See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/262570617 Age and Gender Modulate the Neural Circuitry Supporting Facial Emotion Processing in Adults with Major Depressive Disorder ARTICLE in AMERICAN JOURNAL OF GERIATRIC PSYCHIATRY · MAY 2014 Impact Factor: 4.24 · DOI: 10.1016/j.jagp.2014.05.007 CITATIONS READS 2 27 7 AUTHORS, INCLUDING: Lisa J Rapport Linas A Bieliauskas 61 PUBLICATIONS 1,456 CITATIONS 140 PUBLICATIONS 1,785 CITATIONS Wayne State University SEE PROFILE University of Michigan SEE PROFILE Sara L. Weisenbach Scott A Langenecker 31 PUBLICATIONS 219 CITATIONS 94 PUBLICATIONS 1,430 CITATIONS University of Michigan SEE PROFILE University of Illinois at Chicago SEE PROFILE Available from: Scott A Langenecker Retrieved on: 07 February 2016 Age and Gender Modulate the Neural Circuitry Supporting Facial Emotion Processing in Adults with Major Depressive Disorder Emily M. Briceño, Ph.D., Lisa J. Rapport, Ph.D., Michelle T. Kassel, B.A., Linas A. Bieliauskas, Ph.D., Jon-Kar Zubieta, M.D., Ph.D., Sara L. Weisenbach, Ph.D., Scott A. Langenecker, Ph.D. Objectives: Emotion processing, supported by frontolimbic circuitry known to be sensitive to the effects of aging, is a relatively understudied cognitiveeemotional domain in geriatric depression. Some evidence suggests that the neurophysiological disruption observed in emotion processing among adults with major depressive disorder (MDD) may be modulated by both gender and age. Therefore, the present study investigated the effects of gender and age on the neural circuitry supporting emotion processing in MDD. Design: Cross-sectional comparison of fMRI signal during performance of an emotion processing task. Setting: Outpatient university setting. Participants: One hundred adults recruited by MDD status, gender, and age. Measurements: Participants underwent fMRI while completing the Facial Emotion Perception Test. They viewed photographs of faces and categorized the emotion perceived. Contrast for fMRI was of face perception minus animal identification blocks. Results: Effects of depression were observed in precuneus and effects of age in a number of frontolimbic regions. Threeway interactions were present between MDD status, gender, and age in regions pertinent to emotion processing, including frontal, limbic, and basal ganglia. Young women with MDD and older men with MDD exhibited hyperactivation in these regions compared with their respective same-gender healthy comparison (HC) counterparts. In contrast, older women and younger men with MDD exhibited hypoactivation compared to their respective same-gender HC counterparts. Conclusions: This the first study to report gender- and age-specific differences in emotion processing circuitry in MDD. Gender-differential mechanisms may underlie cognitiveeemotional disruption in older adults with MDD. The present findings have implications for improved probes into the heterogeneity of the MDD syndrome. (Am J Geriatr Psychiatry 2014; -:-e-) Key Words: Emotion processing, depression, fMRI, gender, age Received November 7, 2013; revised May 15, 2014; accepted May 17, 2014. From the Department of Psychiatry, University of Michigan (EMB, MTK, LAB, J-KZ, SLW, SAL), Ann Arbor, MI; Department of Psychology, Wayne State University (EMB, LJR), Detroit, MI; Department of Psychiatry, University of Illinois at Chicago (MTK, SLW, SAL), Chicago, IL; and the Jesse Brown VA Healthcare System (SLW), Chicago, IL. Preliminary results of this work were presented at the 41st Annual Meeting of the International Neuropsychological Society, Waikoloa, HI, February 6e9, 2013. Send correspondence and reprint requests to Scott A. Langenecker, Ph.D., Department of Psychiatry, University of Illinois at Chicago, 1601 W. Taylor St, m/c 912, Chicago, IL 60612. e-mail: slangenecker@psych.uic.edu Supplemental digital content is available for this article in the HTML and PDF versions of this article on the journal’s website (www.ajgponlinle.org). Ó 2014 American Association for Geriatric Psychiatry http://dx.doi.org/10.1016/j.jagp.2014.05.007 Am J Geriatr Psychiatry -:-, - 2014 1 Age and Gender Effects in MDD I ndividuals with major depressive disorder (MDD) exhibit emotion processing inefficiencies, including recognizing facial emotions,1 reflecting potential dysfunction in neural pathways supporting these processes. Neuroimaging studies have demonstrated distinct patterns in the neural circuitry supporting emotion processing in MDD, affecting limbic, basal ganglia, and cortical (e.g., insula, cingulate, fusiform, frontal) regions.2,3 Substantial variability is apparent across these studies with regard to regions affected and direction of differences (e.g., hyperactivation2 versus hypoactivation4) compared with healthy comparison (HC) participants. Although the source of variability in those findings is unknown, methodological differences are common. Studies have varied with regard to task design (e.g., implicit versus explicit paradigms) and regions