# paper.pdf ## Metadata - PDFFormatVersion=1.4 - IsLinearized=false - IsAcroFormPresent=false - IsXFAPresent=false - IsCollectionPresent=false - IsSignaturesPresent=false - Producer=macOS Version 14.2.1 (Build 23C71) Quartz PDFContext - CreationDate=D:20240417153024Z00'00' - ModDate=D:20240417153024Z00'00' ## Contents ### Page 1 1Title Depression, identity functioning, and negative affective reactivity to sexual and genderminority concealment and outness in daily life in a predominantly female bi+ sampleAuthors Sienna R. Nielsen, Craig Rodriguez-Seijas, Sophia Choukas-Bradley, Aidan G.C.WrightAbstractSexual and gender minority (SGM) individuals experience higher rates of depression comparedto their peers. However, little is known about the day-to-day processes that link SGMexperiences to depression. In responses from SGM young adults (N=252), who predominantlyidentified as bi+ (sexual/romantic interest in multiple genders) cisgender women and nonbinaryindividuals assigned female at birth, we used multilevel structural equation modeling toinvestigate whether affective reactivity to SGM concealment and outness over an 8-dayambulatory assessment period (4342 observations) was associated with depression symptoms,and whether there was an indirect effect of momentary identity functioning. Concealment andoutness were differentially associated with self-concept clarity and SGM-identity positivity. Wedid not find a direct link between depression and affective reactivity, though person-centeredaffect linked depression and affective reactivity to concealment. This study indicates thatmomentary identity processes play an important role in the daily life experiences of visibility forbi+ individuals.Keywords: Ambulatory Assessment, Bi+ Sexual Identity, Identity Functioning, Minority Stress ### Page 2 2IntroductionYoung adults with minority sexual and gender identities (such as lesbian, gay, bisexual,transgender, and gender nonbinary) tend to report higher rates of depression and suicidality thantheir cisgender, heterosexual peers (Pakula & Shoveller, 2013; The Trevor Project, 2022). Theyalso tend to report greater identity impairment, including difficulty maintaining a coherent,stable, and positive self-schema. Within this group, individuals who sexually identify as bi, pan,or fluid and/or gender-identify as trans or nonbinary tend to additionally report greaterdepression, identity impairment, and identity invalidation (Feinstein & Dyar, 2017; Garr-Schultz& Gardner, 2021; Johnson et al., 2023; Price-Feeney et al., 2020; Ross et al., 2018).Theory suggests that increased risk of psychopathology in sexual and gender minority(SGM) individuals is caused by minority-specific stressors comprising identity-threatening socialcontexts, such as active or anticipated discrimination or invalidation (Hossain & Ferreira, 2019;Meyer, 2003; Parr & Howe, 2019; Pachankis et al., 2023; Reyes et al., 2020). According to thesetheories, identity-threatening events impede healthy identity functioning, promote distress, andpotentially conduce to depression. A growing body of research investigates the relationshipbetween minority stress experiences and health outcomes (Nicholas & Bresin, 2024). However,few studies have investigated the role of identity functioning by examining reactions to identitythreats in daily life. The present project examines the relation between identity-threatening socialcontexts, identity functioning, and negative affect among SGM young adults in daily life, usingecological momentary assessment (EMA). Further, we explore the relation between reactivity toidentity threats and depression symptoms. In doing so, we aim to clarify the relation betweenidentity functioning and depression for SGM young adults. We broadly seek to understand the ### Page 3 3relation between social context and depression symptoms to facilitate culturally competentconceptualization and treatment of psychopathology in historically marginalized populations.Depression and Identity Impairment in SGM individualsDepression is accompanied by an altered sense of self. Those experiencing depressionmay state that they do not know who they are anymore, or that they feel empty and estrangedfrom the true self that at one point offered direction, purpose, and positivity in their lives(Rhodes et al., 2019). This depressive identity impairment is clinically evidenced in hallmarkdepression symptoms such as low self-esteem and diminished enjoyment in thingscharacteristically important to the depressed individual (American Psychiatric Association,2013). Individuals experiencing depression express a lack of personal agency (Lebowitz & Ahn,2015) and a reduced sense of stable and coherent identity (Van Dijk et al., 2014; Lee-Flynn et al.,2011). Greater identity impairment is associated with more severe depression symptoms,including increased risk of suicidality (Sokol & Eisenheim, 2016; Sekowski et al., 2021; Wong etal., 2019). Although depression and identity have been repeatedly linked, the mechanisms of howidentity impairment relates to depression symptoms over time has received little empiricalattention. Further, it has been theorized that the relationship between identity and depression maybe crucial for understanding the high prevalence of depression among young adults ofmarginalized identities, such as SGM individuals navigating identity development and mentalhealth within identity-threatening environments (Alessi, 2014). The present project explores therole of identity functioning in reactivity to minority stressors, towards investigating SGM-specific stressors as identity threats in the maintenance of depression symptoms for SGMindividuals. ### Page 4 4SGM young adults are significantly more likely to experience depression and suicidalitythan their cisgender, heterosexual peers (Pakula & Shoveller, 2013; The Trevor Project, 2022).SGM individuals are also more likely to report identity impairment, or a sense that they do nothave a coherent, clear, and stable identity (Talley et al., 2011). Identity impairment has beenassociated with other forms of severe psychopathology that are also more prevalent among SGMindividuals, including alcohol and substance use disorders (Stepp et al., 2005), personalitydisorders (Bogaerts et al., 2021), and eating disorders (Stein & Corte, 2007). The high prevalenceof identity impairment and psychopathology in this community has historically contributed toclinician pathologization of SGM identities, to the detriment of patient care (Burckell &Goldfried, 2006; McCrone, 2018; Spengler et al., 2016). Efforts to reform psychotherapytreatment for SGM individuals suggest that the societal negative attitudes towards marginalizedgroups are the root of these higher rates of both identity impairment and psychopathologyexperienced by SGM individuals (Pachankis et al., 2023).Reactivity to Identity ThreatsTheories highlighting the role of minority stress experiences posit that SGM depression isdeveloped and maintained within an identity-threatening environment. In a world of anti-SGMlegislation (American Civil Liberties Union, 2024), prejudice, and violence, threat ofdiscrimination disrupts the maintenance of a stable, positive, and coherent SGM identity (Meyer,2003). For example, an individual identifying as a bisexual cisgender woman may go to schoolwith peers who have negative attitudes towards sexual minority students, with reputations forphysically harming sexual minority students. She may additionally be concerned about socialexclusion from both heterosexual and other sexual minority groups due to her bisexual identity(Hayfield et al., 2014). With this looming threat targeted towards her sexual identity, she is ### Page 5 5forced to consider that a part of her identity may have negative physical and social consequencesfor her.Research indicates exposure to identity-threatening contexts like these impedes SGMindividuals’ self-concept development (Hossain & Ferreira, 2019). She may feel negativelytowards this part of herself. Internalizing this SGM negativity may contribute to overall negativeaffect and lower self-concept clarity. She may feel uncertain about who she is or how she feelsabout herself day-to-day. Research suggests that internalized SGM negativity is associated withincreases in depressed and anxious mood the next day (Dyar, 2024), and that greater SGMnegativity in response to discrimination is associated with greater negative affect (Feinstein et al.,2022). Further, SGM individuals with higher internalized SGM negativity, lower self-conceptclarity, and higher ambiguity about their sexual identity report more severe depression symptoms(Feinstein et al., 2012). Over time, she may feel more ashamed, angry, or depressed.Additionally, she may cope with this threat by concealing her sexual identity from peers, but thethreat still lingers—promising negative consequences for her if her peers ever found out.Concealment to manage stigma may reduce the immediate risk of harm, while also threateningself-concept clarity, SGM-identity positivity, and sense of belonging in the moment (Bosson etal., 2012; Kiekens & Mereish, 2022; Newheiser et al., 2017). When she hears invalidating jokesor observes discrimination, she may similarly experience a momentary reduction in SGM-identity positivity and general self-concept clarity, as well as an increase in negative affect.According to prior research and theory, exposure to identity-threatening situations prohibitspositive integration of SGM identity into an overall cohesive sense of self, diminishing SGMself-concept clarity and self-esteem and thereby inducing negative affect. ### Page 6 6Prior research suggests that identity threats play a role in depression for SGM individuals.For example, gay men in China who reported concealing their sexual identity more often alsoreported higher depression severity and lower self-concept clarity (Li et al., 2021). Bisexualindividuals reporting more identity denial experiences, in which their bisexual identity wasquestioned or invalidated, reported higher depression severity and lower self-concept clarity andself-esteem (Garr-Schultz & Gardner, 2021). Gay and bisexual men indicating higherdiscrimination frequency in the prior six months also reported lower self-esteem and higher ratesof suicidal ideation (Huebner et al., 2004). In more recent EMA studies, SGM outness has beenassociated with greater identity affirmation and lower negative affect (Dyar et al., 2022), whilegreater concealment has been associated with higher negative affect (Kieken & Mereish, 2022).Exposure to stigma has been associated with greater negative affect (Dyar, 2024). Minoritystressors