1. Introduction
In the literature, recent advances in cognitive abilities studies in relation to well-being, social interactions, and sport activities have been observed; in particular, some research focused on the role of chemosensory perception [
1,
2]. In fact, olfactory function could act on cognitive abilities as related to semantic memory, executive function, and verbal fluency [
3,
4,
5]. Moreover, a low language ability has been associated with low olfactory function [
3,
4,
5]. Olfactory function plays an important role in the regulation of emotional function, social interactions, and eating behavior. The decreased olfactory function may show negative implications for human well-being and social relationships.
Olfactory function and cognitive performance showed an overlap in neural substrates and were associated with the orbitofrontal cortex, hippocampus, and entorhinal cortex [
3,
4,
5]. This relationship suggested the need to evaluate olfactory function as a potential early biomarker in patients’ cognitive impairment [
6,
7,
8,
9]. Recently, the importance of olfactory assessment has significantly increased, particularly in cases of mild cognitive decline and in the diagnosis of COVID-19.
The objective evaluation of quantitative olfactory dysfunction comprises three different components: odor threshold (OT), odor discrimination (OD), and odor identification (OI) [
10]. The OT is usually more associated with individual differences in nasal anatomy [
11], and it is related to a specific cognitive domain, such as language function [
6]. Indeed, OD and OI are considered in relation to the subcortical part of the olfactory system and are related to the ability to differentiate and identify different odors, respectively [
12]. Usually, lower scores in cognitive tests associated with a deficit in olfactory function may suggest a mild cognitive impairment [
13]. In this context, the brain’s cognitive reserve is closely related to cortical plasticity and is considered the potential capability of the brain to cope with neuronal damage in relation to individual differences such as brain size and synapse count. Cognitive reserve plays an important role in order to recover brain damage affected by aging or neurodegenerative diseases through the recruitment of pre-existing brain networks. A high score of CRI indicates better neuronal plasticity to compensate for brain atrophy and is considered a protective factor, while a lower score indicates brain vulnerability. In general, setting aside neurocognitive resources during the lifetime may preserve the brain from mild cognitive impairment and pathologies such as Alzheimer’s disease [
14,
15]. Since around 65% of subjects affected by Alzheimer’s disease are women [
16], understanding gender-related differences in the relationship between olfactory function and the specific cognitive domain of the CRI questionnaire appears important in order to obtain the most incisive interventions for an early diagnosis of mild cognitive impairment and Alzheimer’s disease. Since the olfactory deficit is considered an important health issue, which may play a significant role in human social interactions and well-being, in our study, we hypothesized the occurrence of correlations between olfactory function and CRI in healthy subjects in relation to sex. In the literature, few studies have evaluated the gender differences in CRI [
17,
18], but the gender-related association between olfactory function and each specific cognitive domain of the CRI questionnaire has so far not been evaluated.
The aim of this study was first to evaluate gender-related differences in each specific cognitive domain of the CRI questionnaire such as education, working activity, and leisure time in healthy subjects. Then, we evaluated gender-related differences in the relationship between olfactory function and each specific cognitive domain of the CRI.
3. Results
The demographic and clinical information of all participants was indicated in
Table 1. Significant differences between men and women were observed for height, weight, OT, OI, TDI score, cognitive abilities (MoCA scores), and CRI-Working Activity. In particular, women exhibited higher mean values in OT [F
(1,267) = 6.031,
p < 0.05, partial η
2 = 0.022] and TDI score [F
(1,267) = 5.774,
p < 0.05, partial η
2 = 0.021], and lower significant scores in CRI-Working Activity (
Table 1). No significant differences (
p > 0.05) between both sexes for OD, OI, MoCA global score, CRI-Education, CRI-Leisure Time, and CRI-Total Score were found.
Considering all subjects, significant bivariate correlations were observed between OT and CRI-Education (r = −0.150,
p = 0.014), while OD was significantly correlated to CRI-Working Activity (r = 0.401,
p = 0.001), CRI-Leisure Time (r = 0.364,
p = 0.001), and CRI-Total Score (r = 0.197,
p = 0.001) (
Table 2). Moreover, significant bivariate correlations were detected between OI and different CRI sub-domains such as CRI-Education (r = 0.176,
p = 0.004), CRI-Working Activity (r = 0.406,
p = 0.001), CRI-Leisure Time (r = 0.389,
p = 0.001), and CRI-Total Score (r = 0.337,
p = 0.001) (
Table 2). The TDI score was significantly correlated to CRI-Working Activity (r = 0.308,
p = 0.001), CRI-Leisure Time (r = 0.304,
p = 0.001), and CRI-Total Score (r = 0.297,
p = 0.001).
