. This meta-analysis from Gaillard and colleagues [4], encompassing over 600 studies, unveiled a superiority in episodic memory performance among females, accompanied by a discernible difference due to the content that was to be recalled. The results indicated that females had an advantage in verbal tasks, such as naming words, sentences, and prosing nameable images and locations (g = .11 – .28). Conversely, males have been identified to score better than females on spatial tasks involving mental manipulation and visualization of three-dimensional objects, such as mental rotation and matching with effect sizes ranging from medium to large (d = .56 – .94) [3]. Moreover, a recent meta-analysis [5] has also identified that these significant male advantages remain for visual-spatial working memory measures across 180 effect sizes from persons aged 3 to 86 years (mean d = .16). Interestingly, mean age was identified as a significant moderator, whereby the magnitude of the sex differences in visual-spatial working memory increased with age [5].

These sex differences in cognitive performance persist across the lifespan, including in older adulthood. Munro et al. [7] conducted a study involving 957 participants aged 67–88 years (480 females and 477 males), identifying that older females displayed an advantage on tests assessing psychomotor speed, verbal learning, and memory tasks. Conversely, males had an advantage on tasks involving visual construction and perception. These findings tend to suggest that sex differences observed early in life continue to manifest in older adulthood, highlighting the enduring nature of these cognitive disparities [8–10].

Sex differences in cognitive performance are attributed to various biological factors, including neuronal lateralization [11, 12], region-specific brain dimorphisms [12–14], and gonadal hormone levels [15–19]. Research indicates that high circulating gonadal hormone levels correlate with observable effects on cognition in young adults, such as improved performance in visuospatial tasks with higher testosterone levels [19, 20]. Whilst estrogen has been associated with greater verbal and nonverbal memory [21], with cognitive performance varying across different age groups corresponding to estrogen concentrations [21]. Throughout life, individuals are exposed to varying levels and concentrations of hormones due to sex differentiation. Notably, cognitive decline often coincides with significant hormonal transitions such as menopause in females (average age 51), marked by rapid decreases in estrogen [18], and declines in testosterone levels in males [22]. However, these findings have only been observed across a broad age spectrum, with limited investigation focusing on much older adults, specifically those aged 80 and above. While these observations imply a crucial involvement of hormones in cognitive performance, a gap exists in understanding whether they retain their influence in an older population.

Sex and gonadal hormones may influence cognitive abilities, maintenance, and decline with age. However, limited research has explored the interplay between sex, hormones, and cognitive performance, specifically in 80-plus healthy older adults. Consequently, the study aimed to retrospectively investigate the associations between sex, hormonal differences, and cognitive performance within older adults of 80-plus years. By elucidating these intricate relationships, this study aimed to provide valuable insights into the underlying mechanisms contributing to cognitive performance in older adults. Our primary hypothesis posited that sex is the predominant predictor of cognitive performance differences, with females exhibiting superior performance in the cognitive assessments. Moreover, we hypothesized that estrogen is a mediating factor to explain observed sex differences in cognitive abilities, regardless of the loss of estrogen during the menopausal period.

Subjects

A total of 244 participants, aged 79–93 years (M = 82.5, SD = 2.5) and born between 1926 and 1939, were recruited from the Sensor-based Systems for Early Detection of Dementia (SENDA) project [23]. Among them, 160 participants volunteered to provide saliva samples. However, 10 did not return their samples, 17 were cognitively impaired according to the MoCA cut-off criteria (MoCA < 23, [24], and for 2 all cognitive test scores were missing, resulting in a final sample of 131 older adults (60 males and 71 females, Table 1). Recruitment strategies and inclusion and exclusion criteria can be found in the published study protocol [23]. None of the individuals in the final sample received hormone replacement therapy at the time of the study, and no information was available about it prior to the study period. Participants provided written informed consent before participating in the study, which the Research Ethics Committee approved at the Faculty of Behavioral and Social Sciences at Chemnitz University of Technology, Germany (V-232-17-KM-SENDA-07112017). The trial was retrospectively registered at the German Clinical Trials Register (DRKS) with registration number DRKS00013167.

Table 1

Sample characteristics. Data is provided as mean and (standard deviation). Sex differences were test with Student’s t-Test or Welch's unequal variances t-Test and printed in bold when the sexes differed significantly.
	Male (n = 60)	Females (n = 71)	t-Test results
Age (years)	83.68 (2.46)	83.01 (2.30)	t(129) = 1.60, p = .113
MoCA	25.90 (1.85)	26.72 (2.04)	t(129) = -2.39, p = .019
MMSE	27.85 (1.33)	28.49 (1.17)	t(129) = -2.95, p = .004
CCI¹	1.50 (1.73)	1.39 (1.88)	t(114) = 0.33, p = .738
Education (years)	15.47 (3.63)	12.88 (2.47)	t(101.06) = 4.68, p < .001
Estrogen² (pg/mL)	5.61 (3.33)	5.66 (3.19)	t(128) = -0.10, p = .922
Testosterone³ (pg/mL)	65.24 (34.06)	40.53 (29.23)	t(126) = 4.42, p < .001
Note. MoCA = Montreal Cognitive Assessment., MMSE = Mini Mental State Examination, CCI = Charlson Comorbidity Index

¹ Due to missing data this was based on only 54 males and 62 females.

² Estrogen data from 1 female was removed from the analysis because of measurement error.

³ Testosterone data from 2 females and 1 male had to be removed because of measurement error.