have also been associated with other symptoms of depression, including intensity ofsuicidal and nonsuicidal self-injurious ideation (Mereish et al., 2023). Although this researchlinks minority stress, identity impairment, and depression, it does not fully investigate the theorythat identity impairment associated with depression is heightened in the presence of minoritystressors for SGM individuals.Prior experimental research investigating reactions to identity threats demonstrates thatlab-based social identity manipulations are associated with changes in self-concept clarity andself-esteem. For example, lesbian, gay, and bisexual participants concealing their identities in animagined situation reported a decrease in self-esteem and belonging (Newheiser et al., 2017). Inanother experiment, participants were prompted to either describe instances when they actedconsistently or inconsistently with an adjective they used to described themselves, withparticipants in the inconsistent condition reporting lower self-concept clarity after the exercise ### Page 7 7(Boucher et al., 2016). Together, these results suggest that SGM identity functioning may beinfluenced by identity-threatening situations. However, this research is lab-based with relativelysmall sample sizes, raising questions about ecological validity and whether these findingsgeneralize to identity-threatening situations in their typical context of real daily life experiences.Identity Threats and DepressionEcological momentary assessment research assessing experiences in daily life has linkedidentity impairment and identity threats to depression symptoms, though no singular study hasinvestigated the relationships between identity impairment, discrimination, and depressionsymptoms. Linking identity and depression, lower 15-day average self-concept clarity wasassociated with higher depression symptoms in adolescents (Schwartz et al., 2011). On thewithin-person level, momentary decreases in self-concept clarity have been associated withfeeling sad and nervous (Ellison et al., 2022), as well as intent to self-harm (Scala et al., 2018).Linking identity threats and depression symptoms in 14-day EMA studies, SGMs reported higherratings of depressed mood (Livingston et al., 2020) and lower positive mood (Newberger, et al.,2022) on days they experienced identity-based discrimination. Over three weeks, SGMs reportedhigher negative affect following experiences of identity concealment at work (Mohr et al., 2019).Over 30 days, internalized SGM-negativity was associated with depressed and anxious mood(Dyar, 2024). Although prior research links depression symptoms to both identity impairmentand identity threats in daily life, further investigation is needed to assess the theorized role ofidentity impairment in the relationship between identity threats and negative affect in daily life.The theory outlined thus far highlights the within-person momentary relationshipbetween identity threats, identity impairment, and negative affect. However, this process may beassociated with depression through (1) depression severity predicting individual differences in ### Page 8 8reactivity to identity threats and (2) reactivity to identity threats predicting increasing depressionseverity over time. Individuals reporting higher depression symptoms may be more reactive tostressful identity-threatening situations. Given that depression has been linked to greateraffective and self-esteem reactivity to negative events (Cohen et al., 2005; Clasen et al., 2015;Parrish et al., 2011), we would expect individuals with higher depression severity to experiencegreater dips in self-esteem and self-concept clarity and greater increases in negative affect inreaction to identity-threatening events.Identity-threatening events may also contribute to worsening depression symptoms overtime. Individuals who experience greater identity impairment in reaction to identity-threateningevents may be more likely to develop depression symptoms. Prior longitudinal research suggeststhat reactivity to stressors precedes an increase in depression severity, establishing that reactivityto stressors may be an important predictor for the course of depression (Cohen et al., 2005;Parrish et al., 2011). Additionally, longitudinal research links identity impairment to increasingdepression progression, such that identity impairment in adolescence predicts an increase indepression symptoms three years later (Bogaerts et al., 2023). It is unclear whether a greaterimpact of stressors on identity functioning is associated with increasing depression symptoms.Given that reactivity to stressful events and greater identity impairment have been associatedwith worsening depression symptoms, we would expect that greater reactivity in identity andaffect to identity-threatening situations may be associated with increases in depression symptomsover time. Understanding the relationship between identity threats and depression progressionmay help identify individuals at increased risk for depression.Study Aims ### Page 9 9SGM young adults are at increased risk for depression. Although there are many theoriesinvolving the role of stressful, identity-threatening events such as discrimination and situationsrequiring identity concealment, the relationship between depression and reactivity to identity-threatening events in daily life has not been assessed. We explore the relationship between SGM-specific stressors, self-concept clarity, SGM-identity positivity, negative affect, and depressionsymptoms (Figure 1). Specific preregistered aims are as follows:Aim 1: We sought to examine whether individual differences in depression are related toindividual differences in affective reactivity to identity-threatening events in daily life (namely,SGM-invalidating events, higher identity-concealment, and lower identity-outness). Wehypothesized that on average, increases of momentary negative affect relative to one’s averagelevel would be associated with the experience of identity threats reported during randomlydelivered surveys. We anticipated that individuals would differ in the strength of this association,and that individuals reporting greater negative affective reactivity to identity threats over an 8-day period would report greater depression severity at baseline.Aim 2: We sought to investigate the role of identity functioning in the relation betweennegative affective reactivity to identity threats and depression. At the within-person level, wepredicted that momentary identity threats would be associated with decreases in momentary self-concept clarity and SGM-identity self-esteem relative to an individual’s average. We furtheranticipated that the momentary association between identity threats and momentary identityfunctioning will partially account for the momentary association between identity threats andnegative affect. We hypothesized that greater decreases in identity functioning and greaterincreases in negative affect would partially account for the relationship between minoritystressors and depression severity. We also hypothesized that average level of self-concept clarity ### Page 10 10and SGM-identity positivity would relate to the within-person association between identitythreats and negative affect. Given that prior research has linked other SGM identity variables towell-being, such as level of ambiguity about sexual orientation (Talley et al., 2015), weadditionally hypothesized that sexual self-concept ambiguity and identity centrality would berelated to a stronger association between minority stressors and negative affect.Aim 3: We also aimed to explore whether within-person increases in depressionsymptoms between the beginning and end of the EMA protocol are related to the average levelsand within-person associations of negative affect and identity functioning. We hypothesized thatgreater negative affective reactivity to identity threats would predict increases in depressionsymptoms from Time 1 to Time 2. Additionally, greater decreases in momentary identityfunctioning and greater negative affective reactivity to minority stressors would predictdepression symptoms from Time 1 to Time 2, controlling for the number of identity threats andaverage negative affect.Given that these relationships between affect, minority stress, and identity may representgeneralizable social and identity processes in young adulthood, we chose to open our sampling toall sexual and gender minority young adults ages 18 to 29 years. As detailed in the methodssection, our sample consisted of primarily White cisgender women and nonbinary individualsassigned female at birth who indicated sexual/romantic interest in multiple genders. Our resultsare considered in relation to this group.MethodsTransparency and OpennessThe current study was pre-registered, and the pre-registration can be found athttps://osf.io/cmjaz. All deviations from pre-registered procedures are noted in the document. All ### Page 11 11data, code, and output of the models estimated as part of this study are available athttps://osf.io/ngq7p/. We report how we determined our sample size, all data exclusions, allmanipulations, and all measures in the study.Between-person sample size was determined by prior work indicating that correlationcoefficients tend to stabilize at sample sizes around N=250 (Preugini & Schönbrodt, 2013). Awithin-person sampling rate of 10 times per day was designed to achieve the maximal number ofobservations over an 8-day study duration without overburdening participants. To ensure reliableestimates of within-person associations, participants who provided fewer than 10 momentaryobservations over the 8-day period were excluded from analyses.Power was estimated at both the within- and between-persons level using Monte Carlosimulations in Mplus using close approximations to the observed sample sizes (person N = 252,observation N = 4284). Assuming an intraclass correlation of .5, we have > .80 power to detecteffects ata = .05 for standardized regression coefficients as small as .05 at the within-personlevel and .20 at the between-person level. This includes cross-level interactions.ParticipantsParticipants were 252 18- to 29-year-old (mean=23.32; SD=3.13) sexual and/or genderminority young adults recruited in the Pittsburgh, PA area from December 2022 to June 2023.Recruitment sources included Pitt+Me (the University of Pittsburgh community researchrecruitment source), flyers, and social media advertisements by the University of PittsburghClinical and Translational Science Institute. Individuals were considered sexual minorityindividuals if they indicated a