Working memory for emotional facial expressions: role of the estrogen in young women

Physiological hormonal fluctuations during the menstrual cycle, postpartum, and menopause have been implicated in the modulation of mood, cognition, and affective disorders. Taking into account that women's performance in memory tasks can also fluctuate with circulating hormones levels across the menstrual cycle, the cognitive performance in a working memory task for emotional facial expressions, using the six basic emotions as stimuli in the delayed matching-to-sample, was evaluated in young women in different phases of the menstrual cycle. Our findings suggest that high levels of estradiol in the follicular phase could have a negative effect on delayed matching-to-sample working memory task, using stimuli with emotional valence. Moreover, in the follicular phase, compared to the menstrual phase, the percent of errors was significantly higher for the emotional facial expressions of sadness and disgust. The evaluation of the response times (time employed to answer) for each facial expression with emotional valence showed a significant difference between follicular and luteal in reference to the emotional facial expression of sadness. Our results show that high levels of estradiol in the follicular phase could impair the performance of working memory. However, this effect is specific to selective facial expressions suggesting that, across the phases of the menstrual cycle, in which conception risk is high, women could give less importance to the recognition of the emotional facial expressions of sadness and disgust. This study is in agreement with research conducted on non-human primates, showing that fluctuations of ovarian hormones across the menstrual cycle influence a variety of social and cognitive behaviors. Moreover, our data could also represent a useful tool for investigating emotional disturbances linked to menstrual cycle phases and menopause in women.     

Estrogens and memory in physiological and neuropathological conditions

Ovarian hormones can influence brain regions crucial to higher cognitive functions, such as learning and memory, acting at structural, cellular and functional levels, and modulating neurotransmitter systems. Among the main effects of estrogens, the protective role that they may play against the deterioration of cognitive functions occurring with normal aging is of essential importance. In fact, during the last century, there has been a 30 years increase in female life expectancy, from 50 to 83 years; however, the mean age of spontaneous menopause remains stable, 50-51 years, with variability related to race and ethnicity. Therefore, women are now spending a greater fraction of their lives in a hypoestrogenic state. Although many cognitive functions seem to be unaffected by normal aging, age-related impairments are particularly evident in tasks involving working memory (WM), whose deficits are a recognized feature of Alzheimer's disease (AD). Many studies conducted over the past two decades showed that the female gonadal hormone estradiol can influence performance of learning and memory tasks, both in animal and humans. There is a great deal of evidence, mostly from animal models, that estrogens can facilitate or enhance performance on WM tasks; therefore, it is very important to clarify their role on this type of memory. To this aim, in this review we briefly describe the most relevant neurobiological bases of estrogens, that can explain their effects on cognitive functioning, and then we summarize the results of works conducted in our laboratory, both on animals and humans, utilizing the menstrual/estrous cycle as a useful noninvasive model. Finally, we review the possible role of estrogens in neuropathological conditions, such as AD and schizophrenia.     

Vestibular evoked potentials from the vertical semicircular canals in humans evoked by roll-axis rotation in microgravity and under 1-G

It has been suggested that delayed (non-) matching to sample (DNMTS/DMTS) tasks using trial-unique stimuli and short, as well as longer delay intervals, can provide important insights into animal cognition. Therefore, this research examined the capability of the New World capuchin monkey (Cebus apella) in perform trial-unique DMTS and DNMTS tasks across delay intervals ranging between 8 s and 10 min. Subjects were tested using a version of the Wisconsin General Test Apparatus mounted in front of the animal's home cage. They were first trained on a basic DMTS/DNMTS task with an 8 s interval until they reached a learning criterion of nine correct responses in ten consecutive trials. All subjects reached the learning criterion in both DMTS/DNMTS tasks, and the number of trials to criterion did not differ between tasks. After reaching the criterion, subject's memory performance was successively assessed at delay intervals of 15 s, 60 s, 120 s and 10 min. For both DMTS/DNMTS tasks, the mean percentage of correct responses across delays was above chance and, interestingly, performance did not significantly decrease as function of delay increments. Comparisons based on each group's scores, averaged across the four delays, showed no difference between DMTS and DNMTS memory performance. These results indicate that capuchin monkeys are able to learn DMTS/DNMTS tasks in which they are required to respond to new pairs of stimuli on every trial. This demonstrates the capability of 'concept' learning in this species. Moreover, above chance performance on the memory tests indicates a working memory ability similar to that reported for the genus Macaca. Taken together, these data indicates that capuchin monkeys can be a valuable alternative model for investigations of the neuropsychological basis of memory.     