Measures

For a detailed description of all measures in the SENDA project, please refer to [23]. Cognitive assessments were performed at approximately eight-month intervals, with the number of assessments ranging from one to four depending on the enrolment date of each participant. The assessment closest to the date of saliva sample collection was chosen for each participant to represent their current cognitive status in relation to their hormone levels (estrogen and testosterone). The interval between cognitive testing and saliva sampling was between 22.0 and 93.6 weeks (M = 38.9, SD = 16.0).

Cognitive assessment

During two measurement days, each participant underwent a battery of eleven established neuropsychological tests. This included seven subtests of the Consortium to Establish a Registry for Alzheimer's Disease Neuropsychological test battery (CERAD-NP) [26] : Boston Naming Test, Word List Learning, Word List Recognition, Word List Free Recall, Figure Drawing Task, Figure Recall Task, and the Trail Making Test. Furthermore, the Digit Symbol Substitution Test [27], the Serial Sevens Test [28], the Digit Span Forward Test [27] and the Flanker Test [29] were used. From these tests, 17 separate performance scores were obtained for each participant (Table 2). All tests were paper-pencil or oral tests carried out by trained project staff. Only the Flanker Test was a computerized assessment in which participants reacted as fast as possible via keypress to the color of the center target disk (green or red) while ignoring the opposite color (red or green) of the surrounding flanker disk (a detailed description of the same task is already available here [29]). In addition, education was recorded as the number of years spent in school and higher education according to CERAD-NP recommendations [26]. For the cognitive assessments, data was missing from 21 participants across the Trial Making Test and the Flanker Test. One participant abstained from the B condition of the Trial Making Test, and twenty participants were not present on the second measurement day where the Flanker Test was conducted.

Table 2

Test	Score description	Score name	Cognitive function	Domain
Digit Symbol Substitution Test	Number of correctly filled out symbols in 90 seconds	DSST	Updating	EF
CERAD Neuropsycho-logical Test Battery	Number of correctly identified words in the Boston Naming Test (0-12)	-	Visual naming	-
	Number for words recalled Trial 1 of the Word List Task (0-10)	WL1	Verbal memory	MEM
	Number for words recalled Trial 2 of the Word List Task (0-10)	WL2	Verbal memory	MEM
	Number for words recalled Trial 3 of the Word List Task (0-10)	WL3	Verbal memory	MEM
	Percent of correctly identified words in the Word List Recognition Task	Recog	Verbal memory	MEM
	Percent of correctly remembered words in the Word List Recall Task	Recall	Verbal memory	MEM
	Number of s-Words produced in one minute	Fluency.S	Verbal fluency	EF
	Number of animals produced in one minute	Fluency.A	Verbal fluency	EF
	Point score of the Figure Drawing Task (0-11)	-	Visuo-spatial ability	-
	Point score of the Figure Recall Task (0-11)	Fig.Recall	Visual Memory	MEM
	Trail Making Test A time	TMTA	Updating	EF
	Trail Making Test B time	TMTB	Shifting	EF
Serial Sevens Test	Number of correct answers given in 15 seconds	-	Updating	-
Digit Span Forward Test	Number of maximally remembered digits		Updating	-
Flanker Test	Average reaction time of incongruent trials	FL.RT	Inhibition	EF
	Accuracy of incongruent trials	FL.Acc	Inhibition	EF

Saliva hormone assessment

Saliva sample collection was conducted according to standardized guidelines. Participants were provided with saliva sample kits and instructions. On the day following receipt of the kits, participants were asked to provide unstimulated whole saliva samples upon awakening. To account for the pulsatile nature of hormone production and achieve a more precise concentration value, five saliva samples were provided over a period of one and a half hours, with approximately 20 minutes between each sample. Participants were instructed to refrain from consuming excessive amounts of dairy products, alcohol, and chocolate the evening before sample collection. On the day of sample collection, participants were asked to abstain from food and drink (except water), smoking, using toothpaste, and (when possible) medication intake until the sample had been collected. Any medication taken before or during the measurement time was documented and controlled to ensure no influence on hormone concentrations. Unstimulated whole saliva samples (~ 3 mL) were collected into sterile 5 mL plastic containers and sent to the research team within 48 hours of collection. The samples were immediately placed in a -20°C freezer until assayed to maintain sample integrity.

The estrogen and testosterone concentrations in saliva were determined using commercially available enzyme-linked immunosorbent assays (ELISAs) (DRG Instruments GmbH, Marburg, Germany). Estrogens commonly refer to a group of hormones. To maintain consistency with a significant portion of the literature, we will indicate our measurement of estradiol across the manuscript as estrogen. The assays were performed in singlet in accordance with the manufacturer's instructions using the BioTek ELISA Miroplatten-Wascher ELX-50 and BioTek ELISA Mikrotiterplatten-Reader EL-800 (BioTek, Vermont, USA). The intra-assay coefficients of variation for salivary estradiol and testosterone for the collected samples were 4.6% and 7.5%, respectively. For three participants, concentration values that fell outside the normal physiological range were recorded, which warranted consideration as measurement errors. The hormone values of two female subjects were excluded from the dataset due to their hormone levels deviating from the established reference ranges: one female exhibited anomalies in both estrogen and testosterone levels, while another female showed aberrations in testosterone levels. Additionally, the testosterone value of one male subject was omitted from the analysis as his testosterone levels exceeded the accepted normal reference values.