sexual identity other than exclusively heterosexual (i.e., gay,lesbian, bisexual, pansexual, queer, questioning/I am still figuring out who I am attracted to,and/or other), indicated same-gender attraction, or indication same-gender sexual experiences. ### Page 12 12Individuals were considered gender minority individuals if they indicated a non-cisgenderidentity by indicating a gender identity differing from their sex assigned at birth and/or aminority gender identity (i.e., transmasculine, transfeminine, genderqueer, gender fluid,nonbinary, and/or another identity). All participants indicated residence in the United States andownership of a smartphone compatible with the MetricWire survey app used for the EMAsurveys.Inclusion criteria for analyses involved completion of baseline questionnaires and at least10 EMA surveys over the EMA period. The final sample size was 252 participants with a total of4393 observations or 22% of randomly delivered surveys (17.43 /participant). This was a highlyburdensome sampling procedure that included both randomly delivered prompts and event-contingent surveys regarding social interactions. Therefore, the relatively modest response rateon the random prompts is understandable in the context of larger number of surveys responded toeach day. Participants only reported their level of outness and concealment when around others,therefore we used the 4342 observations (17.23/participant) that included these levels fromrandomly-delivered prompts. Sample demographics are provided in Tables 1 and 2. Participantswere primarily White (84%) and assigned female sex at birth (86%). The sample varied in sexualand gender identity, with bisexual (45%) and queer (41%) as the most frequently endorsedsexualities and cisgender woman (63%) and nonbinary (29%) as the most frequently endorsedgender identities. Notably, the sample majority were not current students (60%). Given thatparticipants were largely White, bi+, and assigned female at birth, results will be discussedprimarily as they relate to this group.Study Procedures ### Page 13 13All study procedures were approved by the University of Pittsburgh Institutional ReviewBoard. Participants were screened to determine eligibility before they were invited to reviewadditional study information. Enrolled participants completed baseline questionnaires aboutSGM experiences, depression, and personality. Participants who opted into the EMA portion ofthe study viewed a training video about the EMA protocol and completed a comprehension quizbefore beginning the EMA period via the MetricWire app lasting 8 days and consisting of (a)10/day semi-randomly delivered prompts and (b) a survey on the last day assessing depressionsymptoms over the past week. Compensation was proportional to the number of surveyscompleted, with a maximum of $100.MeasuresMomentary negative affect. Negative affect was assessed in EMA surveys with thefollowing terms on a sliding scale of 0 (not at all) to 100 (extremely): Nervous, Depressed,Rejected, Ashamed, Guilty. An overall momentary negative affect score was calculated as themean of all assessed negative affects. Reliability was good in our sample (a = 0.83).Past-hour SGM-specific invalidating events and other stressors. EMA surveys assessedpast-hour discriminatory experiences via a checklist including 9 items from the EverydayDiscrimination Scale (e.g., “You were treated with less respect than other people are;” Williamset al., 1997) as well as the following SGM-specific experiences: “People denied or invalidatedyour LGBTQ+ identity,” “People expressed discomfort or disapproval of the LGBTQ+experience,” and “People made jokes that involved microaggressions against the LGBTQ+community.” Participants then attributed the source of the discrimination to aspects of theiridentity (e.g., race/ethnicity, disability, mental health, gender, sexual orientation, transgenderidentity, or other). Instances of discrimination attributed to sexual orientation or transgender ### Page 14 14experience were coded as SGM-specific invalidating events. Discrimination experiences in EMAsurveys attributed exclusively to other identities besides SGM identity were coded as an “otherstressor.”Momentary SGM outness and concealment. Prior research suggests that concealment andoutness, or visibility, are distinct processes (Jackson & Mohr, 2016). If participants indicated thatthey were around other people at the time of the EMA survey, they were asked about their levelsof SGM-identity concealment and outness separately. SGM-identity concealment was assessed inEMA surveys with “I hid part of my LGBTQ+ identity from another person or other people” on ascale of 0 (not at all) to 10 (very much). SGM-identity disclosure was assessed with “The otherperson was aware of my LGBTQ+ identity” on a scale of 0 (not at all) to 10 (it was openlyacknowledged).Momentary identity functioning. Identity functioning was assessed as self-concept clarityand SGM-identity positivity via the semi-randomly delivered EMA surveys. Self-concept claritywas assessed using the 4-item Momentary Self-Concept Clarity Scale, which includes items suchas “Right now, I am wondering about what kind of person I really am” and “Right now, I’m notsure what I really believe in” (MSCCS; Ellison et al., 2022). Items are rated on a 4-point Likertscale from “strongly disagree” to “strongly agree.” As 3 of 4 the MSCCS scale items are keyedtowards identity dysfunction, rather than clarity, the one clarity item (“Right now, I have a clearsense of who I am”) was reverse scored when calculating MSCCS score, and the scale isconsidered a measure of identity impairment. Therefore, a lower MSCCS score reflects highermomentary self-concept clarity, and a higher score reflects higher momentary identityimpairment. SGM-identity positivity was assessed with the item “I like my LGBTQ+ identity” ### Page 15 15on a sliding scale of 0 (not at all) to 10 (very much). Reliability in our sample for the MSCCSwas good (a = 0.88).Depression. Depression symptoms were assessed using the Patient Health Questionnaire(PHQ), which asks frequency of depression symptoms over the past week on a 4-point Likertscale from 0 (not at all) to 3 (nearly every day). Depression symptoms were first assessed atbaseline, prior to the EMA period, using the 9-item Patient Health Questionnaire (PHQ-9;Kroenke et al., 1999) and additionally assessed 9 days later at the end of the EMA period viaMetricWire survey using the 8-item version (PHQ-8; Kroenke et al., 2009), which omits thequestion about suicidal ideation. Reliability in our sample was good for the PHQ-8 at Time 2 (a= 0.81) and the PHQ-9 at Time 1 (a = 0.86).Trait SGM identity functioning. We hypothesized that reactivity to SGM identity threatsmay be associated with characteristics of SGM identity, such as uncertainty about one’s sexualorientation, or sexual self-concept ambiguity. We preregistered assessing sexual self-conceptambiguity at baseline using the Sexual Self-Concept Ambiguity Scale (SSAS; Talley & Stevens,2017). As we prepared our data for analyses, we also thought to analyze the relationship betweenreactivity and SGM identity centrality, given that centrality of SGM identity has been indicatedas another influential identity characteristic (Hinton et al., 2022). We assessed SGM identitycentrality using 4 items from the Identity Centrality subscale of the Lesbian, Gay, and BisexualIdentity Scale (LGBIS, Mohr & Kendra, 2011), modified to be inclusive of gender minorities.Items were rated on a 6-point Likert scale from 1 (disagree strongly) to 6 (agree strongly) andincluded “My LGBTQ+ identity is a central part of my identity” and “Being an LGBTQ+ personis a very important aspect of my life.” Reliability in our sample was good for the SSAS (a =0.92) and the centrality subscale (a = 0.92). ### Page 16 16AnalysesAll analyses for the main aims were conducted in R studio version 4.2.2 (RStudio Team,2020) and Mplus version 8.6 (Muthén & Muthén, 2017). Statistical models were preregistered onOSF with the intention of investigating the relationship between depression and daily-lifereactivity to two types of identity-threatening experiences: (a) SGM-specific invalidating events,such as discrimination or identity-invalidating jokes about the LGBTQ+ experience, and (b)actively hiding SGM identity (higher concealment) or limiting open expression of SGM identity(lower outness). However, our sample reported very low rates of SGM-specific invalidatingevents over the 8-day EMA period, with only 67 participants out of 252 reporting at least oneevent (mean=3.49, SD=5.03 for those 67 participants). Potential reasons for this lowendorsement rate are described further in the discussion section. Given that we did not have thestatistical power to conduct our preregistered analyses on reactivity to SGM-specific events, wefocus here on analyses involving reactivity to situations involving modulating SGM concealmentand outness.Analyses involved momentary negative affect, self-concept clarity, SGM-identitypositivity, concealment, and outness, as well as baseline depression severity, sexual self-conceptambiguity, and SGM-identity centrality. We used multilevel structural equation modeling(MSEM) with Bayesian estimation for primary models, nesting momentary assessments withinindividuals to investigate within- and between-person associations among depression, outnessand concealment, affect, and identity functioning (Figure 1). For variables measured duringmomentary assessments, which vary both within and between persons, MSEM performs a latentpartitioning of the total variance in observed variables into their within-person and between-person portions of variance. The between-person variance reflects the average or individual ### Page 17 17differences in the endorsement of momentary variables, whereas the within-person variancereflects momentary deviations from an individual’s average. Thus, the within-person varianceeffectively person-centers the variables, allowing for estimation of association amongmomentary fluctuations among variables relative to the individual’s average levels. All within-person effects were estimated as random, allowing for the strength of within-person associationsto vary across individuals. In MSEM, random effects of within-person associations are treated asbetween-person latent variables (e.g., individual differences in the strength of association amongoutness and affect in the moment) and can be associated with other between-person variables. Inaddition, variables sampled at the between-person level only (e.g., PHQ-9 in these data) can beincluded in these models at the