Implicit memory varies across the menstrual cycle: estrogen effects in young women.

Evidence that ovarian steroid hormones such as estrogen and progesterone affect cognition comes from studies of memory in older women receiving estrogen replacement therapy and studies of sexually dimorphic skills in young women across the menstrual cycle. Sixteen women (ages 18-28) completed tests of memory (implicit category exemplar generation, category-cued recall, implicit fragmented object identification) and sexually dimorphic skills (fine motor coordination, verbal fluency, mental rotations) at the early follicular (low estrogen and progesterone) and midluteal (high estrogen and progesterone) phases of the menstrual cycle. Performance on category exemplar generation, a test of conceptual implicit memory, was better at the midluteal than the follicular phase. In contrast, performance on a test of explicit memory, category-cued recall, did not vary across the menstrual cycle. At Session 1, women in the follicular phase performed better on the fragmented object identification task than did those in the midluteal phase. This unexpected finding suggests that high levels of ovarian hormones might inhibit perceptual object priming. Results confirmed previous reports of decreased mental rotations and improved motor skills and fluency in the midluteal phase. Estradiol levels correlated positively with verbal fluency and negatively with mental rotations and perceptual priming, which suggest that estrogen, and not progesterone, was responsible for the observed changes in cognition. Mood did not vary across the cycle phases. Overall, the findings suggest that estrogen may facilitate the automatic activation of verbal representations in memory.     

Estradiol selectively affects processing of conspecifics' faces in female rhesus monkeys

Estrogen deficiency following ovariectomy or menopause increases the risk of developing diseases such as osteoporosis and may also lead to memory impairment. Although estrogen replacement therapy (ERT) alleviates many symptoms associated with estrogen loss, it is not clear whether it also benefits cognitive function. The effect of estrogens upon cognition can best be studied in an animal model of human menopause, in which estrogen levels can be experimentally manipulated. Six young ovariectomized female rhesus monkeys (6-9 years old) were tested on a battery of touchscreen-based cognitive tasks, including the Matching-to-Sample (MTS) task with mixed delays and the spatial, object, and face conditions of the Delayed Recognition Span Test (DRST). Monkeys were tested 5 days a week, one task per week, for a total of 8 months, while undergoing treatments with placebo and ethinyl estradiol (EE2) in alternating 28-days blocks. Blood samples were collected to verify EE2 levels. We also observed the monkeys by video monitor during test sessions and recorded locomotor activity and response topology. Performance on the face-DRST, a task that involved selecting the new face in an increasing array of rhesus monkey faces, was impaired by EE2 treatment, as compared to placebo. Other tasks were unaffected by EE2. There was no clear evidence of EE2 effects upon motor activity or anxiety. In order to test the reliability of our findings, we conducted an additional experiment in which the monkeys were again given the face-DRST with different categories of face stimuli for 4 months, while receiving placebo and EE2 in alternating 7-days blocks. They performed each task 4-5 days/week for 4 weeks with (1) the same rhesus monkey faces as in the first experiment, (2) human faces, (3) chimpanzee faces, and (4) novel rhesus monkey faces. Face-DRST performance did not vary as a function of treatment when human or chimpanzee faces were used as stimuli. In contrast, periods of EE2 treatment were associated with a lower performance for both sets of rhesus monkey faces. These findings suggest that EE2 treatment has a detrimental effect on processing faces of conspecifics by female rhesus monkeys. We speculate that estrogens may produce this effect by enhancing emotional reactivity to socially relevant stimuli.     

Effects of treatment with leuprolide acetate depot on working memory and executive functions in young premenopausal women

The goal of this study was to assess the influence of sex steroid hormone suppression on performance on tests of prefrontal cortex (PFC) and working memory function (WM) in premenopausal women. Twenty-five women were treated with leuprolide acetate depot (LAD), a gonadotropin releasing hormone (GnRH) analog that chemically suppressed ovarian function as treatment of various benign gynecologic disorders. Performance on tests of PFC and WM of the LAD-treated women was assessed at pretreatment baseline and, again, following 4 weeks of treatment and their performance was compared to that of 25 healthy, control participants matched on age, education, and general intelligence. Following 4 weeks of LAD treatment, estrogen levels decreased to the postmenopausal range whereas progesterone levels fell to the lower limit of the menstrual cycle range of values. Furthermore, the LAD-treated women also experienced a significant deterioration in mood, in health-related symptoms and in performance on two tests of WM following 4 weeks of treatment. It was determined that only the post-treatment declines in estrogen, but not those in progesterone, in mood, or in health-related symptoms were associated with the worsening of performance on the WM tests. These findings provide new evidence that estrogen is influential in the maintenance of WM processes in premenopausal women.     