Statistical Analyses

Statistical analyses were performed in R using MVN package for assessing multivariate normality [30] and lavaan package for structural equation modelling [31]. All structural equation models were estimated using the ‘MLR’ estimator to control for non-normality of the data and using the full-information maximum likelihood option ‘ML’ to deal with missing data in the cognitive test battery and excluded hormone concentrations. First, confirmatory factor analysis (CFA) was performed to combine the cognitive test scores into the two aggregated cognitive factors, executive function and memory, following the results from [32]. Next, the effect of sex directly and mediated via hormones estrogen and testosterone on executive function and memory were estimated while controlling for education. Standardized coefficients (direct paths: β, indirect effects: ab), model fits and R² were reported for each model. Results were considered significant at alpha = .05.

After checking the data distribution of all cognitive scores for normality, the Boston Naming Test and the Figure Drawing Test were not included into the factor analysis because of strong ceiling effects. In addition, the Serial Sevens Test and the Digit Span Forward Test did not load significantly on the executive function factor and were removed from the CFA. The results of the CFA (Fig. 1) indicated acceptable model fit (χ2(63) = 94.24, p = .007; AIC = 9741.81; robust CFI = 0.93; robust RMSEA = .057, p (RMSEA ≤ .05) = .327, SRMR = .078). The two factors of the factor model were used as dependent variable in the structural equation model of the following analysis.

Figure 1 - Results of the confirmatory factor analysis of the cognitive performance scores

PLEASE INSERT FIGURE 1 HERE

Note

EF = Executive Functions, MEM = Memory, DSST = Digit Symbol Substitution Test, TMT = Trail Making Test, FL.RT = Flanker Test: average reaction time of the incongruent condition, FL.Acc = Flanker Test: accuracy in the incongruent condition, Fluency.S = Fluency of S-Words, Fluency.A = Fluency Animals, WWL1 = Wordlist Learning first Trial, WL = Wordlist Learning Trial 2, WL3 = Wordlist Learning Trial 3, Recog = Wordlist Recognition Test, Recall = Wordlist Recall Test, Fig.Recall = Figure Recall Test.

In the mediation model (χ2(108) = 182.80, p < .001; AIC = 11648.25; robust CFI = 0.87; robust RMSEA = .068, p (RMSEA ≤ .05) = .076, SRMR = .081) a direct effect of sex on executive function (β = .41, p < .001) and memory (β = .41, p < .001) was revealed after controlling for education (Fig. 2). No significant indirect effect of sex mediated through hormones on executive function (estrogen: ab = .00, p = .984, testosterone: ab = .05, p = .216) or memory (estrogen: ab = .00, p = .984, testosterone: ab = .05, p = .182) was presented. Looking at the partial paths in the model revealed that males had significantly higher testosterone levels (β = − .34, p < .001) while there was no difference in estrogen levels (β = .002, p = .984) between males and females. Estrogen was a significant predictor of executive function (β = .20, p = .01) but not of memory performance (β = .158, p = .091). Testosterone on the other hand was not significantly associated with the executive function factor (β = − .154, p = .180) or the memory factor (β = − .143, p = .118). Figure 3 includes scatter plots depicting these relationships between hormones and both cognitive performance factors. Education was included as covariate into the model and was associated significantly with the executive function outcome (β = .27, p = .004) but not with the memory factor (β = .10, p = .26). The included predictors (sex, estrogen, testosterone, and education) explained 24% of the variance of executive function and 23% of memory.

Figure 2 - SEM results for the mediation analysis

PLEASE INSERT FIGURE 2 HERE

Note

All coefficients are standardized. The SEM model also included education as additional predictor which were omitted from the display. EF = factor executive functions, MEM = factor memory,

Figure 3 - Residual scatterplot showing the relationship between a) executive function and estrogen, b) executive function and testosterone, c) memory performance and estrogen, and d) memory performance and testosterone after regressing out all other predictors.

PLEASE INSERT FIGURE 3 HERE

Note

EF (executive function) and MEM (memory) residual were obtained by regressing out all other predictors (sex, other sex hormone, education) of the factors in the model. Larger residuals indicate better performance.

The study’s primary objective was to investigate sex differences in cognitive performance and explore the potential association with current levels of gonadal hormones (estrogen and testosterone). Therefore, we utilized a factor analysis to identify cognitive domains based on the applied cognitive battery developed within the SENDA project [23]. The analysis yielded two distinct domains, namely executive function and memory. Consistent with prior research, we found sex differences in memory [3, 33], specifically, females exhibited a better performance compared to males. Furthermore, we found that females outperform males in executive function, a finding that does not consistently align with existing literature [4]. Although notable sex differences were observed in cognitive performance, no indirect effect of sex mediated through hormones was detected. Albeit estrogen emerged as a significant predictor of executive function. The analysis indicated that the observed differences between sexes for memory and executive function could only be partially attributed to the included variables (Table 2). These findings imply that estrogen may exert an influence on the development of these cognitive performance metrics. Nevertheless, it is apparent that multiple additional factors likely contribute to a greater extent in explaining these differences in 80 + year-olds.

Sex differences in cognitive performance vary from small to large effects, as documented in previous studies [6]. The Gender Similarities Hypothesis posits that males and females exhibit considerable similarity across most cognitive variables, though not all. In our investigation, we identified significant differences between sexes on two cognitive domains: executive function and memory. While literature often reports no differences in executive function between sexes [4], recent meta-analyses have highlighted specific executive function tasks that show sex disparities [4]. In our CFA, verbal fluency and inhibition (Table 2) were included and contributed to the executive function factor. Across studies, females consistently demonstrated better performance in verbal fluency compared to males [34], whilst Gaillard and colleagues [4] identify that response inhibition has emerged as an executive function domain where females exhibit moderately better performance than males. Despite literature suggesting overall similarity in EF between sexes, our findings support Gaillard and colleagues' [4] suggestion that sex differences are task-dependent. Specifically, tasks within the SENDA battery that contribute to the executive function domain align with those previously documented to favor female performance over males.