investigated. Studies have also varied with regard to illness characteristics (e.g., phase of illness, severity) and demographic (e.g., gender, age) composition of participants included in study samples. Many studies included a greater proportion of women than men in MDD samples, reflecting the greater prevalence of MDD in women.5 Although samples are typically roughly balanced in the proportion of men and women, this procedure does not identify, and may mask, potential gender-specific neural circuit disruptions in MDD. There is compelling evidence for gender-specific neural circuit disruption underlying emotion processing in MDD. Gender-specific decrements in facial emotion perception accuracy in MDD have been reported:6 Women with MDD were less accurate at detecting emotions than both HC women and men with MDD, whereas men with MDD performed similarly to HC men. Furthermore, gender differences in emotion recognition accuracy in healthy adults are well-established, favoring women,7 as well as in the neural circuitry supporting emotion processing.8,9 These findings, combined with evidence for genderspecific disruption in emotion processing accuracy in MDD, underscore the critical need to evaluate the role of gender in disrupted emotion processing in MDD. The term gender is used rather than sex in the present study to reflect the range of possible biological and sociocultural contributions to these differences. Age also has a high likelihood of impact upon facial emotion processing in MDD. Healthy older adults are less accurate at recognizing emotional 2 expressions than younger adults and tend to categorize emotions as positive.10,11 Older adults also exhibit reduced limbic and greater cortical activation (e.g., insula, frontal cortex12,13) during facial emotion processing compared with younger adults. Because of these effects upon emotion processing in healthy aging, advanced age may conceivably increase burden on emotion processing in late-life MDD. Only two studies to date have investigated emotion processing in late-life depression. One pilot study,14 composed exclusively of 13 women with late-onset depression compared with older HC women, reported reduced engagement of the ventrolateral prefrontal cortex during emotional evaluation of emotionally valenced words. The second15 used a region-of-interest analysis of 27 older adults with primarily late-onset MDD (64% women), showing increased subgenual anterior cingulate and insula activation relative to 33 HC older adults. These initial studies suggest that disruption in emotion processing circuitry continues in MDD into late life. Importantly, no studies have addressed whether emotion processing circuitry disruptions are similar in young and older adults with MDD, and none have systematically evaluated the influence of gender. Supporting the need for such a study, age and gender each have been shown to be determinants of behavioral decrements in face emotion processing accuracy in MDD,16 such that women with MDD performed more poorly than HC women in both young and older adults, whereas men with MDD performed more poorly than HC men only within older adults. Furthermore, older men may be differentially vulnerable to structural abnormalities compared with older women with MDD, including white matter hyperintensities and frontal volume loss.17,18 Taken together, gender and age are known to influence emotion processing accuracy in healthy adults, and to influence emotion processing accuracy in adults with MDD. It is unknown, however, how these characteristics together affect the neural circuitry supporting emotion processing in MDD, which was the aim of the present study. We hypothesized that gender and age each would moderate the effects of MDD on the neurophysiology of emotion processing. Because of the paucity of data addressing this topic, we could not offer specific directional hypotheses for MDD by gender or age. Am J Geriatr Psychiatry -:-, - 2014 Briceño et al. METHODS Participants One hundred ten adults, including 53 with MDD and 57 HC in younger and older age groups (Table 1) were recruited through geriatric psychiatry and primary care clinics, research volunteer databases, and community advertisements. Exclusionary criteria included contraindications for magnetic resonance imaging (MRI), uncontrolled hypertension or diabetes, any neurological disorder, head injury with loss of consciousness of greater than 5 minutes, and major medical conditions that could affect the central nervous system. Participants were also excluded based upon any history of psychotic symptoms, bipolar disorder, schizophrenia, current substance use disorder, or history of substance dependence within 5 years of the MRI. Within the older MDD group, individuals with an age of onset after 45 years were excluded to minimize the likelihood of physiological/acquired contribution to disease pathogenesis. Individuals were not excluded on the basis of taking psychotropic medications, although