between-person level. Model convergence was determined usingthe potential scale reduction factor value of <1.1 (Gelman & Rubin, 1992). Coefficients arepresented along with their 95% credibility intervals (CIs).ResultsDescriptive statistics and bivariate correlations for within- and between-person effects arereported in Table 3. Within-person associations for our four primary models are presented inTable 4. Table 5 reports between-person associations from the combined MSEM includingbaseline variables of centrality, depression, and sexual self-concept ambiguity, as well asnegative affect, outness, and self-concept clarity and SGM positivity in daily life.Negative affective reactivityWe first investigated whether increases of momentary negative affect were associatedwith lower levels of SGM outness and higher SGM concealment. In this model we regressedperson-centered negative affect on person-centered SGM-identity concealment and outness at thewithin-person level. We found that individuals reported greater negative affect when engaged in ### Page 18 18greater concealment of their SGM identity (b=.063, p=.007, 95%CI[.02,.12]), as well as whenoutness was lower (b=-.056, p=.009, 95%CI[-.10,-.01]). These results support our hypothesesthat identity-invalidating social contexts demanding SGM-identity concealment or limitedoutness are associated with negative affect in daily life.We additionally investigated whether higher average levels of negative affect over the 8-day period were associated with higher average levels of concealment, lower outness, and greaternegative affect reactivity. Individuals reporting higher average levels of negative affect over the8-day period also reported higher average concealment (b=.33, p<.001, 95%CI[.20,.45]), but notlower average outness (b=-.15, p=.022, 95%CI[-.28,-.01]). Individuals reporting higher averagenegative affect also reported greater negative affective reactivity to concealment (b=.48, p=.001,95%CI[.19,.74]), but not to limited outness (b=-.07, p=.340, 95%CI[-.38,.28]). Thus, ourhypotheses were supported for concealment, but not outness.We then repeated the above model including baseline depression severity, as well assexual self-concept ambiguity (preregistered) and SGM identity centrality (added) to assesswhether negative affective reactivity to outness and concealment was associated with higherbaseline depression severity, greater sexual self-concept ambiguity, and greater SGM identitycentrality. Individuals with higher baseline depression severity also reported more negative affectover the 8-day EMA period (r=.36, p<.001, 95%CI[.24,.47]), as well as higher sexual self-concept ambiguity (r=.15, p=.010, 95%CI[.11,.37]). Depression severity was not significantlyassociated with SGM identity centrality (r=.05, p=.210, 95%CI[-.08,.18]).Contrary to our hypothesis, individuals with higher baseline depression severity did notreport more concealment (r=.11, p=.053, 95%CI[-.03,.24]), less outness (r=-.09, p=.113,95%CI[-.22,.05]), or higher negative affective reactivity to concealment (r=-.03, p=.410, ### Page 19 1995%CI[-.31,.26]) or outness (r=-.30, p=.024, 95%CI[-.60,-.01]) during the EMA period. Time 2depression severity was also not significantly associated with concealment, outness, or negativeaffective reactivity in this model (see supplemental analyses on OSF). Given that a primarysymptom of depression is heightened negative affect, we anticipated that depression symptomsassessed with the PHQ may be associated with concealment and outness through the indirecteffect of negative affect subsequently experienced in daily life. Follow-up analyses revealed asignificant indirect effect of negative affect on the relationship between depression scores andnegative affective reactivity to concealment (b=.02, p<.001, 95%CI[.01,.03]). There was noindirect effect of negative affect on the relationship between average outness and baselinedepression severity (b=-.01, p=.032, 95%CI[-.02,.00]).The baseline SGM identity variables also were not significantly associated with negativeaffective reactivity to concealment or disclosure. However, individuals reporting higher sexualself-concept ambiguity reported lower average levels of outness (r=-.24, p=.001, 95%CI[-.37,-.10]) and higher average concealment (r=.31, p<.001, 95%CI[.18,.43]). Individuals reportinghigher SGM identity centrality reported higher average levels of outness (r=.24, p<.001,95%CI[.11,.37]), though SGM identity centrality was not significantly associated with averagelevels of concealment (r=-.14, p=.025 95%CI[-.27,.00]).Testing the role of identity functioningSelf-Concept Clarity. We added self-concept clarity to the above model to examine therole of self-concept clarity in accounting for the association between concealment and disclosureand negative affect. We conducted two separate models to first estimate bivariate correlationsamong variables of interest and then estimate regressions to establish unique associations. We ### Page 20 20found that momentary self-concept clarity was significantly associated with momentary negativeaffect, such that when individuals reported lower levels of self-concept clarity, they also reportedhigher levels of negative affect (b=.26, p<.001, 95%CI[.22,.30]). As we hypothesized,momentary self-concept clarity was also significantly associated with momentary level ofconcealment, such that when individuals were concealing at higher levels, they also reportedlower levels of self-concept clarity (b=.06, p=.004, 95%CI[.01,.11]). However, momentary self-concept clarity was not significantly associated with momentary level of outness (b=-.03,p=.069, 95%CI[-.07,.01]). These results suggest that situations with higher levels of SGM-identity concealment are accompanied by diminished self-concept clarity, while modulatingoutness is not associated with self-concept clarity.At the between-person level, higher baseline depression severity associated significantlylower average levels of momentary self-concept clarity over the 8-day EMA period (r=.39,p<.001, 95%CI[.37,.50]). Lower average levels of momentary self-concept clarity were alsosignificantly associated with higher average negative affect (r=.46, p<.001, 95%CI[.34,.56]),higher average levels of concealment (r=.31, p<.001), 95%CI[.17,.43], lower average levels ofoutness (r=-.30, p<.001, 95%CI[-.42,-.16]), higher baseline sexual self-concept ambiguity(r=.47, p<.001, 95%CI[.36,.57]), and lower SGM identity centrality (r=-.20, p=.002, 95%CI[-.32,-.06]). Higher average negative affect was associated with greater self-concept clarityreactivity to concealment situations, such that individuals who reported higher average negativeaffect also reported a greater decrease in self-concept clarity associated with higher levels ofconcealment (r=.30, p=.004, 95%CI[.07,.50]). Higher baseline sexual self-concept ambiguitywas also associated with greater self-concept clarity reactivity to concealment (r=.33, p=.002, ### Page 21 2195%CI[.12,.52]). However, baseline depression severity was not associated with self-conceptreactivity to concealment level (r=-.01, p=.47, 95%CI[-.23,.22]).SGM-Identity Positivity. In separate but similar models, we then added SGM-identitypositivity to the negative affective reactivity model to examine the ability of SGM-identitypositivity to account for the association between concealment and outness and negative affect inthe moment. We conducted two separate models to first estimate bivariate correlations amongvariables of interest and then estimate regressions to establish unique associations. At the within-person level, we found that momentary SGM-identity positivity was significantly associated withmomentary negative affect, such that when individuals reported lower SGM-identity positivitythey also reported higher negative affect (b=-.24, p<.001, 95%CI[-.29,-.19]). Further, addingSGM positivity to the model explained the momentary relationships between negative affect andoutness and concealment, such that the within-person associations between negative affect andconcealment (b=.03, p=.080, 95%CI[-.01,.08]) and between negative affect and outness (b=-.04,p<.001, 95%CI[-.08,.00]) were no longer significant.As we hypothesized, momentary SGM-identity positivity was associated with momentaryoutness, such that individuals reported higher levels of positivity when reporting higher levels ofoutness (b=.08, p<.001, 95%CI[.04,.12]). However, we did not see a significant within-personassociation between SGM-identity positivity and concealment (b=-.05, p=.02, 95%CI[-.10,-.00]).These results suggest that situations with lower levels of SGM-identity outness are accompaniedby diminished SGM-identity positivity, while concealment occurs independently of SGM-identity positivity.At the between-person level, higher average levels of SGM-identity positivity wereassociated with lower average negative affect (r=-.21, p=.001, 95%CI[-.34,-.07]), higher average ### Page 22 22outness (r=.40, p<.001, 95%CI[.27,.51]), lower average concealment (r=-.38, p<.001, 95%CI[-.49,-.25]), lower sexual self-concept ambiguity (r=-.27, p<.001, 95%CI[-.39,-.15]), and higherSGM identity centrality (r=.57, p<.001, 95%CI[.47,.65]). Additionally, SGM identity centralitywas associated with greater SGM-identity positivity reactivity to negative affect (r=-.28, p=.005,95%CI[-.45,-.06]). Contrary to our hypothesis, baseline depression severity was not significantlyassociated with average SGM-identity positivity (r=-.04, p=.29, 95%CI[-.17,-.09]) or positivityreactivity to concealment (r=-.05, p=.34, 95%CI[-.26,.18]) or outness (r=-.03, p=.41, 95%CI[-.28,.23]).Combined Identity Model. We then entered both identity functioning variables (self-concept clarity and SGM-identity positivity) into one model to examine their simultaneous rolesin accounting for the association between concealment and outness and negative affect,controlling for baseline variables (Figure 2). Within this model, all within-person associationsremained significant except the relationship between momentary negative affect and momentaryconcealment and disclosure, suggesting that identity functioning partially accounted for therelationship between negative affect and outness in daily life (Table 5). Higher momentarynegative affect remained significantly associated with lower momentary self-concept clarity(b=.23, p<.001, 95%CI[.19,.27]) and lower momentary SGM-identity positivity (b=-.19, p<.001,95%CI[-.23,-.14]). Momentary SGM positivity remained significantly associated withmomentary outness (b=.09, p<.001, 95%CI[.04,.13]), while momentary self-concept clarityremained significantly associated with momentary