Modeling the role of working memory and episodic memory in behavioral tasks

The mechanisms of goal-directed behavior have been studied using reinforcement learning theory, but these theoretical techniques have not often been used to address the role of memory systems in performing behavioral tasks. This work addresses this shortcoming by providing a way in which working memory (WM) and episodic memory may be included in the reinforcement learning framework, then simulating the successful acquisition and performance of six behavioral tasks, drawn from or inspired by the rat experimental literature, that require WM or episodic memory for correct performance. With no delay imposed during the tasks, simulations with WM can solve all of the tasks at above the chance level. When a delay is imposed, simulations with both episodic memory and WM can solve all of the tasks except a disambiguation of odor sequences task.     

Recovery of motor and cognitive function after cerebellar lesions in a songbird – role of estrogens

In addition to its key role in complex motor function, the cerebellum is increasingly recognized to have a role in cognition. Songbirds are particularly good models for the investigation of motor and cognitive processes but little is known about the role of the songbird cerebellum in these processes. To explore cerebellar function in a songbird, we lesioned the cerebellum of adult female zebra finches and examined the effects on a spatial working memory task and on motor function during this task. There is evidence for steroid synthesis in the songbird brain and neurosteroids may have an impact on some forms of neural plasticity in adult songbirds. We therefore hypothesized that neurosteroids would affect motor and cognitive function after a cerebellar injury. We found that cerebellar lesions produced deficits in motor and cognitive aspects of a spatial task. In line with our prediction, birds in which estrogen synthesis was blocked had impaired performance in our spatial task compared with those that had estrogen synthesis blocked but estrogen replaced. There was no clear effect of estrogen replacement on motor function. We also found that lesions induced expression of the estrogen synthetic enzyme aromatase in reactive astrocytes and Bergmann glia around a cerebellar lesion. These data suggest that the cerebellum of songbirds mediates both motor and cognitive function and that estrogens may improve the recovery of cognitive aspects of cerebellar function after injury.     

Associations between circulating sex steroid hormones and cognition in normal elderly women

OBJECTIVE: To provide exploratory analyses of associations between levels of several sex hormones and cognitive performance in elderly women. BACKGROUND: Sex steroid hormones are implicated in the cognitive processes of the adult brain. Comparing cognitive performance across or between conditions associated with different hormone levels, such as phases of the menstrual cycle, surgical menopause, and estrogen replacement therapy suggests conditions with higher levels of estrogen are associated with better verbal memory and possibly worse visuospatial ability. METHOD: The authors measured circulating sex hormone levels in 39 highly educated, nondemented, predominantly white elderly women. Levels were correlated with neuropsychological performance, controlling for age, education, frequency of prior testing, use of estrogen replacement, and depression. RESULTS: High estradiol levels were associated with better delayed verbal memory and retrieval efficiency, whereas low levels were associated with better immediate and delayed visual memory. Levels of testosterone were related positively to verbal fluency. Levels of progesterone and androstenedione were unrelated to cognitive performance. CONCLUSIONS: Both estrogen and testosterone showed associations with cognitive performance. Estrogen may enhance, and depress, specific cognitive skills.     

Sex and modulatory menstrual cycle effects on sleep related memory consolidation

The benefit of sleep in general for memory consolidation is well known. The relevance of sleep characteristics and the influence of hormones are not well studied. We explored the effects of a nap on memory consolidation of motor (finger-tapping-task) and verbal (associated-word-pairs) tasks in following settings: A: young, healthy males and females during early-follicular phase (n=40) and B: females during mid-luteal and early-follicular phase in the menstrual cycle (n=15). We found a sex and in women a menstrual cycle effect on memory performance following a nap. Men performed significantly better after a nap and women did so only in the mid-luteal phase of their menstrual cycle. Only the men and the women in their mid-luteal phase experienced a significant increase in spindle activity after learning. Furthermore, in women estrogen correlated significantly with the offline change in declarative learning and progesterone with motor learning. The ratio of the 2nd and 4th digit, which has been associated to fetal sex hormones and cognitive sex differences, significantly predicted the average performance of the female subjects in the learning tasks. Our results demonstrate that sleep-related memory consolidation has a higher complexity and more influencing factors than previously assumed. There is a sex and menstrual cycle effect, which seems to be mediated by female hormones and sleep spindles. Further, contrary to previous reports, consolidation of a simple motor task can be induced by a 45 min NREM sleep nap, thus not dependent on REM sleep.     