those using anxiolytics on an as-needed basis were asked to refrain from use 24 hours prior to the scan. HC participants were free from a personal history of psychiatric illness. MDD participants were diagnosed according to the Structured Clinical Interview for the DSM-IV criteria (SCID-IP/NP19). Depression severity was measured with the Hamilton Rating Scale for DepressioneSecond Edition.20 Potential participants were specifically recruited in order to meet minimum cell sizes of 12 for subgroups. All participants completed informed consent procedures per institutional review board guidelines. Procedure The Facial Emotion Perception Test21,22 (Fig. 1) was completed in the fMRI scanner. Details regarding this task have been published elsewhere.2,23 Briefly, participants viewed photographs of faces and were required to categorize each into one of four emotions (happy, sad, angry, fearful). As a control task, blocks of animal photographs were presented, requiring categorization into one of four categories (dogs, cats, primates, birds). Functional neuroimaging acquisition Am J Geriatr Psychiatry -:-, - 2014 parameters were consistent with prior studies in our laboratory.2,9,23 Analysis Imaging data were screened for data corruption. Participants were excluded for extensive signal loss due to signal distortion (N: 6 MDD), excessive movement (N: 3 HC), and unusual head shape resulting in data loss (N: 1 HC). All analyses were conducted with the remaining 100 participants. fMRI data were evaluated with a block design using a 2 (MDD/HC)  2 (gender)  2 (age group) ANOVA, with the Faces minus Animals contrast as the dependent variable. Significance thresholds were derived with AlphaSim24 (p <0.005, k >55). Post hoc, exploratory analyses were conducted using extracted data from significance maps using MarsBaR. Activation maps for the Animals-only contrast were superimposed upon the activation maps for the Faces minus Animals contrast for each result to exclude the control condition as a source of activation difference. Overlap between Animals and Faces minus Animals was absent from nearly all contrasts—except effects of age—and noted in the tables. RESULTS The ANOVA demonstrated modest main effects of MDD status, gender, and age group (Table 2). Two regions had a significant interaction between MDD status and age group, and four regions had a significant interaction between gender and age group (Table 3). There were no regions significant for an MDD status  gender interaction. A three-way interaction between MDD status, gender, and age group was present in seven regions (Table 3). The precuneus was less active in MDD relative to comparison participants, women activated temporal/occipital regions more than men, and there were four regions more active in younger than older adults (Table 2). Two clusters (precentral gyrus and dentate of cerebellum) were significant for an interaction between MDD status and age group (Table 3). For the precentral gyrus cluster, there was a nonsignificant effect of MDD status in the young group, whereas the MDD sample exhibited less activation than HC in the older group. For the dentate cluster, there was greater 3 Age and Gender Effects in MDD TABLE 1. Descriptive Statistic Comparing Demographic, Medical, and Performance Characteristics for Participants with Major Depressive Disorder (n [ 53) and Healthy Controls (n [ 57) Grouped by Age (Younger and Older) and Gender Younger MDD Variables a Age Educationa HDRSa HDRS: Range of scores Charlson Comorbidity Indexa Years of illness (MDD only)a On psychotropic medication (%) Diabetes (N) Hypertension (N) Sleep apnea (N) Heart condition (N) FEPT Task Accuracy (% correct)b Older HC MDD HC Women (N [ 15) Men (N [ 12) Women (N [ 19) Men (N [ 13) Women (N [ 12) Men (N [ 14) Women (N [ 12) Men (N [ 13) M (SD) M (SD) M (SD) M (SD) M (SD) M (SD) M (SD) M (SD) 29.2 (7.8) 16.2 (2.4) 17.4 (4.3) 12e25 0.0 (0.0) 9.2 (8.8) 60 0 1 0 0 85.3 (7.8) 25.5 (3.5) 16.5 (1.9) 14.8 (4.4) 7e21 0.1 (0.3) 8.9 (5.3) 20 0 0 1 0 82.3 (5.6) 26.4 (7.7) 15.9 (2.5) 0.5 (0.8) 0e2 0.0 (0.0) NA NA 0 0 0 0 86.9 (5.7) 24.0 (6.0) 15.2 (2.1) 1.3 (1.7) 0e5 0.0 (0.0) NA NA 0 0 0 0 87.7 (5.9) 64.8 (6.3) 15.4 (2.2) 16.3 (6.3) 7e27 0.2 (0.4) 40.5 (12.3) 75 0 5 2 4 73.8 (9.1) 66.0 (9.6) 16.1 (2.8) 14.6 (4.4) 6e24 0.4 (0.9) 42.9 (17.3) 84 0 8 0 1 80.1 (9.4) 69.2 (8.6) 17.1 (1.5) 0.6 (0.8) 0e2 0.0 (0.0) NA NA 0 3 0 0 77.2 (6.7) 67.1 (8.2) 16.4 (2.4) 1.2 (1.1) 0e3 0.1 (0.3) NA NA 3 4 1 2 74.8 (8.5) Notes: MDD: Major Depressive Disorder; HC: Healthy Comparison; HDRS: Hamilton Depression Rating Scale; FEPT: Facial Emotion Perception Test. a All p values >0.05: age within age group (younger, F(3, 55) ¼ 1.49, p ¼ 0.23; older, F(3, 47) ¼ 0.61, p ¼ 0.61); education (F(7, 102) ¼ 0.85, p ¼ 0.55), and Charlson (F(7, 101) ¼ 1.51, p ¼ 0.17) across entire sample; Hamilton Depression Rating Scale within the MDD group (F(3, 49) ¼ 1.02, p ¼ .39), and years of illness between men and women within the younger (t(23) ¼ 