Our female participants' memory performance was greater than that of their male counterparts, consistent with previous research findings [3, 35]. This pattern also appears to be content-specific, indicating that the nature of the information to be remembered significantly influences recall ability. In previous studies, females have demonstrated particular proficiency in verbal memory tasks [3], while males perform better in spatial memory (e.g., remembering a route). Within our cognitive battery, the memory domain comprises six tests, of which five are verbal memory tasks (see Table 2). The observed sex differences in memory may be further accentuated by the specific tasks within this domain, contributing to the notable disparities between males and females.

It is well-established that cognition generally declines with age [1, 2]. However, research on sex differences in older populations, particularly those aged 80 and above, remains limited. Nonetheless, our findings are consistent with prior studies, indicating that healthy older adult females consistently demonstrate superior performance in tasks associated with executive function and memory across the lifespan, including advanced old age. For instance, de Frias et al. [36] examined episodic and semantic memory, and visuospatial ability in individuals aged 35 to 80 at baseline and found enduring sex differences over a ten-year follow-up period: Females exhibited superior performance in tasks assessing verbal episodic memory and verbal fluency, while males outperformed females in tasks involving visuospatial functions [36]. Additionally, studies by Maitland et al. [37], Pauls et al. [38], and Jockwitz et al. [10] further support the consistency of sex differences, particularly in the verbal versus spatial domains, among older individuals. While this study did not directly assess visuospatial domains, the derivation of the executive function domain through factor analysis, incorporating processing speed, inhibition, attentional control, and working memory (Table 2), provides a different perspective on multiple incorporated cognitive metrics. The current findings confirm that differences in cognitive abilities persist even during the later decades of life.

An additional aim of this study was to explore the influence of gonadal hormones on cognitive performance. It has long been suggested that endogenous sex hormones can impact cognitive abilities; for instance, endogenous estrogen has been linked to memory performance [39]. Our current findings revealed that levels of hormones estrogen and testosterone did not mediate the sex-related differences in executive function or memory. This observation may be attributed to the age group of the participants. As previously mentioned, certain cognitive tasks exhibit sex differences that persist throughout the lifespan. Notably, these differences are less pronounced in early and older adulthood compared to other life stages, as demonstrated in tasks involving verbal abilities [3]. Therefore, discrepancies in cognitive performance would be more evident during periods characterized by substantial differences in hormonal levels between males and females, such as adolescence and adulthood. However, the present study female cohort exclusively comprised postmenopausal females who were not undergoing hormonal replacement therapy. Consequently, the concentration levels of estrogen did not significantly differ between males and females, which may explain the lack of mediating effects.

While no mediating effects of hormones on cognitive performance were observed, estrogen emerged as a predictor of executive function performance. Estradiol, a form of estrogen, has consistently been implicated in influencing cognitive performance, as supported by numerous studies [17, 21, 40–43]. Previous research has highlighted an association between a longer reproductive window, indicative of increased exposure to estrogens, and better cognitive health in later life [44, 45]. In cellular and animal models, estrogens have demonstrated neuroprotective properties, stimulating the production of brain-derived neurotrophic factors and enhancing synaptic spine density in the hippocampus [46]. These findings suggest that exposure to estrogens may contribute to the preservation of neural health in late life, potentially explaining why estrogen was found to predict executive function in our population. However, concerning the direct effects on sex differences, our study suggests that factors beyond gonadal hormonal concentrations play a role in the persistence of these differences in cognitive abilities, underscoring the multifaceted nature of sex disparities in cognition.

Sex differences in cognitive performance may also be influenced by cohort disparities, which could stem from various biological factors such as health conditions or lifestyle choices related to psychosocial gender roles, as well as systematic differences in life course experiences. Regarding biological factors, previous research has indicated that some of the cognitive advantages observed in older females may be mediated by fewer cardiovascular risk factors, including factors such as smoking, waist-to-hip ratio, and cholesterol levels in females [47]. It cannot be disregarded that the sex differences uncovered in our analysis may be associated with systematic differences in health between males and females. According to the Charlson Comorbidity Index [48], there were no significant differences in comorbidities between males and females in our sample, suggesting comparable health conditions between the two groups. Concerning lifestyle factors, educational attainment is of particular interest as it systematically differs between sexes, especially among older birth cohorts as examined in our study [49], and is also positively correlated with cognitive performance in older adults [50, 51]. Our analysis indicated that the sex difference observed could not be explained by differences in educational level. In our sample, females, on average, had fewer years of education yet achieved equal or better cognitive performance scores compared to male participants, a pattern consistent with previous observations [52]. This suggests that rather than explaining the sex differences, educational attainment would have obscured the findings. Numerous other lifestyle factors related to psychosocial gender roles are speculated to contribute to the sex differences in cognitive performance observed in old age [43, 47, 53]. However, as our study solely focused on current gonadal hormone concentrations and did not intend to investigate other non-biological factors, the specific contribution of these non-biological lifestyle factors to sex differences in cognitive performance remains speculative. Nonetheless, it is reasonable to assume that they at least partially contribute to our findings.