Estrogen affects cognition in women with psychosis

Estrogen has been reported to affect aspects of cognition and psychopathology in women, both normal and with psychosis. This study aimed to replicate and extend this research by investigating the effect of estrogen on cognition over the menstrual cycle in a group of normal women and women with psychosis. The sample consisted of 31 premenstrual normal control subjects, and 29 women with psychosis. Subjects were tested twice, 2 weeks apart on a number of cognitive tests. There was no difference in Positive and Negative Symptom Scale scores between the follicular and luteal phases of the menstrual cycle. Both groups of women performed better on the Revised Mental Rotation Test and Trails A during the follicular phase when estrogen levels were low. Contrary to expectation, during the luteal phase, when estrogen was high, the control subjects showed no significant improvement in performance on verbal articulatory-motor tasks, and the women with psychosis performed significantly worse on the Purdue Pegboard. The unexpected adverse effect of high levels of estrogen on motor performance in the psychotic women was hypothesized to be related to their disease process.     

Brain Activation Patterns Associated with the Effects of Emotional Distracters during Working Memory Maintenance in Patients with Generalized Anxiety Disorder

Few studies have assessed the neural mechanisms of the effects of emotion on cognition in generalized anxiety disorder (GAD) patients. In this functional MRI (fMRI), we investigated the effects of emotional interference on working memory (WM) maintenance in GAD patients. Fifteen patients with GAD participated in this study. Event-related fMRI data were obtained while the participants performed a WM task (face recognition) with neutral and anxiety-provoking distracters. The GAD patients showed impaired performance in WM task during emotional distracters and showed greater activation on brain regions such as DLPFC, VLPFC, amygdala, hippocampus which are responsible for the active maintenance of goal relevant information in WM and emotional processing. Although our results are not conclusive, our finding cautiously suggests the cognitive-affective interaction in GAD patients which shown interfering effect of emotional distracters on WM maintenance.     

Competition between frontoparietal control and default networks supports social working memory and empathy

An extensive body of literature has indicated that there is increased activity in the frontoparietal control network (FPC) and decreased activity in the default mode network (DMN) during working memory (WM) tasks. The FPC and DMN operate in a competitive relationship during tasks requiring externally directed attention. However, the association between this FPC-DMN competition and performance in social WM tasks has rarely been reported in previous studies. To investigate this question, we measured FPC-DMN connectivity during resting state and two emotional face recognition WM tasks using the 2-back paradigm. Thirty-four individuals were instructed to perform the tasks based on either the expression [emotion (EMO)] or the identity (ID) of the same set of face stimuli. Consistent with previous studies, an increased anti-correlation between the FPC and DMN was observed during both tasks relative to the resting state. Specifically, this anti-correlation during the EMO task was stronger than during the ID task, as the former has a higher social load. Intriguingly, individual differences in self-reported empathy were significantly correlated with the FPC-DMN anti-correlation in the EMO task. These results indicate that the top-down signals from the FPC suppress the DMN to support social WM and empathy.     

Prefrontal cortex as the site of estrogen's effect on cognition

The hippocampus has long been presumed the primary site of action of estrogens on cognition; and explicit memory is considered the cognitive function most vulnerable to menopausal loss of estrogen. We hypothesize instead that the prefrontal cortex and its neural circuitry are prime mediators of estrogen's role in cognition. We also propose that previously reported menopausal cognitive decline, presumed to be hippocampally mediated, may be secondary to executive dysfunction. We used a cross sectional design to compare the performance of nine menopausal women on hormone replacement therapy (HRT) and 10 menopausal women with no prior exposure to HRT on a battery of neuropsychological tests. The battery was comprised primarily of tests of memory and executive functioning. Executive functioning is mediated by the frontal lobes and encompasses working memory, directed attention, the inhibition of inappropriate responses, cognitive set switching, and behavioral monitoring. Unlike most previous studies, we used a memory measure that yields multiple scores reflecting various problem-solving strategies and error types, thus isolating spared and impaired cognitive processes. Results yielded both qualitative and quantitative evidence for disruption of cognitive processes subserved by the frontal lobes rather than the hippocampus: 1) despite intact free recall on a list-learning task (CVLT), untreated menopausal women were relatively impaired in correctly recognizing words previously learned and distinguishing them from items not on the list (discriminability), 2) untreated women also had difficulty inhibiting inappropriate responses in the form of perseverative errors, and 3) the non-HRT group consistently performed worse on the N-back test of working memory. The prefrontal cortex is critical for intact working memory and estrogen enhances performance on working memory tasks. In conclusion, this study provides preliminary evidence for executive dysfunction in untreated menopausal women as women with HRT outperformed women without HRT on tests requiring directed attention, inhibition of inappropriate responses, and cognitive set switching.     