concealment (b=.06, p=.004,95%CI[.02,.11]). Higher average negative affect was associated with higher average concealment(r=.32, p<.001, 95%CI[.18,.45]), lower SGM-identity positivity (r=-.21, p=.001, 95%CI[-.34,-.07]), lower self-concept clarity (r=.46, p<.001, 95%CI[.34,.56]), and higher self-concept clarity ### Page 23 23reactivity to concealment (r=.28, p=.006, 95%CI[.07,.48]). These results suggest that both self-concept clarity and SGM-identity positivity are associated with modulating SGM social identityin daily life.Baseline depression severity was not associated with reactivity in negative affect oridentity functioning to concealment or outness (Table 5). Higher baseline SGM-identitycentrality was associated with greater negative affective reactivity to SGM-identity positivity,such that decreases in SGM positivity were associated with greater increases in negative affect(r=-.39, p<.001, 95%CI[-.57,-.18]). Higher baseline sexual self-concept ambiguity wasassociated with greater self-concept clarity reactivity to concealment, such that increases inconcealment were associated with greater decreases in self-concept clarity (r=.33, p=.001,95%CI[.12,.52]). Therefore, individuals who considered their SGM status as more central totheir identity reported more negative affect with decreased SGM positivity, and those who weremore uncertain about their sexual orientation reported less self-concept clarity with increasedconcealment.Change in depression severityAs part of our preregistered exploratory analyses, we investigated whether within-personincreases in depression symptoms between the beginning (Time 1) and end of the EMA protocol(Time 2) are related to the levels and within-person associations of momentarily assessednegative affective reactivity to concealment and outness. Past-week PHQ-8 scores from bothtime points were used for comparison (i.e., the suicide question from PHQ-9 was not collected atfollow-up). Time 1 depression severity was strongly associated with Time 2 depression (b=.52,p<.001). PHQ at Time 2 was significantly associated with average levels of negative affect ### Page 24 24during the EMA period (r=.12, p=.001). However, neither depression at Time 1 nor Time 2 wasassociated with negative affective reactivity to concealment or outness.We then included self-concept clarity in the model to explore whether changes indepression severity are associated with self-concept clarity reactivity to concealment andoutness. Time 2 depression scores were associated with lower self-concept clarity (r=.13,p<.001). However, neither depression at Time 1 nor Time 2 was associated with self-conceptclarity reactivity to concealment or outness. When estimating the model with SGM-identitypositivity, average level of SGM-identity positivity was not associated with Time 2 depressionseverity. Given that depression levels at Time 1 and Time 2 were not significantly associatedwith reactivity in negative affect or identity functioning to concealment or outness, we did notrun models exploring change scores in depression severity.DiscussionSexual and gender minority individuals report higher rates of depression and identitydysfunction compared to their cisgender, heterosexual peers (Pakula & Shoveller, 2013). Giventhat situations that threaten or invalidate SGM identity may contribute to depression in SGMindividuals, we investigated the relations among depression symptoms and the reactions ofidentity dysfunction and negative affect to daily-life SGM experiences of modulating outness. Ineight days of smartphone survey responses from a sample of primarily bi+ individuals assignedfemale at birth, we found that negative affect reported in situations of higher concealment andlower outness was associated with momentary identity dysfunction. Specifically, momentaryhigher concealment (compared to the individual’s average), was associated with momentarylower self-concept clarity, while lower outness was associated with lower SGM-identitypositivity. These results suggest that situations eliciting SGM-identity concealment and ### Page 25 25discouraging outness limit identity functioning and promote negative affect. We did not finddirect links between depression symptoms and reactivity to modulating SGM social identity.However, we did find a significant indirect association, such that greater average negative affectlinked depression symptoms and negative affective reactivity to concealment. Given that greaternegative affect is a core feature of many forms of psychopathology, this process of identitythreats contributing to negative affect through identity dysfunction may partially explain higherrates of depression and other forms of psychopathology amongst SGM young adults.Our findings generally support our hypothesis that situations threatening SGM identitytend to promote negative affect by disrupting identity functioning. However, we did notanticipate that SGM experiences of concealment and outness would be differentially associatedwith two types of identity functioning. Higher levels of concealment were associated with lowermomentary general self-concept clarity, but not lower SGM-identity positivity. Lower levels ofoutness were associated with lower SGM-identity self-esteem, but not lower self-concept clarity.Therefore, concealment and outness may be associated with negative affect through differentidentity processes.Given that prior research indicates processes of concealment and outness may beparticularly multifaceted in bi+ populations and gender nonbinary populations (Feinstein et al.,2019; Johnson et al., 2023), our results may indicate processes specific to this sample and maynot be fully generalizable to exclusively gay/lesbian sexual minorities, cisgender male sexualminorities, or binary transgender gender minorities. As we discuss the relationships betweenoutness, identity functioning, and affect, we also consider the potential influence of experiencesspecific to bi+ individuals assigned female at birth investigated in prior research.Concealment ### Page 26 26We found that higher momentary concealment was associated with lower momentaryself-concept clarity, but not lower SGM-identity positivity. Prior research supports the linkbetween concealment and lower self-concept clarity. Concealing one’s identity may cause aninconsistency between one’s internal self-concept and social self-concept. Prior research suggeststhat highlighting inconsistencies in the self-concept reduces self-concept clarity (Boucher et al.,2015). Therefore, conflicting internal and social self-concepts may leave individuals feelingfragmented when concealing their SGM status, contributing to an overall decreased sense of self-concept clarity.Although the link between inconsistency and self-concept clarity may be a more generalprocess, prior research suggests that concealment may not be consistently associated with SGM-identity positivity. Rather, one’s motivation for concealment may indirectly link concealment andSGM positivity. SGM individuals report many motivations for concealment, sometimesconcealing out of fear for their own safety and sometimes concealing out of their ownuncertainty or discomfort towards their identity (Feinstein et al., 2020). Though we do not assessoutness motivations in this study, these prior findings suggest that whether an association isobserved between concealment and SGM positivity may be a function of the motivation forconcealment in the moment. For instance, concealing due to internalized negative attitudes mightbe associated with SGM positivity but concealing to minimize negative social consequenceswould not. Concealment motivation may also explain links between concealment and negativeaffect. For example, Feinstein and colleagues (2020) found that concealing bi+ identity due toconcerns about judgement or victimization was associated with higher levels of depression andanxiety, whereas concealing due to uncertainty about bi+ identity was not. Further research is ### Page 27 27needed to investigate the role of concealment motivation in the link between concealment andSGM-identity positivity.OutnessSGM identity outness was associated with SGM-identity positivity such that lessmomentary outness was associated with lower SGM-identity positivity, but outness was notassociated with self-concept clarity. It may be that choosing not to express SGM social identitycreates less dissonance between the internal and social self-concepts than choosing to activelyconceal identity, such that lower outness is not associated with lower self-concept clarity. Loweroutness may instead be in line with the individual’s preference in the moment to limit self-expression.The relationship between outness and SGM-identity positivity could be explained byseveral processes described in prior literature on SGM outness. First, how one feels about theirSGM identity in the moment contributes to level of outness. The role of internal attitudes aboutSGM identity in outness is supported by prior research on bi+ women, which also found thatgreater disclosure was associated with higher SGM identity centrality and lower SGM identityambiguity (Dyar et al., 2015). Individuals may choose to disclose less when they are alreadyexperiencing lower SGM-identity positivity and disclose more when they are experiencinghigher SGM-positivity. Second, the link between outness and SGM-identity positivity mayindicate the influence of social situations on both self-expression and positivity. Outness mayoccur less in social situations in which personal self-expression is less comfortable (e.g., withstrangers or acquaintances) or perceived as not immediately relevant (e.g., at work). Higherlevels of outness may occur in instances in which self-expression is more comfortable (e.g., withclose friends) or perceived as immediately relevant (e.g., around a potential romantic interest). ### Page 28 28Prior research on bi+ visibility suggests that, while outness is generally associated with positiveaffect and identity affirmation, these positive outcomes were dependent on a supportive socialcontext (Dyar et al., 2022). The social context may be particularly relevant to bi+ individuals,who report difficulty expressing their sexual minority status through appearance (Hayfield et al.,2013) and may need to express verbally. Further, the strategy for bi+ identity visibility (e.g.,direct communication, indirect communication, visual displays, etc.) may influence visibilityoutcomes (Feinstein et al., 2021). Given that we assessed momentary SGM identity outness with“The other person was aware of my LGTBQ+ identity,” our largely bi+ sample may beindicating outness primarily in