Gender Differences in Memory and Cognition
KeywordsGender Difference Episodic Memory Mental Rotation Verbal Ability General Cognitive Ability
Age and gender are variables that have been found to influence cognitive performance. In the following entry, findings of gender differences in cognitive functions will be reviewed, with a specific focus on examining to what extent the pattern and magnitude of these differences change over the adult life span. First, results on gender differences found in memory and cognition tasks in childhood and young adulthood are presented. This is followed by a discussion of the extent to which these differences change over the life span and/or over geographical regions. Lastly, somewhat contradictory findings reported in cross-sectional and longitudinal studies will be highlighted.
In this review, the following main conclusions are drawn: There is little evidence to suggest that the rate of decline in cognitive performance over the life span is different for men and women. However, cross-sectional studies, comparing individuals of different ages, suggest that the magnitude of gender differences may change over time. Improvements in living conditions and better educational opportunities are factors that may lead to increased gender differences favoring women in some cognitive functions and decreased or eliminated differences in other cognitive abilities. These changes in gender differences seem to take place due to a general increase in cognitive performance over time, associated with societal improvements, where women improve more than men.
Gender Differences in Cognitive Abilities
Not all cognitive tasks give rise to differences in test scores between men and women, but some do and have consistently been shown to do so. The focus in this entry will be on tasks assessing visuospatial, verbal, numerical, and episodic memory abilities – cognitive abilities that typically yield differences between men and women.
Visuospatial ability. Most research indicates that men perform at a higher level than women when it comes to visuospatial ability. The most recent meta-analysis on the topic (Voyer et al. 1995) concluded that men perform at a substantially higher level than women on most visuospatial tasks, with the overall gender difference in visuospatial ability being d = −.37 (effect size, d = (Mwomen − Mmen) / Sdtotal). Although men outperform women, the size of the difference may vary depending on the task being assessed. The largest difference is found in mental rotations (d = −.56; the ability to rapidly rotate two or three dimensional figures in mind), while it is somewhat smaller for spatial perception (d = −.44; the ability to determine a spatial relation regardless of distracting information) and considerably smaller in spatial visualization (d = −.19; the ability to manipulate complex spatial information when several steps are required to arrive at the correct solution).
The differences in visuospatial ability have also been examined in children. For example, studies conducted on 3- to 4-month-old (Quinn and Liben 2008) and 5-month-old babies (Moore and Johnson 2008) have shown that male infants are able to differentiate between rotated (compared to the orientation it had during the encoding phase) familiar figures and novel ones. In both of these studies, male infants displayed a novelty preference for the new figure, indicating that they recognized the familiar figure and therefore spent less time viewing it. This was not found for the female infants, who divided their attention between the two items equally. Others (Levine et al. 1999) have demonstrated that 4.5-year-old boys display the same kind of male advantage when it comes to spatial visualization and mental rotation tasks. The edge that boys have over girls at this age also persists into adolescence, as demonstrated by Herlitz and colleagues (2013) who found that these differences also exist around puberty and that the magnitude of the difference is similar across the examined age groups (12–14 years) (Herlitz et al. 2013).
Verbal ability. When it comes to verbal ability, the prevailing opinion has been that women outperform men. However, the picture is slightly more complicated. A meta-analysis on the existence of gender differences in verbal abilities (Hyde and Linn 1988) showed an overall modest advantage for women over men (d = .11). However, different tasks yielded different effect sizes, with the largest differences found in speech production tasks (d = .33). In anagram solving, women’s advantage was smaller (d = .22), in vocabulary it was nonexistent (d = .02), and in verbal analogies men had a slight advantage over women (d = −.16). Further, in verbal production tasks, such as category fluency, an advantage favoring women is most often found (Maylor et al. 2007), but depending on the topic that the participants are asked to generate words from, no gender differences are sometimes found (Weber et al. 2014). Similarly to visuospatial abilities, studies examining the development of verbal abilities, specifically concerning verbal production tasks, find gender differences favoring girls as young as 5 years of age (Hyde and Linn 1988). In line with this, another study (Herlitz et al. 2013) has found that the magnitude of the gender differences is similar for 12- to 14-year-olds.
Episodic memory. Episodic memory refers to the conscious recollection of unique personal experiences in terms of their content (what), location (where), and temporal occurrence (when). It is typically assessed by first presenting some information (e.g., episodes, words, objects, or faces) and then asking the person to recall or recognize the earlier-presented material.