History of Previous Midlife Estradiol Treatment Permanently Alters Interactions of Brain Insulin-like Growth Factor-1 Signaling and Hippocampal Estrogen Synthesis to Enhance Cognitive Aging in a Rat Model of Menopause

Across species, including humans, elevated levels of brain estrogen receptor (ER) α are associated with enhanced cognitive aging, even in the absence of circulating estrogens. In rodents, short-term estrogen treatment, such as that commonly used in the menopausal transition, results in long-term increases in ERα levels in the hippocampus, leading to enhanced memory long after termination of estrogen treatment. However, mechanisms by which increased levels of brain ERα enhances cognitive aging remain unclear. Here we demonstrate in aging female rats that insulin-like growth factor-1 (IGF-1), which can activate ER via ligand-independent mechanisms, requires concomitant synthesis of brain-derived neuroestrogens to phosphorylate ERα via MAPK signaling, ultimately resulting in enhanced memory. In a rat model of menopause involving long-term ovarian hormone deprivation, hippocampal neuroestrogen activity decreases, altering IGF-1 activity and resulting in impaired memory. However, this process is reversed by short-term estradiol treatment. Forty days of estradiol exposure following ovariectomy results in maintenance of neuroestrogen levels that persist beyond the period of hormone treatment, allowing for continued interactions between IGF-1 and neuroestrogen signaling, elevated levels of hippocampal ERα, and ultimately enhanced memory. Collectively, results demonstrate that short-term estradiol use following loss of ovarian function has long-lasting effects on hippocampal function and memory by dynamically regulating cellular mechanisms that promote activity of ERα in the absence of circulating estrogens. Translational impacts of these findings suggest lasting cognitive benefits of short-term estrogen use near menopause and highlight the importance of hippocampal ERα, independent from the role of circulating estrogens, in regulating memory in aging females.SIGNIFICANCE STATEMENT Declines in ovarian hormones following menopause coincide with increased risk of cognitive decline. Because of potential health risks, current recommendations are that menopausal estrogen therapy be limited to a few years. Long-term consequences for the brain and memory of this short-term midlife estrogen therapy are unclear. Here, in a rodent model of menopause, we determined mechanisms by which short-term midlife estrogen exposure can enhance hippocampal function and memory with cognitive benefits and molecular changes enduring long after termination of estrogen exposure. Our model indicates long-lasting benefits of maintaining hippocampal estrogen receptor function in the absence of ongoing estrogen exposure and suggests potential strategies for combating age-related cognitive decline.     

Stress shifts brain activation towards ventral 'affective' areas during emotional distraction

Acute stress has been shown to impair working memory (WM), and to decrease prefrontal activation during WM in healthy humans. Stress also enhances amygdala responses towards emotional stimuli. Stress might thus be specifically detrimental to WM when one is distracted by emotional stimuli. Usually, emotional stimuli presented as distracters in a WM task slow down performance, while evoking more activation in ventral 'affective' brain areas, and a relative deactivation in dorsal 'executive' areas. We hypothesized that after acute social stress, this reciprocal dorsal-ventral pattern would be shifted towards greater increase of ventral 'affective' activation during emotional distraction, while impairing WM performance. To investigate this, 34 healthy men, randomly assigned to a social stress or control condition, performed a Sternberg WM task with emotional and neutral distracters inside an MRI scanner. Results showed that WM performance after stress tended to be slower during emotional distraction. Brain activations during emotional distraction was enhanced in ventral affective areas, while dorsal executive areas tended to show less deactivation after stress. These results suggest that acute stress shifts priority towards processing of emotionally significant stimuli, at the cost of WM performance.     