0.08, p ¼ 0.94) and older (t(24) ¼ 0.39, p ¼ 0.70) MDD groups separately. Data on medical conditions were missing from 10 younger HC participants. Years of illness was missing from two younger MDD participants. HDRS was missing for 7 HC participants. b F(7, 102) ¼ 7.30, p <0.001. Posthoc analyses included in the supplemental text. FIGURE 1. Schematic of the Facial Emotion Perception Test (FEPT). Note: Presentation begins with fixation cross (þ) followed by 300 msec presentation of face (or animal), then 100 msec mask, followed by a 2,600 msec response window. activation in MDD than in HC in the young group and less activation in MDD compared with the HC group within the older group. Regions significant for the interaction between MDD status, gender, and age group included clusters in the inferior, middle, and superior frontal gyri, cingulate, and putamen (Table 3; Fig. 2). Post hoc 4 analysis with activation data extracted with MarsBaR revealed that for young adults, the interaction between MDD status and gender was significant for all seven clusters (Supplemental Table 1; available online). Within young women, activation in the MDD sample was significantly greater than the HC sample for four of the seven clusters, with the same Am J Geriatr Psychiatry -:-, - 2014 Briceño et al. TABLE 2. Regions Significantly Associated with Main Effects of MDD, Gender, and Age Group Lobe Gyrus BA mm3 7 Z Group Difference 46 3.49 D<C x y z 472 20 65 Main effect of MDD Parietal Precuneus Temporal Occipital Middle Temporal Fusiform 19 19 888 608 40 38 42 73 2 11 3.50 4.26 W>M W>M Frontal Occipital Occipital Posterior Inferior Frontala Cuneus/Linguala Linguala Pyramisa 47 17/18 18 536 4496 808 1192 40 17 13 20 34 79 73 73 1 8 1 29 3.57 7.98 3.73 3.41 Y Y Y Y Main effect of gender Main effect of age > > > > Oa Oa Oa Oa Notes: D: Major Depressive Disorder group; C: Healthy Comparison group. a Animals condition only O > Y. TABLE 3. Regions Significantly Associated with Interactions Between MDD Status, Gender, and Age Group Lobe Gyrus MDD  Age Group Interaction Frontal Precentral Posterior BA mm3 6 y z 512 15 19 59 3.63 664 15 60 19 3.83 9 488 41 28 26 3.05 Cerebellum; Dentate Gender  Age Group Interaction Frontal Middle Frontala Z x Frontal Precentral 43 680 31 7 29 3.37 Subcortical Claustrum/Insula 13 560 31 6 9 3.22 Subcortical Pulvinar 3704 3 32 16 4.15 1464 472 504 920 2368 616 1328 20 38 45 36 24 20 4 20 26 12 10 16 39 11 32 6 29 7 36 35 22 3.66 3.28 3.09 3.39 3.90 3.28 3.84 MDD  Gender  Age Frontal Frontal Frontal Frontal/Subcortical Frontal Frontal Limbic Group Interaction Cingulate/Middle Frontal Inferior Frontal Inferior Frontal Inferior Frontal/Putamen Middle Frontal Superior Frontal Cingulate 32 47 9 13 8 9 23 Contrast of Interest t df p YD OD YD OD ¼ < > < YC OC YC OC 1.38 3.95 3.01 2.68 52 44 52 44 .174. 001 .004 .010 YW OW YW OW YW OW YW OW >  > <  < > < YM OM YM OM YM OM YM OM 2.95 1.66 2.42 3.97 2.41 3.11 2.04 2.88 52 44 52 44 49.89 44 52 44 .005 .105 .019 .001 .020 .003 .046 .006 Notes: Y: Younger; O: Older; W: Women; M: Men; D: Major Depressive Disorder; C: Healthy Comparison. Contrasts of interest for regions significant for the 3-way interaction are detailed in Supplemental Table 1. Effect sizes and post hoc testing were conducted with extracted data from clusters that met height by threshold whole-brain correction in the main analysis. a Significant MDD  gender  age group interaction, F(1, 92) ¼ 3.99, p ¼ 0.049. directionality of results (MDD > HC) for the remaining nonsignificant clusters. Within young men, the MDD sample exhibited significantly less activation than the HC group in three of the seven clusters, with the same directionality (MDD < HC) in the remaining clusters. In the older adults, activation was significantly less in MDD women compared with HC women in five of the seven clusters, with the same directionality (MDD < HC) in the remaining clusters. Within older men, Am J Geriatr Psychiatry -:-, - 2014 activation was significantly greater in MDD participants relative to HC participants in two of the seven clusters, with the same directionality (MDD > HC) in the remaining clusters. Because consistent patterns between groups were present across these clusters in the main and subordinate interactions, activation values for each of the seven clusters were averaged into one value for the purpose of illustration (Fig. 3; Supplementary Fig. 1 [available online] displays individual activation). 