Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Dec 1.
Published in final edited form as: J Anxiety Disord. 2021 Sep 1;84:102474. doi: 10.1016/j.janxdis.2021.102474

Are people with social anxiety disorder happier alone?

Fallon R Goodman a, Ruba Rum a, Gabriella Silva a, Todd B Kashdan b
PMCID: PMC9199593  NIHMSID: NIHMS1739920  PMID: 34509949

Abstract

Quality contact with other people serves as a reliable mood enhancement strategy. We wondered if the emotional benefits of socializing are present even for those with a psychological disorder defined by social distress and avoidance: social anxiety disorder (SAD). We conducted two ecological momentary assessment (EMA) studies and analyzed 7,243 total surveys. In both studies, community adults diagnosed with SAD and healthy controls received five surveys each day for two weeks. Consistent with research on positivity deficits in SAD, between-person analyses in both studies suggest that, on average, participants with SAD reported lower positive and higher negative affect in social and non-social situations than healthy controls. Within-person analyses, however, revealed that in both studies participants with SAD and healthy controls reported higher positive affect when with others than when alone; no differences were found for negative affect. The difference in positive affect between social and nonsocial situations was smaller for participants with SAD in Study 1, suggesting that people with SAD may experience diminished reward responding when socializing. Our results suggest that even those with a mental illness defined by interpersonal distress can and do derive positive emotions from social interactions.

Keywords: social anxiety, happiness, positive affect, negative affect, experience-sampling

1. Introduction

Social interactions are a primary source of anxiety for people with social anxiety disorder (SAD). People with SAD worry that they will be negatively evaluated and that their perceived flaws will result in rejection (Moscovitch, 2009). To minimize opportunities for scrutiny, people with SAD avoid social interactions or tolerate them with considerable distress (Moscovitch et al., 2013). Evidence suggests that socially anxious adults experience lower positive affect (PA) and higher negative affect (NA) during social interactions compared to less anxious people (Goodman et al., 2018; Vittengl & Holt, 1998). However, this research has relied on between-person comparisons, where someone high in social anxiety (or with SAD) is compared to someone low in social anxiety (or absent of mental illness). While we can conclude that socially anxious people have lower PA and higher NA when socializing than non-anxious peers, we know little about if/how they experience different emotions in different situations.

A within-person comparison can assess if a socially anxious person experiences higher or lower PA and NA when in a social situation than when alone. Decades of social psychology research suggest that social relationships are critical for well-being (see Diener et al., 2018), and people feel happier when they are interacting with others than when alone (e.g., Sun et al., 2019; Mehl et al., 2010), even during brief exchanges with a stranger (e.g., short conversation with a barista; Sandstrom & Dunn, 2014). The paradox of social anxiety is that fear of social interactions coexists with a desire for social connection (Alden & Taylor, 2010). Thus, we wondered if people with SAD, who long for social connection yet are typically devoid of it, experience similar emotional benefits of social interactions as non-anxious peers. To date, this question has been examined in a small number of undiagnosed college student convenience samples. Across three studies, people with higher social anxiety symptoms reported similar levels of PA and NA in social situations as when they were alone (Brown et al., 2007; Geyer et al., 2018; Kashdan & Collins 2010). Thus, despite concerns and anxiety about socializing, socially anxious individuals did not report elevated PA or NA when socializing.

In this research program, we build off this work and synthesize new literature suggesting that people with SAD may report ever higher PA in social situations than when they are alone. Using clinical samples of adults recruited from the community, we conducted two ecological momentary assessment (EMA) studies of adults diagnosed with SAD and psychologically healthy control groups. We aimed to replicate work on emotion deficits in SAD with traditional between-person analyses that compare average affect between people with SAD and healthy controls. We then conducted within-person analyses that compare how people with SAD experience affect in social situations compared to when they are alone. We also examined how participants’ emotional experiences differed based on the persons with whom they interacted.

1.1. Social Anxiety and Affect During Socializing

We propose three reasons why people with SAD may report higher PA in social situations than when they are alone.