Although the first comprehensive review of gender differences in cognition did not find any differences between men and women when it comes to memory (Maccoby and Jacklin 1974), many more recent studies have found gender differences favoring women in episodic memory tasks (see Herlitz and Rehnman 2008 for an overview). Women consistently outperform men on tasks that require remembering items that are verbal in nature or can be verbally labeled. However, women also excel on tasks requiring little or no verbal processing, such as recognition of unfamiliar odors or faces. In contrast, there is a male advantage on episodic memory tasks requiring visuospatial processing. Thus, the pattern of gender differences in episodic memory mirrors the pattern seen in verbal and visuospatial tasks, with the notable caveat that gender differences favoring women are also found in tasks requiring little or no verbal or visuospatial processing (Lowe et al. 2003). Studies of episodic memory function in children have found the same patterns as in adults, with girls having a slight overall advantage compared to boys as well as girls being better at verbal memory tasks and boys being better at visuospatial memory tasks (e.g., Lowe et al. 2003). Also, the magnitude of the difference in adolescents is similar to the difference found in adults, shown for example in a study examining memory in 12- to 14-year-olds (Herlitz et al. 2013).
Mathematics. Mathematics is an umbrella term that includes several different cognitive abilities concerning quantities, space, and numbers. Performing mathematical tasks therefore involves the recruitment of several cognitive abilities, and as a result, gender differences in mathematics often vary as a function of type of task. Although gender differences exist in school grades in mathematics, with girls having a small advantage over boys (d = .07) (Voyer and Voyer 2014), this edge is typically not present on tasks assessing mathematical ability. A meta-analysis taking type of task into consideration showed that boys and men have an advantage over girls in general tasks (d = −.15), with differences being larger in cognitively more demanding mathematical tasks such as mathematical complex problem solving. Further, the differences are also more prominent in samples performing in the upper percentiles of the distribution (Hyde et al. 1990). Interestingly, boys have a larger advantage over girls in later school years as compared to earlier (Hyde et al. 1990), indicating that the magnitude of gender difference may increase throughout childhood and adolescence.
Taken together, gender differences exist in some cognitive tasks, with girls and women outperforming boys and men in some of them, whereas the reverse is true in others. With the possible exception of mathematical ability, these differences seem to be present already in childhood and are preserved through adolescence and young adulthood without any change in magnitude.
Gender, Age Decline, and Age Differences
Are there any reasons to expect that men and women decline cognitively at different rates in old age? Biologically, there are some factors that would suggest age-related variation in the magnitude of gender differences. One such factor is accelerated brain aging, or brain atrophy, with some reports suggesting that men show more brain atrophy than women (Raz et al. 2004). Such gender differences in age-related brain atrophy may lead to greater gender differences with increasing age on tasks in which women excel, and smaller differences on tasks in which men perform at a higher level than women. On the other hand, the positive health selection of men with advanced age (i.e., the men who survive into old age may be healthier than the average), together with an increased risk of women being affected by Alzheimer’s disease, may lead to a minimization of gender differences in which women excel.
When investigating differences related to age, it is important to differentiate between longitudinal and cross-sectional assessments of gender differences across the life span. With longitudinal studies, it is possible to examine to what extent the same individuals or groups of individuals deteriorate with increasing age, whereas the cross-sectional assessment will show to what degree different age groups, measured at the same time point, vary with regard to performance. Naturally, age groups assessed at the same time differ not only with regard to age, but also with regard to the environment they have been exposed to. For example, later-born generations have typically grown up in societies in which individuals have received more years of education, better nutrition and health care, and more complex and stimulating environments. However, the exposure to these societal improvements may also vary between men and women, with women in many societies and age groups receiving less favorable exposure (Else-Quest and Grabe 2012).
Longitudinal assessment. There have been several longitudinal studies looking at gender differences in cognitive decline. One of these studies investigated cognition over a 10-year period in a population-based sample (De Frias et al. 2006), with over 600 participants (initially 35–80 years old). Regardless of initial age, it was shown that gender differences remained stable over the examined 10-year period: women performed at a higher level than men on some cognitive tasks (episodic memory, verbal production), whereas men performed at a higher level on a task assessing visuospatial ability. In line with this, results from the Berlin Aging Study, with participants aged between 70 and 100 years of age, showed that men and women declined virtually in parallel, with no evidence of any differences in decay over the 13-year period (Gerstorf et al. 2006). Further, a recent meta-analysis, where altogether 13 studies were included, came to the same conclusion (Ferreira et al. 2014). With very few exceptions, gender was not found to affect the rate of decline in the cognitive tasks assessed.