contexts where verbal self-expression is comfortable and SGM-positivity may be higher (e.g., with close others). Therefore, it is impossible to determine fromour findings whether the link between momentary outness and SGM-identity positivity may bebetter explained by internal attitudes about SGM identity or the interpersonal situation.DepressionWe aimed to investigate the relationship between depression symptoms and SGMconcealment and outness processes in daily life. Although we did not find a direct associationbetween depression and reactivity to concealment and outness, we did find a significant indirecteffect between the two with average negative affect, such that individuals with higher averagenegative affect had greater negative affective reactivity to concealment and higher depressionseverity. However, the reason for this effect is unclear. Negative affect may be a more proximaloutcome to identity threats in daily life, whereas heightened depression symptoms are moredistally related to identity threats. The depression symptoms may be an outcome and/or riskfactor for the negative affect associated with identity threats. Alternatively, there may be amethods effect such that a true direct link between depression symptoms and affective reactivity ### Page 29 29to concealment and outness could not be detected when depression was assessed at baseline andconcealment, outness, and negative affect were assessed through EMA surveys. Further researchusing different assessments for negative affect and depression symptoms over different timescales is needed to assess the link between outness and depression.Another possibility is that identity threats are associated with other forms ofpsychopathology besides depression. Although low mood is a core feature of depression,negative affect is also a feature or outcome of many other psychiatric disorders (Stanton, 2020).There may be a direct link between reactivity to concealment and other disorders with higherprevalence amongst SGM individuals, such as substance use disorders and personality disorders.Further research investigating the relationship between concealment, outness, and other forms ofpsychopathology may help elucidate the role of identity threats in the manifestation andmaintenance of different disorders in SGM individuals.SGM Experiences of our sample: Bi+, assigned female at birthThe identities represented in our SGM sample may influence these results. Our sampleprimarily identified sexually as bi+, meaning they experience sexual/romantic interest in morethan one gender. Bi+ individuals report more concealment and less outness than gay and lesbianindividuals (Mohr et al., 2017; van der Star et al., 2019), and report identity invalidation fromboth heterosexual and gay/lesbian individuals (Feinstein & Dyar, 2017), suggesting that thesituations and processes of modulating outness in our sample may differ from those observed ina predominantly gay/lesbian sample. Bi+ individuals also report higher levels of general identitydysfunction (Talley et al., 2011) and sexual identity uncertainty (Talley & Stevens, 2017), as wellas more exposure to stereotypes about their identities being unstable and phasic fromheterosexual individuals and other sexual minority groups (Feinstein & Dyar, 2017). Further, ### Page 30 30outness may be associated with different health outcomes for bi+ individuals compared togay/lesbian individuals (Feinstein et al., 2019). Therefore, the relation between outness andidentity functioning observed in our sample may represent bi+-specific processes, or processesamplified among bi+ individuals.The gender makeup of our sample may also restrict interpretation of our results. Oursample was also predominantly assigned female at birth, with participants largely identifying asexclusively cisgender women or as gender nonbinary individuals. Concealment and outnessprocesses may be different for gender minority individuals compared to cisgender sexualminority individuals. There may be a greater risk of victimization for gender minorityindividuals, with more immediate negative physical and social consequences when one is openabout gender minority identity, making outness a potentially riskier process than concealment incertain contexts (Flynn & Smith, 2021). Identity threats impacting mental health for genderminority individuals may often take the form of identity invalidation, in which the genderidentity is directly challenged or ignored. For example, the process of “passing” as binary-gendered has been associated with distress for nonbinary individuals, suggesting that lowerdisclosure may be particularly associated with momentary identity dysfunction and negativeaffect for nonbinary individuals (Flynn & Smith, 2021). Nonbinary individuals report morefrequent exposure to identity invalidation compared to binary-trans individuals, with identityinvalidation partially explaining variance in psychopathology severity (Johnson et al., 2023).Therefore, the nonbinary individuals in our sample may experience more identity threatsprimarily involving identity invalidation than other SGM groups. Further research is needed toinvestigate differences in the impact of identity-invalidating events compared to other types ofidentity threats in different SGM populations. ### Page 31 31LimitationsGiven that our sample is primarily bi+ 18- to 29-year-old individuals assigned female atbirth, with a mix of cisgender and nonbinary gender identities, our results are contextualized inthe experiences of women and nonbinary bi+ individuals. To meet demands of statistical powerfor our analyses, we were unable to separate our sample into groups based on sexual orientationor gender identity. Further studies are needed to investigate whether these findings replicate insamples of other SGM groups across age cohorts. Given that racial and ethnic experiences mayinfluence reactions to SGM-specific experiences, further research is needed to examine the roleof intersectionality across racial and ethnic groups for SGM individuals (Fox et al., 2020; Jardaset al., 2023; Mereish et al., 2021).Due to low frequency of identity-invalidating events, we were unable to test ourhypotheses about the relationship between depression and discrimination. The low frequency inreporting may be a result of the sociopolitical climate in Pittsburgh, PA, a city with several largeuniversities and a more liberal climate than the surrounding area. Additionally, our assessment ofevents may have been lacking important manifestations of identity invalidating contexts orbehavior, leaving participants without the ability to share all they experienced. Our events weretypically direct and in person. However, future studies may benefit from including vicariouslyexperienced events both in-person and via media (e.g., “I saw a negative social media commentabout my identity”) to better capture the types of identity-invalidation faced by SGM youngadults today (Marciano & Antebi-Gruszka, 2022), and expand the sampling period beyond eightdays. Inclusion of events specific to the SGM sample may also be beneficial. For example, bi+samples and nonbinary samples may be more likely to endorse identity-threatening eventscharacterized by invalidation rather than discrimination in a more liberal area. ### Page 32 32Our assessment of depression symptoms was also limited. Although we found an indirectlink between depression and negative affective reactivity to concealment through average levelsof negative affect during the EMA period, we did not find a direct link between depression andnegative affective reactivity to concealment or outness. Additional research is needed to furtherinvestigate the relationship between depression and identity threats for SGM individuals. Futureresearch may investigate this process in a sample with higher depression severity and assess therelationship between depression, concealment, and outness over a longer time scale.ConclusionYoung adult sexual and gender minorities experience higher rates of psychopathologycompared to their peers (Pakula & Shoveller, 2013). In this project, we investigated whether day-to-day processes of reactivity to concealment and outness explain individual differences indepression severity. We found that concealment and outness were differentially associated withself-concept clarity and SGM-identity positivity. Our results support that situations promotingconcealment and limiting outness relate to negative affect through identity processes, functioningas identity threats. Further, concealment and outness may relate to negative affect throughdifferent identity processes. Contrary to our hypotheses, we did not find a direct link betweendepression symptoms and reactivity to modulating concealment and outness over 8 days. Furtherresearch is needed to explore the relationship between concealment, identity functioning, anddepression across different SGM groups and age cohorts, as well as SGM social identityexpression and other forms of psychopathology. This project provides a potential framework forthe relationship between SGM-specific experiences, identity dysfunction, and negative affect indaily life. ### Page 33 33Author Contributions: Conceptualization: S. Nielsen, A. Wright, C. Rodriguez-Seijas, and S.Choukas-Bradley; Methodology: S. Nielsen and A. Wright; Formal Analysis: S. Nielsen and A.Wright; Investigation: S. Nielsen; Writing—Original Draft Preparation: S. Nielsen; Writing—Review & Editing: A. Wright, C. Rodriguez-Seijas, and S. Choukas-Bradley.Conflicts of Interest: The authors declare that there were no conflicts of interest with respect tothe authorship or publication of this article.Funding: Funding for this project was provided by the University of Pittsburgh Clinical andTranslational Science Institute.ReferencesAlessi, E. J. (2014). A framework for incorporating minority stress theory into treatment withsexual minority clients. Journal of Gay & Lesbian Mental Health, 18(1), 47-66.American Psychiatric Association. (2013). Diagnostic and statistical manual of mentaldisorders (5th ed.). https://doi.org/10.1176/appi.books.9780890425596Bogaerts, A., Claes, L., Bastiaens, T., & Luyckx, K. (2021). Personality disordersymptomatology in Belgian emerging adults: Associations with identity processes andstatuses. European Journal of Developmental Psychology, 18(1), 75-95.Bogaerts, A., Claes, L., Raymaekers, K., Buelens, T., Bastiaens, T., & Luyckx, K. (2023).Trajectories of adaptive and disturbed identity dimensions in adolescence: developmentalassociations with self-esteem, resilience, symptoms of depression, and borderlinepersonality disorder features. Frontiers in