1.1.1. #1) People with SAD Desire Social Connection

Psychologists have suggested that social interactions facilitate positive emotions because they fulfill a fundamental need to develop close connections with other people (Baumeister & Leary, 1995). There is no evidence that people with SAD are devoid of this desire; to the contrary, people with SAD desire strong, intimate social relationships to the same degree as healthy adults (Goodman, Kashdan et al., 2019). People with elevated social anxiety may even have a stronger need for belongingness than those with lower social anxiety (e.g., Lavigne et al., 2011). In this way, social anxiety can be distinguished from social anhedonia. Social anhedonia is characterized by a diminished need to belong and lack of reward from social interactions, whereas social anxiety is characterized by a normative need to belong and desire to connect with others but fear of doing so (Silvia & Kwapil, 2011). Empirical work supports their distinction; the two constructs are moderately correlated (Brown et al., 2007), and very few people score highly on both (Silva & Kwapil, 2011). Taken together, because people with SAD desire social connection, they may experience elevated PA that social situations elicit—even in the presence of anxiety or other negative emotions.

1.1.2. #2) Negative Emotions Do Not Preclude Positive Emotions

PA and NA tend to be inversely correlated, but the experience of any negative emotion does not preclude the experience of any positive emotion (e.g., Tellegen et al., 1999). People with SAD, in part by definition, will feel distressed in certain social interactions, as they elicit self-presentation concerns and fear of rejection (Moscovitch et al., 2013). As a result, people high in social anxiety tend to experience elevated anxiety and other negative emotions during social interactions (e.g., Lazarus & Shahar, 2018; Oren-Yagoda et al., 2021). Nonetheless, they may also experience positive emotions, such as feeling excited for an opportunity to develop a friendship, interested in what someone is saying, or proud for speaking up. Social interactions tend to elicit a mix of positive and negative emotions (Berry & Hansen, 1996). Emotions elicited during social interactions may be analogous to the mixed emotions elicited during pursuit of important life goals (e.g., anxiety about failing; excitement about an opportunity for success). For people with SAD, social interactions may facilitate both negative emotions (borne out of social fears and anxiety) and positive emotions (borne out of social connectedness).

1.1.3. #3) Social Interactions Might Facilitate a “Mood Brightening” Effect

Despite persistent and chronic deficits in PA (Kashdan et al., 2011), people with SAD may derive enhanced emotional benefits from social experiences. The “mood brightening” effect is a phenomenon in which people with mood difficulties experience greater emotional benefits from positive experiences than people without mood difficulties. Research on the mood brightening effect has primarily been studied in depression. Following positive life events, people with depression experience greater decreases in NA (Bylsma et al., 2011; Rottenberg, 2017) and greater increases in PA (Peeters et al., 2003) than non-depressed people. Depression and SAD have similar mood profiles, characterized by high NA and low PA (Brown et al., 1998; Chorpita et al., 2000; Hughes et al., 2006), suggesting that people with SAD may also experience mood brightening effects.

A small set of experience-sampling studies offer ancillary support for this hypothesis. One daily diary study found that people with elevated social anxiety experienced larger reductions in anxiety on days following intimate sexual experiences than those with low social anxiety (Kashdan & Adams et al., 2014); one EMA study found that people with elevated social anxiety reported larger decreases in NA following social interactions with close companions (i.e., close friends, family, romantic partners), although they did not report changes in PA (Hur et al., 2020); and one EMA study found that people with elevated social anxiety reported greater psychological benefits (less anxiety, less motivation to avoid social situations, greater sense of belonging) following positive events than people with lower social anxiety (Doorley et al., 2020). For people with SAD, we might expect a “mood brightening” effect when they enter social situations, where PA increases to a greater degree than controls. Given their anxiety about socializing, it is unlikely their NA will decrease when they enter a social situation.