As is often the case in geropsychological research, studies are only rarely conducted in less affluent regions. This is also the case for studies on gender difference in cognitive decline, with most of them being conducted in Europe or in North America. Since it is possible that societal differences could influence men and women’s rate of cognitive decline differently, future research is needed to determine if the same type of patterns are present in less affluent regions of the world where there often also is less gender equity.
Cross-sectional assessment. An issue related to the discussion above is whether the magnitude of cognitive gender differences varies when different age groups are compared at the same point in time. Indeed, cross-sectional studies have found that the magnitude of gender differences may vary depending on the age groups assessed. One such study (Maylor et al. 2007) gathered data on about 200,000 individuals between the ages of 20 and 65. The participants, stemming from 53 countries, completed web-based tasks in English assessing visuospatial ability, verbal production, and episodic memory. As usual, men performed at a higher level than women on the visuospatial tasks, whereas women outperformed men on the category fluency and episodic memory tasks. However, although performance on all tasks declined with age, gender differences were smaller in the older age groups for the tasks where men outperformed women, and larger for the tasks in which women outperformed men. Although the underlying reasons for this pattern are unclear, it can be speculated that it is related to sample differences; older women around the world who both have access to computers and master English as a second language may have had to pass a higher cognitive threshold to acquire these capabilities, compared to participating men of the same age.
Why would gender differences in cognitive performance be larger in some age groups? As discussed previously, there have been substantial and continuous increases in living condition with regard to, for example, health, economy, family size, nutrition, and education. Such improvements, taking place over time, have been connected to improvements in cognitive performance, called the Flynn effect (Flynn 1987). An example of this comes from a study of Norwegian male conscripts who were tested on tasks assessing general cognitive ability from the mid-1950s to early 2000, where each yearly cohort was approximately 19 years old (Sundet et al. 2008). Results showed that the general cognitive ability increased more or less linearly and rather steeply (i.e., IQ increased from 100 to 108) from the mid-1950s to 1970, and to a lesser extent thereafter (i.e., IQ increased from 108 to 111). These increases in cognitive ability were strongly associated with increases in height, thereby pointing to the importance of improvements in nutrition and health care, factors that also influence cognitive performance.
In the SHARE study mentioned above, the investigators wanted to determine whether improvements in living conditions and educational opportunities, with women initially being more disadvantaged than men (Else-Quest and Grabe 2012), would explain why the magnitude of gender differences varied systematically across birth cohorts and regions (Weber et al. 2014). In doing so, a regional development index was created, which was specific for each country and age group. Historical information about each nation’s economy, fertility rate, infant mortality, life expectancy, and national education levels corresponding to the years when participants were 25–45 years old was collected. It was found that in countries with a higher regional development index, cognitive performance was also higher. More importantly, it could be seen that women improved in cognitive performance more than men did, suggesting that women were more affected by improved living conditions and education. Although it is unclear why this is the case, one could hypothesize that women benefit disproportionately from societal improvements because they may start from a more disadvantaged level (Else-Quest and Grabe 2012). If this hypothesis is correct, women would be expected to improve their cognitive abilities the most in regions that advance from relatively low levels of living conditions and educational opportunities to higher. Future research need to explore whether this assumption is correct.
Taken together, there is little evidence to suggest that the rate of decline in cognitive performance over the life span is different for men and women. However, cross-sectional studies, comparing individuals of different ages at the same time point, suggest that the magnitude of gender differences may have changed over time. Improvements in living conditions and better educational opportunities are factors that may lead to increased gender differences favoring women for some cognitive functions (e.g., episodic memory) and decreased (e.g., numeracy) or elimination (e.g., category fluency) of differences in other cognitive abilities. These changes seem to take place due to a general increase in cognitive performance over time, associated with societal improvements in living conditions and educational opportunities, where women are more affected than men.
- Ferreira, D., Rivero-Santana, A., Perestelo-Pérez, L., Westman, E., Wahlund, L.-O., Sarría, A, et al. (2014). Improving CSF biomarkers’ performance for predicting progression from Mild Cognitive Impairment to Alzheimer’s disease by considering different confounding factors: A meta-analysis. Frontiers in Aging Neuroscience. 6, Published online ahead of print. doi: 10.3389/fnagi.2014.00287.Google Scholar
- Maccoby, E. E., & Jacklin, C. N. (1974). The psychology of sex differences. Stanford, California: Stanford University Press.Google Scholar