Psychiatry, 14, 1125812.Bosson, J. K., Weaver, J. R., & Prewitt-Freilino, J. L. (2012). Concealing to belong, revealing to ### Page 34 34be known: Classification expectations and self-threats among persons with concealablestigmas. Self and Identity, 11(1), 114-135.Boucher, H. C., Bloch, T., & Pelletier, A. (2016). Fluid compensation following threats to self-concept clarity. Self and Identity, 15(2), 152-172.Burckell, L. A., & Goldfried, M. R. (2006). Therapist qualities preferred by sexual-minorityindividuals. Psychotherapy: Theory, research, practice, training, 43(1), 32.Clasen, P. C., Fisher, A. J., & Beevers, C. G. (2015). Mood-reactive self-esteem and depressionvulnerability: Person-specific symptom dynamics via smart phone assessment. PLoSOne, 10(7), e0129774.Cohen, L. H., Gunthert, K. C., Butler, A. C., O'Neill, S. C., & Tolpin, L. H. (2005). Dailyaffective reactivity as a prospective predictor of depressive symptoms. Journal ofpersonality, 73(6), 1687-1714.Dyar, C. (2023). Prospective examination of mechanisms linking minority stress andanxious/depressed affect at the event level: The roles of emotion regulation strategies andproximal minority stressors. Journal of Psychopathology and Clinical Science.Dyar, C., Feinstein, B. A., Bettin, E., & Davila, J. (2022). Bisexual+ visibility attempts:Associations with minority stress, affect, and substance use in a daily diarystudy. Psychology of sexual orientation and gender diversity, 9(2), 201.Dyar, C., Feinstein, B. A., & London, B. (2015). Mediators of differences between lesbians andbisexual women in sexual identity and minority stress. Psychology of Sexual Orientationand Gender Diversity, 2(1), 43.Ellison, W. D., Yun, J., Lupo, M. I., Lucas-Marinelli, A. K., Marshall, V. B., Matic, A. F. R., & ### Page 35 35Trahan, A. C. (2022). Development and initial validation of a scale to measuremomentary self-concept clarity. Self and Identity, 21(8), 995-1014.Feinstein, B. A., & Dyar, C. (2017). Bisexuality, minority stress, and health. Current SexualHealth Reports, 9, 42–49.Feinstein, B. A., Dyar, C., Li, D. H., Whitton, S. W., Newcomb, M. E., & Mustanski, B. (2019).The longitudinal associations between outness and health outcomes among gay/lesbianversus bisexual emerging adults. Archives of sexual behavior, 48, 1111-1126.Feinstein, B. A., Dyar, C., Milstone, J. S., Jabbour, J., & Davila, J. (2021). Use of differentstrategies to make one’s bisexual+ identity visible: Associations with dimensions ofidentity, minority stress, and health. Stigma and Health, 6(2), 184.Feinstein, B. A., Dyar, C., Poon, J. A., Goodman, F. R., & Davila, J. (2022). The affectiveconsequences of minority stress among bisexual, pansexual, and queer (bi+) adults: Adaily diary study. Behavior therapy, 53(4), 571-584.Feinstein, B. A., Goldfried, M. R., & Davila, J. (2012). The relationship between experiences ofdiscrimination and mental health among lesbians and gay men: An examination ofinternalized homonegativity and rejection sensitivity as potential mechanisms. Journal ofconsulting and clinical psychology, 80(5), 917.Feinstein, B. A., Xavier Hall, C. D., Dyar, C., & Davila, J. (2020). Motivations for sexualidentity concealment and their associations with mental health among bisexual,pansexual, queer, and fluid (bi+) individuals. Journal of Bisexuality, 20(3), 324-341.Flynn, S., & Smith, N. G. (2021). Interactions between blending and identity concealment:Effects on non-binary people’s distress and experiences of victimization. Plos one, 16(3),e0248970. ### Page 36 36Fox, K. R., Choukas-Bradley, S., Salk, R. H., Marshal, M. P., & Thoma, B. C. (2020). Mentalhealth among sexual and gender minority adolescents: Examining interactions with raceand ethnicity. Journal of Consulting and Clinical Psychology, 88(5), 402.Garr-Schultz, A., & Gardner, W. (2021). “It’s just a phase”: Identity denial experiences, self-concept clarity, and emotional well-being in bisexual individuals. Self and Identity, 20(4),528-544.Gelman, A., & Rubin, D. B. (1992). Inference from iterative simulation using multiplesequences. Statistical science, 7(4), 457-472.Hayfield, N., Clarke, V., & Halliwell, E. (2014). Bisexual women’s understandings of socialmarginalisation:‘The heterosexuals don’t understand us but nor do thelesbians’. Feminism & Psychology, 24(3), 352-372.Hayfield, N., Clarke, V., Halliwell, E., & Malson, H. (2013, September). Visible lesbians andinvisible bisexuals: Appearance and visual identities among bisexual women. In Women'sStudies International Forum (Vol. 40, pp. 172-182). Pergamon.Hinton, J. D., De la Piedad Garcia, X., Kaufmann, L. M., Koc, Y., & Anderson, J. R. (2022). Asystematic and meta-analytic review of identity centrality among LGBTQ groups: Anassessment of psychosocial correlates. The Journal of Sex Research, 59(5), 568-586.Hossain, F., & Ferreira, N. (2019). Impact of social context on the self-concept of gay andlesbian youth: A systematic review. Global Psychiatry, 2(1), 51-78.Huebner, D. M., Rebchook, G. M., & Kegeles, S. M. (2004). Experiences of harassment,discrimination, and physical violence among young gay and bisexual men. Americanjournal of public health, 94(7), 1200-1203.Jackson, S. D., & Mohr, J. J. (2016). Conceptualizing the closet: Differentiating stigma ### Page 37 37concealment and nondisclosure processes. Psychology of Sexual Orientation and GenderDiversity, 3(1), 80.Johnson, K. C., LeBlanc, A. J., Dolezal, C., Singh, A. A., & Bockting, W. O. (2023). Invalidationand mental health among nonbinary individuals. Psychology of Sexual Orientation andGender Diversity.Kiekens, W. J., & Mereish, E. H. (2022). The association between daily concealment and affectamong sexual and gender minority adolescents: The moderating role of family and peersupport. Journal of Adolescent Health, 70(4), 650-657.Kroenke, K., Spitzer, R. L., & Williams, J. B. (1999). Patient health questionnaire-9. CulturalDiversity and Ethnic Minority Psychology.Kroenke, K., Strine, T. W., Spitzer, R. L., Williams, J. B., Berry, J. T., & Mokdad, A. H. (2009).The PHQ-8 as a measure of current depression in the general population. Journal ofaffective disorders, 114(1-3), 163-173.Lee-Flynn, S. C., Pomaki, G., DeLongis, A., Biesanz, J. C., & Puterman, E. (2011). Dailycognitive appraisals, daily affect, and long-term depressive symptoms: The role of self-esteem and self-concept clarity in the stress process. Personality and Social PsychologyBulletin, 37(2), 255-268.Lebowitz, M. S., & Ahn, W. K. (2015). Emphasizing malleability in the biology of depression:Durable effects on perceived agency and prognostic pessimism. Behaviour research andtherapy, 71, 125-130.Li, F., Liao, J., Sun, X., Yang, T., Li, T., Wang, Y., & Mei, Y. (2021). Does self-concept clarity ### Page 38 38relate to depressive symptoms in Chinese gay men? The mediating effects of sexualorientation concealment and gay community connectedness. Sexuality Research andSocial Policy, 1-13.Livingston, N. A., Flentje, A., Brennan, J., Mereish, E. H., Reed, O., & Cochran, B. N. (2020).Real-time associations between discrimination and anxious and depressed mood amongsexual and gender minorities: The moderating effects of lifetime victimization andidentity concealment. Psychology of sexual orientation and gender diversity, 7(2), 132.Marciano, A., & Antebi-Gruszka, N. (2022). Offline and online discrimination and mentaldistress among lesbian, gay, and bisexual individuals: the moderating effect of LGBTQfacebook use. Media Psychology, 25(1), 27-50.McCrone, S. (2018). LGBT healthcare disparities, discrimination, and societal stigma: Themental and physical health risks related to sexual and/or gender minoritystatus. American Journal of Medical Research, 5(1), 91-96.Mereish, E. H., Peters, J. R., Brick, L. A., Killam, M. A., & Yen, S. (2023). A daily diary study ofminority stressors, suicidal ideation, nonsuicidal self-injury ideation, and affectivemechanisms among sexual and gender minority youth. Journal of psychopathology andclinical science, 132(4), 372.Mereish, E. H., Miranda Jr, R., Liu, Y., & Hawthorne, D. J. (2021). A daily diary study ofminority stress and negative and positive affect among racially diverse sexual minorityadolescents. Journal of Counseling Psychology, 68(6), 670.Meyer, I. H. (2003). Prejudice, social stress, and mental health in lesbian, gay, and bisexualpopulations: conceptual issues and research evidence. Psychological bulletin, 129(5),674. ### Page 39 39Mohr, J. J., Jackson, S. D., & Sheets, R. L. (2017). Sexual orientation self-presentation amongbisexual-identified women and men: Patterns and predictors. Archives of SexualBehavior, 46, 1465-1479.Mohr, J. J., & Kendra, M. S. (2011). Revision and extension of a multidimensional measure ofsexual minority identity: the Lesbian, Gay, and Bisexual Identity Scale. Journal ofcounseling psychology, 58(2), 234.Mohr, J. J., Markell, H. M., King, E. B., Jones, K. P., Peddie, C. I., & Kendra, M. S. (2019).Affective antecedents and consequences of revealing and concealing a lesbian, gay, orbisexual identity. Journal of Applied Psychology, 104(10), 1266.Muthén, L. K., & Muthén, B. O. (2017). Mplus: Statistical Analysis with Latent Variables: User'sGuide (Version 8). Los Angeles, CA: Authors.Newberger, N. G., Hinds, Z., Mahoney, C. T., Bryant, W. T., Herbitter, C., & Livingston, N. A.