1.2. Social Interaction Partners

Emotional experiences may differ based on who is present in a social interaction. Some studies have found that participants reported higher PA when interacting with close others (e.g., romantic partner) compared to distant others (e.g., colleagues) (Mueller et al., 2019; Venaglia & Lemay, 2017). Other studies, however, have found no differences in affect across social partners (e.g., stranger vs. romantic partner: Dunn et al., 2007). Findings on social interaction partners and social anxiety are similarly mixed. In terms of frequency, one ESM study found that participants with SAD spent more time with romantic partners and supervisors than controls (Russell et al., 2011); another study found that people with elevated social anxiety spent less time with close companions (a composite measure that included romantic partners; Hur et al., 2020), and yet another study found no differences in how often people with SAD and controls interacted with social interaction partners (Oren-Yagoda et al., 2021). In terms of emotional experiences across different social interaction partners, research is surprisingly sparse, especially among socially anxious adults. Two studies found that those with higher social anxiety reported higher PA when with close companions (e.g., close friend) than less close persons (adults: Hur et al., 2020; adolescents: Morgan et al., 2017). Taken together, a small set of experience-sampling studies suggest that social anxiety symptoms differentially may predict the frequency of and emotional experiences during social interactions depending on who is present in the interaction.

1.3. An Ecological Momentary Assessment Approach

Retrospective assessments may be ill-suited to study emotional experiences in social anxiety. In addition to general recall bias issues posed by these assessments (Stone & Shiffman, 2002), people with SAD display negativity biases when recalling life events (Edwards et al., 2003). Following social interactions, people with SAD tend to ruminate on the social interaction (Mellings & Alden, 2000). They have difficulties sustaining and savoring positive emotions (Eisner et al., 2009), which can hijack positive memories. People with SAD might underreport felt positive emotions when reporting them several hours after the event.

To address these limitations, we conducted two EMA studies. We assessed participants’ momentary PA and NA several times throughout each day for two weeks while also recording whether they were with other people at the time. We hypothesized that 1) participants with SAD would report higher average NA and lower average PA across situations than healthy controls; 2) participants with and without SAD would report higher PA during social situations than when alone, and that this difference would be stronger for those with SAD (consistent with a mood brightening effect); and 3) participants with SAD would report higher NA during social situations than when alone, and healthy controls would report the opposite (i.e., higher negative when alone than during social situations). To increase confidence in study findings, and in light of larger concerns about replicability (Maxwell et al., 2015), we tested these questions in a second study with a different sample of community adults (Study 2). In addition to replication, Study 2 addressed a limitation of Study 1 by examining patterns of affect across different social partners. In the absence of consistent evidence of emotional experiences across different social partners, we did not specify a priori hypotheses for analyses examining social interactions partners and consider them exploratory.

2. Study 1

2.1. Method

2.1.1. Participants and Procedure

Participants were community adults recruited from a large northeastern metropolitan area via flyers and posts on online forums (e.g., VolunteerMatch.org). Trained research assistants conducted semi-structured phone interviews to assess mood and anxiety symptoms. Eligible participants—those with likely SAD or the absence of mental illness—attended a laboratory session (N = 111). Clinical psychology doctoral students administered the Structured Clinical Interview (SCID) to assess anxiety, mood, substance use, eating, obsessive-compulsive and related, trauma- and stressor-related, and psychotic disorders. Eligibility criteria for the SAD group was a SAD diagnosis; comorbid diagnoses were permitted if SAD was the primary diagnosis. Eligibility criteria for the control group was the absence of any current mental illness, lifetime psychosis or mania, and past (two years) depressive episodes. Half of the SCID videos (N = 56) were randomly selected and coded by doctoral students; inter-rater reliability for SAD diagnoses was acceptable (κ = .93).

Of the 111 participants who completed SCIDs, 45 were diagnosed with SAD, 49 were absent of mental illness, and 17 were diagnosed with disorders other than SAD and therefore excluded. Seven eligible participants (3 with SAD; 4 healthy controls) declined to participate in the EMA portion, yielding a final sample of 42 participants with SAD and 45 healthy controls. In the SAD group, nine (21%) participants met criteria for generalized anxiety disorder, three (7%) for panic disorder, and two (5%) for agoraphobia; 13 (31%) met criteria for current depressive disorder, seven (17%) for alcohol use disorder, and six (14%) for post-traumatic stress disorder. The average participant age was 30.3 years (SD = 9.63); 62% were female; and self-reported race/ethnicity was 48.3% white, 19.5% Black/African American, 13.8% Asian/Pacific Islander, 6.9% Latino/Hispanic, 2.3% Arab/Middle Eastern, and 9.2% other. Participants with SAD and healthy controls did not differ on demographics (ps > .05).