In the current study, males were not found to perform better than females across the cognitive performance measures. Additionally, no associations were observed between testosterone levels and cognitive performance across executive function and memory. Androgens, including testosterone, have been linked to cognitive performance, particularly in spatial rotation and awareness tasks. Evidence suggests that prenatal androgens influence spatial performance in girls with congenital adrenal hyperplasia [54] and serve as predictors of spatial cognition in males [55]. Furthermore, in older males, lower testosterone concentrations have been associated with poorer performance on selected cognitive tests. Although a significant difference in testosterone levels between males and females was observed in our study, as depicted in Fig. 2A, this difference did not translate into sex differences benefitting males in executive function or memory performance. One potential explanation for these findings lies within the CFA results of the SENDA test battery used in our study, which did not include specific tests assessing visual-spatial functioning. While one test of visual memory (Table 2) was included in the executive function factor, research suggests that males may have an advantage in tasks requiring visual-spatial processing, such as those involving figures, shapes, and routes [5]. However, since this test was only one of six tests comprising the executive function domain, its contribution may not have been substantial enough to detect a better male performance in executive function. Consequently, our ability to discern male abilities in cognitive tasks was limited. Future research studies employing a similar design would benefit from exploring the relationship between cognitive tasks and testosterone, particularly tasks involving high demands of visual-spatial processing.

The present study exhibits both strengths and limitations. Notably, the utilization of confirmatory factor analysis enabled the development of comprehensive variables that capture a holistic perspective of cognition, thereby reflecting real-life cognitive processes rather than isolating specific aspects.

Across research, the age range of participants often varies widely, resulting in significant diversity in cognitive and hormonal states within the population [7, 10]. By exclusively enrolling older adults aged ≥ 80 years, the current study achieved a more homogeneous sample in terms of cognitive and hormonal characteristics compared to other studies. When investigating sex differences, it is crucial to consider the various factors that might contribute to their development.

Aside from hormonal influences, genetic and social factors may also contribute to the observed sex differences. Our study extended its investigation to explore potential hormonal effects on cognitive performance while controlling for other known factors that may affect sex differences, such as education level. However, participants' general health is also a significant contributor to cognitive performance. Although the Charlson Comorbidity Index was utilized, it may not provide a comprehensive assessment of participants' health, especially considering that females in this age group may generally exhibit better overall health.

While significant differences in cognitive performance favoring females were identified in our study, there may be limitations regarding the cognitive outcomes. We choose a two-factor model with one latent factor for executive functions and the other latent factor operationalizing memory based on findings from [32]. Reliable cognitive tests were used to derive the indicators of these latent factors. Surprisingly, the Serial Sevens Test and the Digit Span Forward test did not load highly on the executive function factor and were, therefore, removed from the analysis. Because of this, it may be speculated that, rather than assessing purely executive function as intended, the factor instead or additionally represent processing speed. Disentangling the relationship between processing speed and executive function is challenging, and it has been reported that a portion of age-related deficits in executive function performance may be attributed to a slowing in information processing [56]. In our model three of the executive function indicators (Digit Symbol Substitution Test, Trail Making Test A and B) are also frequently used as measures for processing speed [57, 58]. Hence, our executive function factor may not represent a pure measure of executive function but also processing speed, it remains a useful outcome for detecting sex differences in neurocognitive aging. In particular, the use of latent factors to study sex differences is superior to other commonly used methods, such as using z-scores to obtain composite scores, which is done without checking model fit [32], or analyzing individual cognitive tests, which does not provide a comprehensive model of cognition [59].

Furthermore, it is worth noting that the absence of specific cognitive tests tailored specifically for males (e.g., spatial visual tasks) may limit the identification of tests demonstrating better male performance. Consequently, the cognitive profiles of both sexes may not have been fully represented in the study.

Finally, hormonal concentrations were collected retrospectively and at the same time point for all participants, although discrepancies might exist between the time elapsed since their last cognitive assessments and their hormonal evaluations. However, this potential limitation was minimized by the study's recruitment strategy, which targeted a significant age group where substantial hormone fluctuations are not expected within the observed timeframes.

Collectively, our study makes a significant contribution to the existing literature base on sex differences in cognitive performance. The results robustly demonstrate the presence of sex differences in memory and task-specific executive function, with females consistently outperforming males. Notably, these differences persist into advanced adulthood, including individuals aged 80 years and above. Furthermore, our findings suggest that current gonadal hormone concentrations do not mediate these sex differences in older age groups. It is noteworthy that while estrogen predicts executive function performance, indicating its potential importance for cognitive development, this association may not extend into older age. These findings underscore the complexity of factors contributing to sex differences in cognitive abilities, warranting further research to unravel the underlying mechanisms, particularly among older adults, even at the advanced age of 80 years and beyond. Future investigations can thus focus on elucidating the interplay of various biological, environmental, and social factors in shaping cognitive performance disparities between sexes in older age groups.

Ethics approval and consent to participate

The local research ethics committee approved at the faculty of Behavioral and Social Sciences at Chemnitz University of Technology, approved the study protocols that were performed in accordance with the Declaration of Helsinki. Written informed consent was obtained from all patients involved in the study.

Consent for publication

Not applicable

Competing interests

The authors declare that they have no competing interests.

Funding

This work was supported by the European Social Fund for Germany and the Sächsische Aufbaubank-Förderbank (SAB) of the Free State of Saxony (Grant No. 100310502). This funding source had no role in the study design, in the collection, analysis, and interpretation of data, or in the writing of the report.