Functional anatomy of visuo-spatial working memory during mental rotation is influenced by sex, menstrual cycle, and sex steroid hormones

Recent observations indicate that sex and level of steroid hormones may influence cortical networks associated with specific cognitive functions, in particular visuo-spatial abilities.

The present study probed the influence of sex, menstrual cycle, and sex steroid hormones on 3D mental rotation and brain function using 3-T fMRI. Twelve healthy women and 12 men were investigated. Menstrual cycle and hormone levels were assessed. The early follicular and midluteal phase of the menstrual cycle were chosen to examine short-term cyclical changes.

Parietal and frontal areas were activated during mental rotation in both sexes. Significant differences between men and women were revealed in both phases of menstrual cycle. In men we observed a significant correlation of activation levels with testosterone levels in the left parietal lobe (BA 40). In women, a cycle-dependent correlation pattern was observed for testosterone: brain activation correlated with this male hormone only during the early follicular phase. In both cycle phases females’ brain activation was significantly correlated with estradiol in frontal and parietal areas.

Our study provides evidence that fMRI-related activity during performance of cognitive tasks varies across sex and phases of the menstrual cycle. The variation might be partly explained by better task performance in men, but our results indicate that further explanations like basic neuronal or neurovascular effects modulated by steroid hormones must be considered. Both estradiol and testosterone levels may influence fMRI signals of cognitive tasks, which should affect selection of subjects for future fMRI studies.     

Preliminary investigation of the influence of dopamine regulating genes on social working memory

Working memory (WM) refers to mental processes that enable temporary retention and manipulation of information, including information about other people (“social working memory”). Previous studies have demonstrated that nonsocial WM is supported by dopamine neurotransmission. Here, we investigated in 131 healthy adults whether dopamine is similarly involved in social WM by testing whether social and nonsocial WM are influenced by genetic variants in three genes coding for molecules regulating the availability of dopamine in the brain: catechol-O-methyltransferase (COMT), dopamine active transporter (DAT), and monoamine-oxidase A (MAOA). An advantage for the Met allele of COMT was observed in the two standard WM tasks and in the social WM task. However, the influence of COMT on social WM performance was not accounted for by its influence on either standard WM paradigms. There was no main effect of DAT1 or MAOA, but a significant COMT x DAT1 interaction on social WM performance. This study provides novel preliminary evidence of effects of genetic variants of the dopamine neurotransmitter system on social cognition. The results further suggest that the effects observed on standard WM do not explain the genetic effects on effortful social cognition.     

Altered Working Memory Processing of Emotion in Adolescents with Dysphoric Symptomatology: An Eye Tracking Study

It has been suggested that altered processing of emotion during cognitive control plays an important role in the etiology of depressive symptoms. The current study investigates the influence of emotional stimuli on working memory (WM) in adolescents with dysphoric symptomatology (DS). Twenty-five adolescents with DS and 40 adolescents with no dysphoric symptomatology (NDS) completed a memory-guided eye movement task. This task assessed the influence of irrelevant affective information on WM processes during high and low cognitive load. Latency analyses showed that, in the high load WM condition, negative distractors disturbed WM performance in adolescents with NDS, but not in adolescents with DS. Accuracy analyses revealed that adolescents with NDS had higher accuracy rates in the presence of positive distractors relative to negative and neutral distractors, and in comparison to adolescents with DS. The findings indicated altered WM performance in the context of emotional distractors in adolescents with DS and may contribute to theoretical knowledge and early prevention of youth depression.     

Age and gender interactions on verbal memory performance.

Age and gender effects on verbal episodic memory are well established. However, the possibility of interactions between age and gender has been raised by studies linking estrogen and verbal memory performance, and by research suggesting gender differences in age-related cortical atrophy. We evaluated whether age by gender interactions in verbal memory were present. Subjects within three years of the median age of menopause were excluded from a large cohort of normal subjects, resulting in a younger sample (16-47 years) of 288 men and 285 women, and an older sample (55-89 years) of 201 men and 245 women. All subjects were administered the CVLT-2, a multiple-trial list-learning task. Verbal memory was negatively correlated with age for younger men, older men, and older women, but not for younger women. Multivariate analyses indicated age by gender interactions on memory for the younger group but not the older group. Results indicate that verbal memory declines with age for younger men but not younger women, whereas both older men and older women show age-related declines. These findings are consistent with hypotheses linking estrogen and verbal memory performance, and with imaging data suggesting that age-related hippocampal atrophy is found in younger men but not younger women. The role of estrogen on cognition in normal aging warrants further study.