Eligible participants received training and then completed a two-week EMA study. EMA surveys were sent five times per day randomly during 12pm-8pm (at least 30 minutes apart) through ESMCapture, an experience-sampling platform; this study used the text-based version, where participants received text messages containing survey links at each prompt. Trained research assistants led participants through a detailed training session explaining how to use the app, troubleshooting, and the compliance structure; participants received a printed and emailed handout with FAQs. Participants received emailed compliance updates at the beginning and end of each week that included earnings to date. Participants received $40 for the baseline session and up to $35 for the EMA portion, which included bonus payments for answering 80% or more surveys each week ($5.00 and $5.50, respectively).

2.2. Measures

2.2.1. Baseline Questionnaires

Social Anxiety Symptoms.

Participants completed the Social Interaction Anxiety Scale (SIAS; Mattick & Clarke, 1998) at baseline. The SIAS is a 20-item measure that assesses fear and avoidance of social situations. Participants indicated the degree to which they felt the statements are characteristic for them on a 5-point Likert scale from 0=not at all to 4=extremely. The full SIAS contains three reverse-coded items; factor analysis suggests removing these items improves validity and reliability (Rodebaugh et al., 2007). Therefore, we calculated a total average score using the 17 straightforward items. The SIAS scores reliably discriminate between people with and without SAD (clinical cutoff score for 17-item SIAS = 28; Rodebaugh et al., 2011). The 17-item SIAS demonstrates construct validity, as evidenced by strong positive correlations with other social anxiety measures (e.g., Liebowitz Social Anxiety Scale: Liebowitz, 1987; Social Phobia Scale: Mattick & Clarke, 1998). Reliability was acceptable (α = .97).

2.2.2. Ecological Momentary Assessment Measures

Momentary PA and NA.

Affect was measured with six positive emotion items (enthusiastic, content, joyful, proud, interested, relaxed) and six negative emotion items (angry, sluggish, anxious, sad, irritable, guilty) from the Positive and Negative Affect Schedule (PANAS; Watson et al., 1988). Participants responded using a 5-point Likert scale from 1=not at all to 5=extremely. Items were averaged to create subscales scores of PA and NA at each prompt.

Social Interactions.

Participants responded to a single yes or no item indicating whether they were with anyone else at the moment answering the survey.

3. Results

3.1. Analytical Plan

All analyses were conducted in R version 3.6.1 (R Core Team, 2019). Simulation studies of multilevel power suggest that designs with approximately 80 Level 2 units (e.g., participants) and 14 Level 1 units (e.g., days) provide sufficient power (i.e., greater than .80) to detect small (d = .2) effects, and increasing the number of responses per participant increases power to a greater degree than increasing the number of participants (Nezlek, 2011, 2012; Raudenbush & Liu, 2000). Thus, our sample size of 87 with up to 70 responses per participant is well-powered to detect small effects. This was confirmed via a simulation using EMAtools::ema.powercurve (Kleiman, 2017), which demonstrated sufficient power to detect small effects, even at lower levels of compliance.

Three-level multilevel model analyses were fit to accommodate hierarchical data nesting (prompts within days within persons). We estimated models using restricted maximum likelihood (REML) with the lme4 package (Bates et al., 2007); full statistical information for each model was examined using the sjPlot::tab_model function (Lüdecke & Lüdecke, 2017). Following centering guidelines (Nezlek, 2011), PA and NA were entered as continuous uncentered outcomes; socializing (1=yes, 0=no) was entered as a dichotomous, uncentered Level 1 (i.e., prompt) predictor; and diagnostic group was entered as a categorical Level 3 (i.e., person) predictor (1=SAD, 0=control). Participants were treated as random effects with a random intercept and no random slopes. Effect sizes were calculated using multilevel model R2 indices that account for fixed and random effects: marginal R2m (i.e., proportion of the total variance explained by the fixed effects), and conditional R2c (i.e., proportion of the total variance explained by both fixed and random effects; Nakagawa et al., 2017).1

3.2. Descriptive Statistics

Participants submitted 4,559 reports with an average of 52.4 reports per participant. Participants completed at least three of five surveys on 88.3% of study days. Most participants (75.9%) completed at least one survey each study day; only six participants completed surveys on less than 10 days.