Authors' contributions

CVR developed the global research design of the SENDA project and provided overall project administration, funding acquisition, resources, and supervision. RJ, CVR, and KM contributed to the specific study design of the current research question. CVR, RJ, SF and KM curated data. SF performed statistical analyses. RJ and SF drafted the manuscript. CVR, KM and MD critically revised the manuscript. All authors read and approved the final manuscript.

Acknowledgements

The authors would like to thank all the participants for their contribution. A special thank you is also extended to all members of the SENDA team from the Chemnitz University of Technology for their support in subject recruitment, attainment, and data collection.

Availability of data and materials

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

1. Salthouse TA. When does age-related cognitive decline begin? Neurobiology of Aging. 2009;30(4):507 − 14. https://doi.org/10.1016/j.neurobiolaging.2008.09.023.
2. Whitley E, Deary IJ, Ritchie SJ, Batty GD, Kumari M, Benzeval M. Variations in cognitive abilities across the life course: Cross-sectional evidence from: The UK Household Longitudinal Study. Intelligence. 2016;59:39–50. https://doi.org/10.1016/j.intell.2016.07.001.
3. Asperholm M, Hogman N, Rafi J, Herlitz A. What did you do yesterday? A meta-analysis of sex differences in episodic memory. Psychol Bull. 2019;145(8):785–821. https://doi.org/10.1037/bul0000197.
4. Gaillard A, Fehring DJ, Rossell SL. A systematic review and meta-analysis of behavioural sex differences in executive control. Eur J Neurosci. 2021;53(2):519 − 42. https://doi.org/10.1111/ejn.14946.
5. Voyer D, Voyer SD, Saint-Aubin J. Sex differences in visual-spatial working memory: A meta-analysis. Psychon Bull Rev. 2017;24(2):307 − 34. https://doi.org/10.3758/s13423-016-1085-7.
6. Hyde JS. Gender Similarities and Differences. Annu Rev Psychol. 2014;65:373 − 98. https://doi.org/10.1146/annurev-psych-010213-115057.
7. Munro CA, Winicki JM, Schretlen DJ, Gower EW, Turano KA, Munoz B, et al. Sex differences in cognition in healthy elderly individuals. Neuropsychol Dev Cogn B Aging Neuropsychol Cogn. 2012;19(6):759 − 68. https://doi.org/10.1080/13825585.2012.690366.
8. Jorm AF, Anstey KJ, Christensen H, Rodgers B. Gender differences in cognitive abilities: The mediating role of health state and health habits. Intelligence. 2004;32(1):7–23. https://doi.org/10.1016/j.intell.2003.08.001.
9. van Hooren SAH, Valentijn AM, Bosma H, Ponds RWHM, van Boxtel MPJ, Jolles J. Cognitive Functioning in Healthy Older Adults Aged 64–81: A Cohort Study into the Effects of Age, Sex, and Education. Aging, Neuropsychology, and Cognition. 2007;14(1):40–54. https://doi.org/10.1080/138255890969483.
10. Jockwitz C, Wiersch L, Stumme J, Caspers S. Cognitive profiles in older males and females. Sci Rep. 2021;11(1):6524. https://doi.org/10.1038/s41598-021-84134-8.
11. Gur RC, Alsop D, Glahn D, Petty R, Swanson CL, Maldjian JA, et al. An fMRI Study of Sex Differences in Regional Activation to a Verbal and a Spatial Task. Brain and Language. 2000;74(2):157 − 70. /https://doi.org/10.1006/brln.2000.2325.
12. Ruigrok AN, Salimi-Khorshidi G, Lai MC, Baron-Cohen S, Lombardo MV, Tait RJ, et al. A meta-analysis of sex differences in human brain structure. Neurosci Biobehav Rev. 2014;39(100):34–50. https://doi.org/10.1016/j.neubiorev.2013.12.004.
13. Cowell PE, Allen LS, Zalatimo NS, Denenberg VH. A developmental study of sex and age interactions in the human corpus callosum. Developmental Brain Research. 1992;66(2):187 − 92. https://doi.org/10.1016/0165-3806(92)90079-C.
14. Lopez-Ojeda W, Hurley RA. Sexual Dimorphism in Brain Development: Influence on Affective Disorders. J Neuropsychiatry Clin Neurosci. 2021;33(2):A485-89. https://doi.org/10.1176/appi.neuropsych.20100269.
15. Ali SA, Begum T, Reza F. Hormonal Influences on Cognitive Function. 2018(1394-195X (Print)).
16. Auyeung B, Lombardo MV, Baron-Cohen S. Prenatal and postnatal hormone effects on the human brain and cognition. Pflügers Archiv - European Journal of Physiology. 2013;465(5):557 − 71. https://doi.org/10.1007/s00424-013-1268-2.
17. Hampson E. Sex differences in human brain and cognition: The influence of sex steroids in early and adult life. Behavioral endocrinology, 2nd ed. Cambridge, MA, US: MIT Press; 2002. p. 579–628.
18. Russell JK, Jones CK, Newhouse PA. The Role of Estrogen in Brain and Cognitive Aging. Neurotherapeutics. 2019;16(3):649 − 65. https://doi.org/10.1007/s13311-019-00766-9.
19. Silverman I, Kastuk D, Choi J, Phillips K. Testosterone levels and spatial ability in men. Psychoneuroendocrinology. 1999;24(8):813 − 22. https://doi.org/10.1016/s0306-4530(99)00031-1.