5 Age and Gender Effects in MDD FIGURE 2. Illustration of regions significant for an interaction between MDD status, gender, and age group during facial emotion processing, including the inferior frontal gyrus and putamen (A), inferior/middle frontal gyrus and cingulate (B), middle and superior frontal gyri (C), and middle frontal gyrus/cingulate and superior frontal gyrus (D). Note: The z coordinate is displayed in the Talairach system; the scale displays corresponding F values. DISCUSSION The present study demonstrated that gender and age are pertinent issues to consider among depressed adults and are potentially confounding to the interpretation of data related to emotion processing circuitry and function in MDD. Young women with MDD demonstrated hyperactive emotion processing circuits, particularly within the right prefrontal cortex, compared with HC young women. In contrast, young men with MDD demonstrated hypoactivation compared with young HC men in these regions. The opposite pattern of results was evident in older adults: Older women with MDD exhibited hypoactivation 6 compared with older HC women; older men with MDD exhibited hyperactivation compared with older HC men. It is important to note that when these subgroups defined by age and gender were combined into a single group, effects of MDD appeared negligible, as they were obscured by the underlying gender- and age-specific processes in MDD. The present study is the first to investigate the neural circuitry underlying emotion processing in older adults with early-onset, recurrent MDD. This population is relatively understudied in fMRI investigations, and there have been only two studies to date investigating emotion processing circuitry in individuals with late-life MDD,14,15 composed Am J Geriatr Psychiatry -:-, - 2014 Briceño et al. FIGURE 3. Illustration of the averaged extracted activation signal for the areas significant for the MDD status 3 gender 3 age group interaction. Top panel displays activation across each of the eight subgroups; the bottom two panels display these values separated by young and older subgroups. Note: D: Major Depressive Disorder (MDD); C: Healthy Comparison; W: Women; M: Men; Y: Younger, O: Older. Error bars indicate 1 SEM. primarily of women, and those with late-onset MDD. The present study found that in older adults with MDD, there were gender-specific effects that were opposite those observed in younger adults with MDD. Due to the small group sizes, especially given the known heterogeneity in late-life depression, these findings must be replicated. Longitudinal investigation of facial emotion processing with adults with MDD throughout the age spectrum would be informative in clarifying the nature of age-specific inversions of MDD by gender effects. Regions in the present study that were observed to vary in activation based upon MDD status, gender, and age group have been described by prior research as central to facial emotional processing, in the disease expression of MDD, and in topedown regulation of affect. The inferior frontal gyrus (IFG) has been shown to be important for facial emotion processing in healthy Am J Geriatr Psychiatry -:-, - 2014 adults25 and is aberrantly active during emotion processing in MDD.2 The IFG is involved in inhibitory control26 and in the regulation of negative affect.27 Our group2 reported that greater right-versus-left activation of the IFG in women with MDD was associated with increased accuracy in facial emotion processing, suggesting that the IFG plays a compensatory role in emotion processing among women with MDD. The middle and superior frontal gyri and cingulate are associated with facial emotion processing25,27 and the superior frontal gyrus and cingulate are involved in regulating negative affect.28 The putamen’s role in facial emotion processing may relate to preparing for and coordinating responses to emotional stimuli.8,29 These critical emotion processing and regulation areas demonstrate age and gender sensitivity, suggesting that even matching for age and gender may obscure gender- and age-specific processes. 7 Age and Gender Effects in MDD Hypothesized Mechanisms for Aberrant Activation Patterns by Gender and Age in MDD Young women with MDD exhibited hyperactivation in frontal, basal ganglia, and limbic areas compared with young HC women, reflecting a pattern that has been frequently reported2,3 but not previously revealed to be gender- and age-specific. Hypotheses for processes underlying hyperactive emotion processing circuitry include heightened sensitivity to affective material, difficulty regulating emotional response to affective stimuli, and increased recruitment of resources to support performance.2 Hypoactivation, less frequently reported in related prior studies,4 has been hypothesized to reflect dysregulated functional connectivity or underlying structural dysfunction. With regard to the compensation hypothesis, similar to hypotheses in the healthy aging literature,30 utilization of neural resources to support task performance may depend on the relationship between task demands and neural resource capacity. As task difficulty increases, recruitment of neural resources may also increase until it reaches its maximum capacity, after which activation plateaus or drops to reduced levels. Young women and older men with MDD may have been able to recruit additional neural resources to support performance, whereas young men and older women with MDD may have exceeded their capacity