In addition, in order to better clarify the role of bivariate correlations, multivariate linear regression analyses were performed to predict olfactory dysfunction in relation to CRI-Education, CRI-Working Activity, and CRI-Leisure Time. The multivariate linear regression analyses showed that CRI-Education was a significant predictor when using OT as a dependent variable [F
(4,264) = 10.724,
p = 0.002] (
Table 3). The model with OT as a dependent variable explained 4% of the variance (R
2 = 0.041) (
Figure 1A). In the second model, CRI-Working Activity and CRI-Leisure Time were significant predictors for the OD [F
(4,264) = 16.790,
p = 0.0001] and this model explained around the 20% of the variance (R
2 = 0.203) (
Figure 1B,C). Similarly, in the third model, significant associations were observed between CRI-Working Activity and CRI-Leisure Time and the OI [F
(4,264) = 16.986,
p = 0.0001] and this model explained around 19% of the variance (R
2 = 0.193) (
Figure 1D,E).
In women, significant correlations were found between CRI-Working Activity and OD (r = 0.313,
p < 0.001), and OI (r = 0.426,
p < 0.001), and TDI score (r = 0.355,
p < 0.001). Moreover, we also observed significant correlations between the CRI-Leisure Time and OT (r = 0.278,
p < 0.001), and OD (r = 0.398,
p < 0.001), and OI (r = 0.546,
p < 0.001), and TDI score (r = 0.526,
p < 0.001) (
Table 4). The CRI-Education was significantly correlated only with the OI (r = 0.184,
p = 0.020) and TDI score (r = 0.333,
p < 0.001). Finally, the CRI-Total score was correlated with OD (r = 0.325,
p < 0.001), OI (r = 0.337,
p < 0.001), and TDI score (r = 0.298,
p < 0.001) (
Table 4).
In women, the multivariate linear regression analyses showed that in model 1, obtained using OT as a dependent variable, CRI-Leisure Time was significantly associated with OT [F
(4,153) = 3.514,
p = 0.002] (
Table 5), and the model explained 8% of the variance (R
2 = 0.084 (
Figure 2A). Similarly, in model 2, using OD as a dependent variable, a significant association was found between OD and CRI-Leisure Time [F
(4,153) = 8.396,
p = 0.004] with the 18% of the variance (R
2 = 0.180) (
Figure 2B). While, in model 3, performed using OI as a dependent variable, two different significant associations were observed between CRI-Working Activity [F
(4,153) = 18.450,
p = 0.001] and CRI-Leisure Time [F
(4,153) = 18.450,
p = 0.018] and OI (
Figure 2C,D), with around 33% variance (R
2 = 0.325) (
Table 5).
In men, the following significant correlations were found: between OT and CRI-Education (r = −0.346,
p = 0.001), between OI and CRI-Education (r = 0.194,
p = 0.042), between OD and CRI-Working Activity (r = 0.379,
p = 0.001), between OI and CRI-Working Activity (r = 0.342,
p = 0.001), between TDI score and CRI-Working Activity (r = 0.290,
p = 0.002), between OD and CRI-Leisure Time (r = 0.333,
p = 0.001), between OI and CRI-Leisure Time (r = 0.334,
p = 0.001), between TDI score and CRI-Leisure Time (r = 0.205,
p = 0.031), and between OI and CRI-Total Score (r = 0.387,
p = 0.001) (
Table 6).
Moreover, in men, multivariate linear regression analyses showed a significant association only between OT and CRI-Education [F
(4,106) = 4.440,
p = 0.002] with the model explaining around 14% of the variance (R
2 = 0.143) (
Figure 3). Instead, no significant associations (
p > 0.05) were found between OD and OI and each specific sub-score of CRI (
Table 7).
4. Discussion
This study focused on the evaluation of gender-related differences in the association between olfactory function and cognitive reserve index. The brain cognitive reserve is closely related to cortical plasticity and is considered the potential capability of the brain to cope with neuronal damage in relation to individual differences such as brain size and synapse count. Cognitive reserve is important in order to recover brain damage affected by aging or neurodegenerative diseases through the recruitment of pre-existing brain networks [
27]. Robertson indicated that the right hemisphere plays an important role in cognitive reserve using a noradrenergic pathway [
28].