(2022). Real-time associations between discrimination, cannabis use, and mood amongsexual and gender minority individuals. Psychology of Addictive Behaviors.Newheiser, A. K., Barreto, M., & Tiemersma, J. (2017). People like me don't belong here:Identity concealment is associated with negative workplace experiences. Journal ofSocial Issues, 73(2), 341-358.Nicholas, J., & Bresin, K. (2024). Everyday Sexual and Gender Minority Stress and Health: ASystematic Review of Experience Sampling Studies. Archives of Sexual Behavior, 1-23.Pachankis, J. E., Hatzenbuehler, M. L., Klein, D. N., & Bränström, R. (2023). The role of shamein the sexual-orientation disparity in mental health: A prospective population-based studyof multimodal emotional reactions to stigma. Clinical Psychological Science,21677026231177714. ### Page 40 40Pachankis, J. E., Soulliard, Z. A., Morris, F., & van Dyk, I. S. (2023). A model for adaptingevidence-based interventions to be LGBQ-affirmative: Putting minority stress principlesand case conceptualization into clinical research and practice. Cognitive and BehavioralPractice, 30(1), 1-17.Pakula, B., & Shoveller, J. A. (2013). Sexual orientation and self-reported mood disorderdiagnosis among Canadian adults. BMC public health, 13(1), 1-7.Parr, N. J., & Howe, B. G. (2019). Heterogeneity of transgender identity nonaffirmationmicroaggressions and their association with depression symptoms and suicidality amongtransgender persons. Psychology of Sexual Orientation and Gender Diversity, 6(4), 461.Parrish, B. P., Cohen, L. H., & Laurenceau, J. P. (2011). Prospective relationship betweennegative affective reactivity to daily stress and depressive symptoms. Journal of Socialand Clinical Psychology, 30(3), 270-296.Price-Feeney, M., Green, A. E., & Dorison, S. (2020). Understanding the mental health oftransgender and nonbinary youth. Journal of Adolescent Health, 66(6), 684-690.Reyes, M. E. S., Davis, R. D., Yapcengco, F. L., Bordeos, C. M. M., Gesmundo, S. C., & Torres,J. K. M. (2020). Perceived Parental Acceptance, Transgender Congruence, andPsychological Well-Being of Filipino Transgender Individuals. North American Journalof Psychology, 22(1).Rhodes, J. E., Hackney, S. J., & Smith, J. A. (2019). Emptiness, engulfment, and life struggle: aninterpretative phenomenological analysis of chronic depression. Journal of ConstructivistPsychology, 32(4), 390-407.Ross, L. E., Salway, T., Tarasoff, L. A., MacKay, J. M., Hawkins, B. W., & Fehr, C. P. (2018). ### Page 41 41Prevalence of depression and anxiety among bisexual people compared to gay, lesbian,and heterosexual individuals: A systematic review and meta-analysis. The Journal of SexResearch, 55(4-5), 435-456.RStudio Team (2020). RStudio: Integrated Development for R.Schwartz, S. J., Klimstra, T. A., Luyckx, K., Hale III, W. W., Frijns, T., Oosterwegel, A., ... &Meeus, W. H. (2011). Daily dynamics of personal identity and self–conceptclarity. European Journal of Personality, 25(5), 373-385.Sekowski, M., Gambin, M., & Sharp, C. (2021). The relations between identity disturbances,borderline features, internalizing disorders, and suicidality in inpatientadolescents. Journal of Personality Disorders, 35(Supplement B), 29-47.Sokol, Y., & Eisenheim, E. (2016). The relationship between continuous identity disturbances,negative mood, and suicidal ideation. The primary care companion for CNSdisorders, 18(1), 26150.Spengler, E. S., Miller, D. J., & Spengler, P. M. (2016). Microaggressions: Clinical errors withsexual minority clients. Psychotherapy, 53(3), 360.Stanton, K. (2020). Increasing diagnostic emphasis on negative affective dysfunction: Potentiallynegative consequences for psychiatric classification and diagnosis. ClinicalPsychological Science, 8(3), 584-589.Stein, K. F., & Corte, C. (2007). Identity impairment and the eating disorders: Content andorganization of the self‐concept in women with anorexia nervosa and bulimianervosa. European Eating Disorders Review: The Professional Journal of the EatingDisorders Association, 15(1), 58-69. ### Page 42 42Stepp, S. D., Trull, T. J., & Sher, K. J. (2005). Borderline personality features predict alcohol useproblems. Journal of Personality Disorders, 19(6), 711-722.Talley, A. E., & Stevens, J. E. (2017). Sexual orientation self-concept ambiguity: Scaleadaptation and validation. Assessment, 24(5), 632-645.Talley, A. E., Tomko, R. L., Littlefield, A. K., Trull, T. J., & Sher, K. J. (2011). The influence ofgeneral identity disturbance on reports of lifetime substance use disorders and relatedoutcomes among sexual minority adults with a history of substance use. Psychology ofAddictive Behaviors, 25(3), 530.van der Star, A., Pachankis, J. E., & Bränström, R. (2019). Sexual orientation openness anddepression symptoms: A population-based study. Psychology of Sexual Orientation andGender Diversity, 6(3), 369.Van Dijk, M. P., Branje, S., Keijsers, L., Hawk, S. T., Hale, W. W., & Meeus, W. (2014). Self-concept clarity across adolescence: Longitudinal associations with open communicationwith parents and internalizing symptoms. Journal of youth and adolescence, 43, 1861-1876.Williams, D.R., Yu, Y., Jackson, J.S., and Anderson, N.B. “Racial Differences in Physical andMental Health: Socioeconomic Status, Stress, and Discrimination.” Journal of HealthPsychology. 1997; 2(3):335-351.Wong, A. E., Dirghangi, S. R., & Hart, S. R. (2019). Self-concept clarity mediates the effects ofadverse childhood experiences on adult suicide behavior, depression, loneliness,perceived stress, and life distress. Self and Identity, 18(3), 247-266. ### Page 43 43Figures/TablesDepiAiBiCiThreatiNAiIdeniBetween-Person ModelWithin-Person ModelThreat#Iden#NA#BiCiAiFigure 1. Diagram of the general multilevel structural equation model estimatedfor our main aims. Iden = identity; NA = negative affect; Threat = identity threat.Variables with a subscript of ti vary across time and individuals; variables with ani subscript vary across individuals only. Single-headed arrows representregression paths. At the within-person level, solid circles along regression pathsrepresent random effects, which are also depicted as between-person latentvariables. All random intercepts and slopes at the between-person level wereallowed to freely covary. Covariates (SSCA and centrality) are not depicted. ### Page 44 44ConcealOutnessMSCCSGM PNA.06 .23-.19.09Figure 2. Diagram of the within-person associations in the combined multilevelstructural equation model. Conceal = momentary concealment; Outness =momentary outness; MSCC = momentary self-concept clarity; SGM P = SGM-identity positivity; NA = negative affect. Single-headed arrows representregression paths. Solid arrows represent significant associations (p<.01). Dashedlines represent nonsignificant associations. Estimates are provided for significantassociations. Covariates included for between-person associations (SSCA,depression, and centrality) are not depicted.-.23-.02.04-.04-.05 ### Page 45 45Table 1. Sexual and gender minority categoriesBi+ SM Mono SM Unspecified SMCisgender woman 78 27 19Cisgender man 11 14 0GM-AFAB 45 17 29GM-AMAB 3 2 4Total 137 60 52Cisgender = Indicated male sex at birth and exclusively “man” gender identity or female sex at birth andexclusively “woman” gender identity; GM = Gender minority (not cisgendered); AFAB = Assigned female atbirth; AMAB = Assigned male at birth. SM = Sexual minority. None of the participants identified as exclusivelyheterosexual or exclusively questioning. Bi+ SM = indicated identifying as bisexual or pansexual; Mono SM =indicated identifying as gay or lesbian, but not bisexual, pansexual, or heterosexual. Unspecified SM = indicatedsexual orientation only through terms without gender(s) implied, e.g., " queer" or "other". This category mayrepresent individuals who are sexually fluid, do not experience sexual attraction, or do not identify with genderedsexuality terms. ### Page 46 46Table 2. DemographicsVariable Number %Race and EthnicityAmerican Indian or Alaska Native 2 0.79Asian 29 11.51Black or African American 21 8.33Native Hawaiian or Other PacificIslander 3 1.19White 212 84.13Other race (LatinX, MiddleEastern) 6 2.38Multiracial 24 9.52Hispanic 15 5.95Sex at BirthMale at birth 34 13.49Female at birth 217 86.11Intersex 0 0.00GenderMan 41 16.27Woman 161 63.89Transmasculine 26 10.32Transfeminine 3 1.19Genderqueer 32 12.70Gender fluid 20 7.94Nonbinary 77 30.56Other gender identity 8 3.17Sexual OrientationHeterosexual 7 2.78Gay 30 11.90Lesbian 47 18.65Bisexual 115 45.63Pansexual 44 17.46Queer 108 42.86Questioning 19 7.54Prefer not to answer 0Other sexual orientation (asexual,demisexual) 19 7.54Employment StatusStudent 100 39.68Employed, Full-time 91 36.11Employed, Part-time 38 15.08Unemployed, looking for work 12 4.76Self-employed 5 1.98Unemployed, not looking for work 3 1.19Receiving disability 2 0.79Homemaker 1 0.40 ### Page 47 47Table 3. Between-person correlations and means for baseline variables and person-centered 8-day averages for momentary variables, andwithin-person correlations for momentary variables.Mean, SD PHQ-9 Centrality SSCA NA MSCC SGM P Outness ConcealPHQ-9 1.11 (.66)Centrality 4.29 (1.33) .05[-.07,.18]SSCA 1.99 (.66) .15[.03,.27]-.13[-.25,-.01]NA 1.29 (1.09) .34[.22,.44]-.05[-.17,.07].22[.10,.34].35[.33, .38]-.19[-.22, -.16]-.09[-.12,-.06].10[.07, .13]MSCC 2.05 (.82) .37[.26,.48]-.20[-.31,-.07].46[.36,.56].44[.33,.53]-.25[-.28, -.22]-.08[-.11,-.05].10[.07, .13]SGM P 7.77 (2.28) -.03[-.16,.09].56[.47,.64]-.26[-.38,-.14]-.19[-.30,.06]-.25[-.37,-.13].12[.09, .15]-.09[-.12,-.06]Outness 6.58 (2.40) -.08[-.20,.05].20[.08, .32]-.19[-.31,-.07]-.15[-.27,-.02]-.26[-.37,-.14].33[.22,.44]-.59[-.61,-.57]Conceal 2.02 (1.96) .09[-.04,.21]-.13[-.25,-.01].28[.17,.39].28[.17,.39].26[.14,.37]-.33[-.44,-.22]-.64[-.71,-.56]Between-person associations shown under the diagonal; within-person associations shown over the diagonal. NA = negative affect; SGMP = SGM positivity; MSCC = momentary self-concept clarity; PHQ-9 = Patient Health Questionnaire (depression symptoms); SSCA =sexual self-concept ambiguity. Bold = p<.01; 95% CI reported. ### Page 48 48Table 4. Within-person associations with momentary outness and concealment predicting negative affectand identity functioning.NA Model MSCC Model SGM P Model Combined ModelConceal Outness Conceal Outness Conceal Outness Conceal OutnessNA.06[.01,.11]-.05[-.09,.00].04[.00,.08]-.06[-.10,-.02].03[-.01,.08]-.04[-.08,.00].04[-.00,.08]-.04[-.08,-.00]MSCC -- --.06[.01,.11]-.03[-.07,.01] -- --.06[.02,.11]-.03[-.08,.02]SGM P -- -- -- ---.05[-.10,-.00].08[.04,.12]-.04[-.09,.01].09[.04, .13]NA = negative affect; SGM P = SGM positivity; MSCC = momentary self-concept clarity. Bold = p<.01;95% CI reported.All models include between-person associations with baseline variables of PHQ-9, SSCA, and Centrality.Table 5. Between-person associations from the combined model including baseline variablesand negative affect, identity, and concealment and outness in daily life.PHQ-9 Centrality SSCANA .36 [.23, .47] -.05 [-.19, .09] .25 [.12, .38]Conceal .11 [-.03, .24] -.15 [-.29, -.01] .30 [.17, .43]Outness -.08 [-.22, .06] .25 [.11, .38] -.24 [-.37, -.10]MSCC .39 [.27, .50] -.20 [-.33, -.06] .47 [.36, .57]SGM P -.04 [-.18, .10] .57 [.47, .66] -.28 [-.40, -.15]Conceal à NA -.06 [-.33, .20] -.10 [-.37, .19] -.17 [-.41, .09]Outness à NA -.27 [-.51, -.01] -.09 [-.35, .18] -.04 [-.22, .30]MSCC à NA .13 [-.06, .30] -.06 [-.26, .14] -.01 [-.20, .18]SGM S-E à NA -.10 [-.32, .12] -.39 [-.57, -.18] .05 [-.17, .27]Conceal à MSCC .02 [-.20, .23] -.04 [-.25, .18] .33 [.12, .52]Outness à MSCC -.09 [-.35, .17] -.08 [-.33, .18] .12 [-.15, .38]Conceal à SGM P -.04 [-.26, .19] -.13 [-.35, .10] -.13 [-.34, .09]Outness àSGM P .00 [-.27, .27] -.15 [-.40, .11] .22 [-.05, .47]NA = negative affect; SGM P = SGM positivity; MSCC = momentary self-concept clarity;PHQ-9 = Patient Health Questionnaire (depression symptoms); SSCA = sexual self-conceptambiguity. Bold = p<.01; 95% CI reported.