to support a highly demanding task, resulting in reduced activation. Differential recruitment patterns, particularly within the prefrontal system, may also reflect genderand age-differential emotion regulation strategies in healthy and MDD adults. For example, women are more likely than men to ruminate as a coping strategy, which has been hypothesized to be associated with the greater prevalence of MDD in women than in men.31 Rumination has been positively related to neural responses to negative facial emotions and negatively related to positive emotions in remitted MDD.32 The current study did not specifically investigate rumination or activation to negative emotions, although approximately two-thirds of the facial emotions in the present design were negative, indicating that activation differences were likely driven more by negative than positive or neutral emotional content. Men, in contrast, tend to be less emotionally focused7 and are more likely to cope through distraction than are women.33 Some evidence suggests a link between 8 disengagement coping and depressive symptoms.34 Alexithymia, more common in men and related to depressive symptoms,35 is related to reduced activity of neural regions supporting emotion processing.36 As such, different emotion regulation strategies associated with MDD in men (difficulty maintaining engagement with processing of affective material) and women (difficulty disengaging from processing of affective material) could result in discrepant patterns of activation supporting facial emotion processing. Aging is associated with a tendency to be less emotionally reactive and with a bias towards positive emotions.37 One study38 reported that increasing emotional stability in later life was associated with greater medial prefrontal control over negative emotional stimuli. Other studies13 have found that healthy older adults recruit greater frontal and less limbic regions during emotion processing compared to younger adults. One study39 reported distraction to be superior to reappraisal as an effective emotion regulation strategy in older adults with MDD. In older adults, women are more likely than men to utilize suppression as a regulation strategy, which is associated with depressive symptoms.40 Emotion suppression during memory encoding has been related to reduced functional connectivity of hippocampus and right dorsolateral prefrontal cortex.41 Altered connectivity of prefrontal regulatory regions during an emotion regulation task has been shown in older adults with genetic susceptibility to depression.42 As such, altered engagement of prefrontal regulatory regions may be partially explained by gender differences in emotion regulatory strategies in older adults. Chronic MDD is likely to interfere with these aspects of the healthy aging process and with emotion regulation strategies. General Effects of MDD When the overall effect of MDD status was evaluated without regard to age and gender in a sample balanced across gender and age, only one region differed between groups. Deactivation was present in the precuneus for the MDD group, with minimal differential use of this region for the HC group. The precuneus is part of the task-positive network.43 Emotional salience may interfere with the ability to engage the task-positive network in those with MDD, resulting in disengagement of this region during Am J Geriatr Psychiatry -:-, - 2014 Briceño et al. emotion processing. The disengagement of the lateral precuneus and potentially other parts of this network during emotion processing might explain why performance is typically diminished in MDD. Given the multiple interactions between gender, age, and MDD status, the lack of robust differences between the undifferentiated MDD and HC samples is not surprising. The critical implication of this finding is that evaluation of brain activation during emotion processing without accounting for and investigating separately the effects of age and gender may mask important network dysfunction pertinent to emotion processes in MDD. The inclusion of disproportionally few men and older adults in studies of emotion processing in MDD could be responsible for some proportion of the variability in prior literature. In addition, MDD is a heterogeneous illness. Disease severity, psychiatric comorbidities, medication status, and other clinical factors are potential complicating characteristics in interpreting findings within and across studies. Gender and age must also be seriously considered, and may be factors in the heterogeneity of findings with regard to the pathophysiology of MDD. Limitations Although the present sample size was substantially larger than is typical of neuroimaging studies, the number of participants within each cell was modest. These modest cell sizes reflect the challenge in conducting investigations of age and gender effects in MDD, as testing for these effects require large sample sizes. Replicating these findings with a larger