Our results showed statistical differences between men and women for the OT, global olfactory function (TDI score), and CRI-Working Activity. According to previous studies, in women, higher mean values in OT and TDI scores were found compared to men [
29,
30]. Moreover, in our study, women showed significantly decreased scores in CRI-Working Activity compared to men, while no significant differences were observed for CRI-Education and CRI-Leisure Time. This result may be explained as a difference in the employment and retirement age between the two sexes, as reported by Boots and Colleagues [
31]. In fact, some studies reported that women with healthy working conditions (e.g., crafts worker, shopkeeper, and farmer) may reduce the risk of mild cognitive impairment and dementia [
18,
32].
Considering all subjects, our data showed a noteworthy association between olfactory function and CRI. In particular, we found significant associations between the OT and CRI-Education and between the OD and OI and CRI-Working Activity and CRI-Leisure Time. These data suggested and highlighted the close association between olfactory function and cognitive abilities. Subjects with lower scores in olfactory function usually exhibit weaker cognitive performance [
2]. In addition, higher olfactory scores are usually associated with better semantic memory and verbal abilities [
33]. Craick and colleagues [
34] showed that, both in women and men, Alzheimer’s disease symptoms appeared five years later in bilinguals than in monolinguals. Another study indicated that cognitive ability and vocabulary were associated with OI [
35]. A possible explanation of these data is due to partial overlapping in the brain areas involved in cognitive abilities and those involved in olfactory function such as the orbitofrontal cortex and amygdala. Our data also suggested a significant positive correlation between OD and OI scores and CRI-Leisure Time and are similar to those obtained in a previous study [
36], suggesting a relationship between OI and social life. On the other hand, the relationship between OI and OD and CRI-Leisure Time is not clearly understood. In the multivariate linear regression analyses, we found that there was a positive significant association between OI and OD and CRI-Leisure Time only in women, but not in men. These data support the hypothesis that CRI-Leisure Time is closely correlated to OI and OD performance only in women. Generally, women showed better olfactory performance compared to men [
10,
37] and also had more social connections. It is likely that women perform differently in social relationships and there could possibly be an association between leisure time, social networks, and health measures. In fact, Codina and Pestana showed that men had more leisure time, but women had a higher positive leisure experience than men [
38]; women enjoyed themselves more with less leisure time and were more positive about time orientations. Moreover, Larsson and colleagues observed that women identified more odors than men due to gonadal hormones, fluctuations of the menstrual cycle, and neuroendocrine influences on brain regions involved in olfactory function, but sex differences disappeared in older age [
35]. Although the potential cause of the difference between men and women remains unclear, the higher identification in women may be due to sex differences in verbal abilities, prior experience, and odor memory [
39]. However, social factors may also contribute as women generally experience greater olfactory pleasantness, odor familiarity, and greater exposure to odors in their social environment. A better performance in OD and OI is considered a measure of general good health in the population. Indeed, good health is often connected with social lives and the number of social contacts that the individuals have in their life.
The association between OT and CRI-Education was observed only in men and not in women. Our data suggested that the CRI-Education sub-score may have a minor contribution in women, as indicated in a previous study [
17]. Instead, Heian and colleagues showed that men with low education had lower olfactory function scores after a comparison between self-reported tests and Sniffin’ sticks data analysis [
40].
Our data suggested that in men, CRI-Education and not CRI-Working activity was associated with the odor threshold. Moreover, both in men and in women, working activity had no relation with the total olfactory function. Education probably has a protective role in mild cognitive decline, as previously indicated by Meng and D’Arcy [
41]. In fact, people with a high level of education correctly identified more odors, as indicated by Larsson and colleagues [
35].
Considering these results, our data suggested a potential role of biomarkers for olfactory function in the early diagnosis of mild cognitive impairment. Similarly, other previous studies indicated that olfactory impairment represents a peculiar biomarker in neurodegenerative disorders [
6,
7,
8,
42,
43,
44,
45,
46]. Recently, there has been an increased interest in the evaluation of olfactory dysfunction in the early stage of neurodegenerative disorders such as Parkinson’s disease. In our previous study on Parkinson’s disease, patients’ significant correlations were observed between OT and language, between OD and visuospatial domain, and between OI and executive index scores and attention [
6], suggesting that the OT, OD, and OI are differently related to the cognitive abilities of the subjects.
One limitation of our study is the cross-sectional design, thereby it did not allow us to evaluate these associations over time.