20. Thilers PP, MacDonald SWS, Herlitz A. The association between endogenous free testosterone and cognitive performance: A population-based study in 35 to 90 year-old men and women. Psychoneuroendocrinology. 2006;31(5):565 − 76. https://doi.org/10.1016/j.psyneuen.2005.12.005.
21. Sherwin BB. Estrogen and cognitive functioning in women. Endocr Rev. 2003;24(2):133 − 51. https://doi.org/10.1210/er.2001-0016.
22. McBride JA, Carson CC, 3rd, Coward RM. Testosterone deficiency in the aging male. Ther Adv Urol. 2016;8(1):47–60. https://doi.org/10.1177/1756287215612961.
23. Müller K, Fröhlich S, Germano AMC, Kondragunta J, Agoitia Hurtado M, Rudisch J, et al. Sensor-based systems for early detection of dementia (SENDA): a study protocol for a prospective cohort sequential study. BMC Neurology. 2020;20(1):84. https://doi.org/10.1186/s12883-020-01666-8.
24. Carson N, Leach L, Murphy KJ. A re-examination of Montreal Cognitive Assessment (MoCA) cutoff scores. Int J Geriatr Psychiatry. 2018;33(2):379 − 88. https://doi.org/10.1002/gps.4756.
25. Nasreddine ZS, Phillips NA, Bedirian V, Charbonneau S, Whitehead V, Collin I, et al. The Montreal Cognitive Assessment, MoCA: a brief screening tool for mild cognitive impairment. J Am Geriatr Soc. 2005;53(4):695-9. https://doi.org/10.1111/j.1532-5415.2005.53221.x.
26. Basel MC. Manual zum Auswertungsprogramm CERAD-Plus. Memory Clinic Basel, Switzerland; 2005.
27. Wechsler D. The psychometric tradition: developing the wechsler adult intelligence scale. Contemporary Educational Psychology. 1981.
28. Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”: A practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research. 1975;12(3):189 − 98. https://doi.org/10.1016/0022-3956(75)90026-6.
29. Fröhlich S, Kutz DF, Müller K, Voelcker-Rehage C. Cardiorespiratory fitness is associated with cognitive performance in 80+-year-olds: Detangling processing levels. Geroscience. 2024. https://doi.org/10.1007/s11357-024-01065-8.
30. Korkmaz S, Goksuluk D, Zararsiz G. MVN: An R Package for Assessing Multivariate Normality. R J. 2014;6(2):151 − 62. https://doi.org/DOI 10.32614/rj-2014-031.
31. Rosseel Y. lavaan: An R Package for Structural Equation Modeling. Journal of Statistical Software. 2012;48(2):1–36. https://doi.org/10.18637/jss.v048.i02.
32. Wagner M, Wolfsgruber S, Gaertner B, Kleineidam L, Buttery AK, Jacobi F, et al. Cognitive functioning in the general population: Factor structure and association with mental disorders-The neuropsychological test battery of the mental health module of the German Health Interview and Examination Survey for Adults (DEGS1-MH). Int J Methods Psychiatr Res. 2018;27(1). https://doi.org/10.1002/mpr.1594.
33. Loprinzi PD, Frith E. The Role of Sex in Memory Function: Considerations and Recommendations in the Context of Exercise. J Clin Med. 2018;7(6). https://doi.org/10.3390/jcm7060132.
34. Hirnstein M, Stuebs J, Moe A, Hausmann M. Sex/Gender Differences in Verbal Fluency and Verbal-Episodic Memory: A Meta-Analysis. Perspect Psychol Sci. 2023;18(1):67–90. https://doi.org/10.1177/17456916221082116.
35. Sundermann EE, Maki PM, Rubin LH, Lipton RB, Landau S, Biegon A, et al. Female advantage in verbal memory: Evidence of sex-specific cognitive reserve. Neurology. 2016;87(18):1916-24. https://doi.org/10.1212/WNL.0000000000003288.
36. de Frias CM, Nilsson LG, Herlitz A. Sex differences in cognition are stable over a 10-year period in adulthood and old age. Neuropsychol Dev Cogn B Aging Neuropsychol Cogn. 2006;13(3–4):574 − 87. https://doi.org/10.1080/13825580600678418.
37. Maitland SB, Herlitz A, Nyberg L, Backman L, Nilsson LG. Selective sex differences in declarative memory. Mem Cognit. 2004;32(7):1160-9. https://doi.org/10.3758/bf03196889.
38. Pauls F, Petermann F, Lepach AC. Gender differences in episodic memory and visual working memory including the effects of age. Memory. 2013;21(7):857 − 74. https://doi.org/10.1080/09658211.2013.765892.
39. Hara Y, Waters EM, McEwen BS, Morrison JH. Estrogen Effects on Cognitive and Synaptic Health Over the Lifecourse. Physiol Rev. 2015;95(3):785–807. https://doi.org/10.1152/physrev.00036.2014.
40. Halari R, Hines M, Kumari V, Mehrotra R, Wheeler M, Ng V, et al. Sex Differences and Individual Differences in Cognitive Performance and Their Relationship to Endogenous Gonadal Hormones and Gonadotropins. Behavioral Neuroscience. 2005;119(1):104 − 17. https://doi.org/10.1037/0735-7044.119.1.104.
41. Gargiulo AT, Hu J, Ravaglia IC, Hawks A, Li X, Sweasy K, et al. Sex differences in cognitive flexibility are driven by the estrous cycle and stress-dependent. Front Behav Neurosci. 2022;16:958301. https://doi.org/10.3389/fnbeh.2022.958301.