sample would provide additional power to pursue more nuanced inquiry within this area. This MDD sample was composed of individuals with varied disease severity, illness chronicity, medication status, and psychiatric and medical comorbidities. Although this is a limitation with regard to sample homogeneity, this heterogeneity reflects the typical MDD expression within the population. There were very small cells for linear analyses of correlations between these characteristics, in addition to performance, and the regions of interest, and these results are not reported due to highly variable estimates of effect sizes. Future, larger, or more targeted studies will be needed to examine the potential role of these characteristics in the observed interactions. Our study also did not allow for hypothesis testing regarding the relative contributions of biological, learning/socialization, or interactive factors to the observed gender differences. For example, information regarding phase of menstrual cycle in women was not collected, which has been shown to influence emotion processing circuitry.44 CONCLUSIONS This study has critical implications for future research in the neurobiology of MDD. Gender and age must be carefully evaluated when conducting these studies through either balancing and evaluating gender and age effects within samples, or utilizing more homogenous samples. Research should also be directed towards better understanding the neurobiological mechanisms underlying gender- and agespecific activation differences in MDD. This research will provide enhanced understanding of the heterogeneity of MDD and better targeting of individualized treatments for depression. We gratefully acknowledge the contributions of Erich Avery and Brennan Haase in their assistance in data collection and processing. Funding for the present study included the following: Wayne State University Thesis/ Dissertation Research Support (EMB); National Alliance for Research on Schizophrenia and Depression (SAL); K-23 Career Development Award mh074459 (SAL); University of Michigan Department of Psychiatry Research Committee (SAL); University of Michigan Depression Center (SLW, SAL); University of Michigan fMRI Lab Pilot Scans (SAL, SLW); University of Michigan Claude D. Pepper Center (SLW); Alzheimer’s Disease Research Center, University of Michigan (SAL, SLW); MICHR Pilot Grant Fund (SLW); Rachel Upjohn Clinical Scholars Award (SAL, SLW); Phil F. Jenkins Research Fund (JKZ). The authors have no diclosures to report. References 1. Csukly G, Czobor P, Szily E, et al: Facial expression recognition in depressed subjects: the impact of intensity level and arousal dimension. J Nerv Ment Dis 2009; 197:98e103 Am J Geriatr Psychiatry -:-, - 2014 2. Briceño EM, Weisenbach SL, Rapport LJ, et al: Shifted inferior frontal laterality in women with major depressive disorder is related to emotion-processing deficits. Psychol Med 2013; 43:1433e1445 9 Age and Gender Effects in MDD 3. Surguladze S, Brammer MJ, Keedwell P, et al: A differential pattern of neural response toward sad versus happy facial expressions in major depressive disorder. Biol Psychiatry 2005; 57:201e209 4. Lawrence NS, Williams AM, Surguladze S, et al: Subcortical and ventral prefrontal cortical neural responses to facial expressions distinguish patients with bipolar disorder and major depression. Biol Psychiatry 2004; 55:578e587 5. Weissman MM, Bland RC, Canino GJ, et al: Cross-national epidemiology of major depression and bipolar disorder. JAMA 1996; 276:293e299 6. Wright SL, Langenecker SA, Deldin PJ, et al: Gender-specific disruptions in emotion processing in younger adults with depression. Depress Anxiety 2009; 26:182e189 7. Thayer JF, Johnsen BH: Gender differences in judgement of facial affect: a multivariate analysis of recognition errors. Scand J Psychol 2000; 41:243e246 8. Wager TD, Phan KL, Liberzon I, et al: Valence, gender, and lateralization of functional brain anatomy in emotion: a meta-analysis of findings from neuroimaging. Neuroimage 2003; 19:513e531 9. Weisenbach SL, Rapport LJ, Briceño EM, et al: Reduced emotion processing efficiency in healthy men relative to women. Soc Cogn Affect Neurosci 2014; 9:316e325 10. Calder AJ, Keane J, Manly T, et al: Facial expression recognition across the adult life span. Neuropsychologia 2003; 41:195e202 11. Sullivan S, Ruffman T: Emotion recognition deficits in the elderly. Int J Neurosci 2004; 114:403e432 12. Fischer H, Sandblom J, Gavazzeni J, et al: Age-differential patterns of brain activation during perception of angry faces. Neurosci Lett 2005; 386:99e104 13. Gunning-Dixon FM, Gur RC, Perkins AC, et al: Age-related differences in brain activation during emotional face processing. Neurobiol Aging 2003; 24:285e295 14. Brassen S, Kalisch R, Weber-Fahr W, et al: Ventromedial prefrontal cortex processing during emotional evaluation in lateonset depression: A longitudinal fMRI-study. Int J Psychol 2008; 43:265e265 15. Aizenstein HJ, Andreescu C, Edelman KL, et al: fMRI correlates of white matter hyperintensities in late-life depression. Am J Psychiatry 2011; 168:1075e1082 16. Wright SL, Langenecker, SA: Differential risk for emotion processing difficulties by gender and age in Major Depresive Disorder, in Women and Depression. Edited by Hernandez P, Alonso S. Hauppauge, NY, Nova Science Publishers, Inc., 2009, pp 183e216 17. Lavretsky H, Lesser IM, Wohl M, et al: Relationship of age, age of onset, and sex to depression in older adults. Am J Geriatr Psychiatry 1998; 6:248e256 18. Lavretsky H, Kurbanyan K, Ballmaier M, et al: Sex differences in brain structure in geriatric depression. Ame J Geriatr Psychiatry 2004; 12:653e657 19. First MB, Spitzer RL, Gibbon M, et al: Structured Clinical Interview for DSMeIV Axis I Disorders. Washington, DC, American Psychiatric Press, 1996 20. Hamilton M: A rating scale for depression. J Neurol Neurosurg Psychiatry 1960; 23:56e62 21. Langenecker SA, Bieliauskas LA, Rapport LJ, et al: Face emotion perception and executive functioning deficits in depression. J Clin Exper Neuropsychol 2005; 27:320e333 22. Rapport LJ, Friedman SL, Tzelepis A, et al: Experienced emotion and affect recognition in adult attention-deficit hyperactivity disorder. Neuropsychology 2002; 16:102e110 23. Langenecker SA, Weisenbach SL, Giordani B, et al: Impact of chronic hypercortisolemia on affective processing. Neuropharmacology 2012; 62:217e225 10 24. Ward BD: Simultaneous inference for fMRI data. Analysis of Functional Images (AFNI) Documentation, 2000. (online). Available at: http://afni.nimh.nih.gov/pub/dist/doc/manual/AlphaSim.pdf 25. Fusar-Poli P, Placentino A, Carletti F, et al: Laterality effect on emotional faces processing: ALE meta-analysis of evidence. Neurosci Lett 2009; 452:262e267 26. Nielson KA, Langenecker SA, Garavan H: Differences in the functional neuroanatomy of inhibitory control across the adult life span. Psychol Aging 2002; 17:56e71 27. Payer DE, Baicy K, Lieberman MD, et al: Overlapping neural substrates between intentional and incidental down-regulation of negative emotions. Emotion 2012; 12:229e235 28. Mak AKY, Hu ZG, Zhang JX, et al: Neural correlates of regulation of positive and negative emotions: an fMRI study. Neurosci Lett 2009; 457:101e106 29. Fusar-Poli P, Placentino A, Carletti F, et al: Functional atlas of emotional faces processing: a voxel-based meta-analysis of 105 functional magnetic resonance imaging studies. J Psychiatry Neurosci 2009; 34:418e432 30. Rypma B, D’Esposito M: The roles of prefrontal brain regions in components of working memory: effects of memory load and individual differences. Proc Natl Acad Sci USA 1999; 96:6558e6563 31. Nolen-Hoeksema S: Gender differences in depression. Curr Dir Psychol Sci 2001; 10:173e176 32. Thomas EJ, Elliott R, McKie S, et al: Interaction between a history of depression and rumination on neural response to emotional faces. Psychol Med 2011; 41:1845e1855 33. Hampel P, Petermann F: Perceived stress, coping, and adjustment in adolescents. J Adolesc Health 2006; 38:409e415 34. Tompkins TL, Hockett AR, Abraibesh N, et al: A closer look at co-rumination: gender, coping, peer functioning and internalizing/externalizing problems. J Adolesc 2011; 34:801e811 35. Honkalampi K, Saarinen P, Hintikka J, et al: Factors associated with alexithymia in patients suffering from depression. Psychother Psychosom 1999; 68:270e275 36. Lee B-T, Lee H-Y, Park S-A, et al: Neural substrates of affective face recognition in alexithymia: a functional magnetic resonance imaging study. Neuropsychobiol 2011; 63:119e124 37. Scheibe S, Carstensen LL: Emotional aging: recent findings and future trends. J Gerontol B Psychol Sci Soc Sci 2010; 65:135e144 38. Williams LM, Brown KJ, Palmer D, et al: The mellow years? Neural basis of improving emotional stability over age. J Neurosci 2006; 26:6422e6430 39. Smoski, MJ, LaBar KS, Steffens DC: Relative effectiveness of reappraisal and distraction in regulating emotion in late-life depression [Epub ahead of print September 7, 2013]. Am J Geriatr Psychiatry 40. Nolen-Hoeksema S, Aldao A: Gender and age differences in emotion regulation strategies and their relationship to depressive symptoms. Pers Individ Dif 2011; 51:704e708 41. Binder J, de Quervain DJF, Friese M, et al: Emotion suppression reduces hippocampal activity during successful memory encoding. Neuroimage 2012; 63:525e532 42. Waring JD, Etkin A, Hallmayer JF, et al: Connectivity underlying emotion conflict regulation in older adults with 5-HTTLPR short allele: a preliminary investigation [Epub ahead of print October 8, 2013]. Am J Geriatr Psychiatry 43. Cavanna AE, Trimble MR: The precuneus: a review of its functional anatomy and behavioural correlates. Brain 2006; 129: 564e583 44. Guapo VG, Graeff FG, Zani ACT, et al: Effects of sex hormone levels and phases of the menstrual cycle in the processing of emotional faces. Psychoneuroendocrinology 2009; 34:1087e1094 Am J Geriatr Psychiatry -:-, - 2014