42. McEwen BS, Milner TA. Understanding the Broad Influence of Sex Hormones and Sex Differences in the Brain. J Neurosci Res. 2017;95(1–2):24–39. https://doi.org/10.1002/jnr.23809.
43. Gurvich C, Hoy K, Thomas N, Kulkarni J. Sex Differences and the Influence of Sex Hormones on Cognition through Adulthood and the Aging Process. Brain Sci. 2018;8(9). https://doi.org/ARTN 16310.3390/brainsci8090163.
44. Ryan J, Carriere I, Scali J, Ritchie K, Ancelin ML. Life-time estrogen exposure and cognitive functioning in later life. Psychoneuroendocrinology. 2009;34(2):287 − 98. https://doi.org/10.1016/j.psyneuen.2008.09.008.
45. McLay RN, Maki PM, Lyketsos CG. Nulliparity and late menopause are associated with decreased cognitive decline. J Neuropsychiatry Clin Neurosci. 2003;15(2):161-7. https://doi.org/10.1176/jnp.15.2.161.
46. Luine V, Frankfurt M. Interactions between Estradiol, Bdnf and Dendritic Spines in Promoting Memory. Neuroscience. 2013;239:34–45. https://doi.org/10.1016/j.neuroscience.2012.10.019.
47. van Zutphen EM, Rijnhart JJM, Rhebergen D, Muller M, Huisman M, Beekman A, et al. Do Cardiovascular Risk Factors and Cardiovascular Disease Explain Sex Differences in Cognitive Functioning in Old Age? Journal of Alzheimers Disease. 2021;80(4):1643-55. https://doi.org/10.3233/Jad-201173.
48. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40(5):373 − 83. https://doi.org/10.1016/0021-9681(87)90171-8.
49. Hasselgren C, Ekbrand H, Hallerod B, Mellqvist Fassberg M, Zettergren A, Johansson L, et al. Sex differences in dementia: on the potentially mediating effects of educational attainment and experiences of psychological distress. BMC Psychiatry. 2020;20(1):434. https://doi.org/10.1186/s12888-020-02820-9.
50. Liu Y, Lachman ME. Education and Cognition in Middle Age and Later Life: The Mediating Role of Physical and Cognitive Activity. J Gerontol B Psychol Sci Soc Sci. 2020;75(7):e93-e104. https://doi.org/10.1093/geronb/gbz020.
51. Lovden M, Fratiglioni L, Glymour MM, Lindenberger U, Tucker-Drob EM. Education and Cognitive Functioning Across the Life Span. Psychol Sci Public Interest. 2020;21(1):6–41. https://doi.org/10.1177/1529100620920576.
52. Gerstorf D, Herlitz A, Smith J. Stability of sex differences in cognition in advanced old age: the role of education and attrition. J Gerontol B Psychol Sci Soc Sci. 2006;61(4):P245-9. https://doi.org/10.1093/geronb/61.4.p245.
53. Lee BH, Richard JE, de Leon RG, Yagi S, Galea LAM. Sex Differences in Cognition Across Aging. In: Gibson C, Galea LAM, editors. Sex Differences in Brain Function and Dysfunction. Cham: Springer International Publishing; 2023. p. 235 − 84.
54. Puts DA, McDaniel MA, Jordan CL, Breedlove SM. Spatial ability and prenatal androgens: meta-analyses of congenital adrenal hyperplasia and digit ratio (2D:4D) studies. Arch Sex Behav. 2008;37(1):100 − 11. https://doi.org/10.1007/s10508-007-9271-3.
55. Vuoksimaa E, Kaprio J, Eriksson CJP, Rose RJ. Pubertal testosterone predicts mental rotation performance of young adult males. Psychoneuroendocrinology. 2012;37(11):1791 − 800. https://doi.org/10.1016/j.psyneuen.2012.03.013.
56. Albinet CT, Boucard G, Bouquet CA, Audiffren M. Processing speed and executive functions in cognitive aging: How to disentangle their mutual relationship? Brain Cognition. 2012;79(1):1–11. https://doi.org/10.1016/j.bandc.2012.02.001.
57. Salthouse TA. What cognitive abilities are involved in trail-making performance? Intelligence. 2011;39(4):222 − 32. https://doi.org/10.1016/j.intell.2011.03.001.
58. Jaeger J. Digit Symbol Substitution Test: The Case for Sensitivity Over Specificity in Neuropsychological Testing. J Clin Psychopharm. 2018;38(5):513-9. https://doi.org/10.1097/Jcp.0000000000000941.
59. Mitchell MB, Shaughnessy LW, Shirk SD, Yang FM, Atri A. Neuropsychological Test Performance and Cognitive Reserve in Healthy Aging and the Alzheimer's Disease Spectrum: A Theoretically Driven Factor Analysis. J Int Neuropsych Soc. 2012;18(6):1071-80. https://doi.org/10.1017/S1355617712000859.

The authors declare no competing interests.

Download PDF

Version 1

posted

You are reading this latest preprint version

Sex Differences in Cognitive Performance Persist into Your 80s

Status:

Version 1

Abstract

Figures

BACKGROUND

METHODS

Subjects

Measures

Cognitive assessment

Saliva hormone assessment

Statistical Analyses

RESULTS

DISCUSSION

CONCLUSION

Declarations

Competing interests

Funding

Authors' contributions

Acknowledgements

Availability of data and materials

References

Additional Declarations

Status:

Version 1