Descriptive statistics are reported in Table 1. For continuous momentary measures (i.e., PA and NA), we computed two indices of multilevel reliability using psych::multilevel.reliability (Revelle, 2020): RCN (reliability of within-person variations averaged over items) and RKRN (reliability of between-person differences averaged over time and items). We computed bivariate correlations between PA and NA by decomposing between and within-subjects effects. Intraclass correlation coefficients (ICC) refer to the percentage of variance attributable to differences between persons relative to the total variance.

Table 1.

Descriptive Statistics of Daily Variables

Study 1
Positive affect Negative affect
Positive affect -- −0.47*
Negative affect −0.42* --
RCN/RKRN .72/.99 .51/.99
Grand Mean (SD) 2.59 (1.01) 1.48 (0.57)
ICC 0.64 0.48
Study 2
Positive affect Negative affect
Positive affect -- −0.46*
Negative affect −0.69* --
RCN/RKRN .65/.99 .18/.99
Grand Mean (SD) 2.79 (1.08) 1.53 (0.63)
ICC 0.49 0.37
Open in a new tab

Note. ICC = Intraclass correlation coefficient. RCN = reliability of within-person variations averaged over items. RKRN = reliability of between-person differences averaged over time and items. Within-person correlations are above the diagonal; between-person correlations are below the diagonal.

*

p < .001

Participants with SAD reported higher social anxiety symptoms at baseline (M = 42.12, SD = 9.79) than controls (M = 11.71, SD = 8.88; Cohen’s d = 0.86, t = −15.04, p < .001)2. Participants with and without SAD were equally likely to be with others when prompted (OR = 0.73, t = −0.92, 95% CI [0.37, 1.43], p = .36, R2m = .004, R2c = .529).

3.3. Primary Analyses

3.3.1. Between-Person Analyses

We first examined if participants with SAD and healthy controls differed in affect across situations (i.e., between-groups analyses). When with others, participants with SAD reported lower PA (b = −0.73, t = −4.78, 95% CI [−1.03, −.43], p < .001) and higher NA (b = 0.43, t = 5.75, 95% CI [.28, .57], p < .001) than did healthy controls. When alone, participants with SAD reported lower PA (b = −0.65, t = −4.27, 95% CI [−.95, −.35], p < .001) and higher NA (b = 0.40, t = 5.40, 95% CI [.25, .54], p < .001) than did healthy controls. Our results are consistent with prior research suggesting that on average, participants with social anxiety experience lower PA and higher NA than healthy controls in both social and nonsocial situations. See Table 2.

Table 2.

Between-Person Analyses: Positive and Negative Affect When with Others versus Alone

Positive Affect
 Negative Affect
b CI t p b CI t p
Study 1
SAD Group (With Others) −0.73 −1.03, −0.43 −4.78 <.001 0.43 0.28, 0.57 5.75 <.001
SAD Group (Alone) −0.65 −0.95, −0.35 −4.27 <.001 0.40 0.25, 0.54 5.40 <.001
Study 2
SAD Group (With Others) −0.94 −1.22, −0.65 −6.47 <.001 0.40 0.23, 0.56 4.67 <.001
SAD Group (Alone) −0.88 −1.16, −0.60 −6.22 <.001 0.39 0.23, 0.55 4.73 <.001
Open in a new tab

Note. CI = confidence interval; t = t-statistic. All reported test statistics represent the main effects from separate models comparing differences in affect between social anxiety disorder (SAD) diagnostic groups, where SAD group is a categorical predictor (SAD = 1; healthy controls = 0). “With others” represents main effects of affect differences between participants with and without SAD when they were with others at the time of the report; “Alone” represents main effects of affect differences between participants with and without SAD when they were alone at the time of the report.

3.3.2. Within-Person Analyses

We conducted within-person analyses to examine if participants with SAD and healthy controls experienced higher PA and higher NA when they were with others than when alone, and then examined if the magnitude of difference in affect between social and nonsocial situations differed between SAD diagnostic groups. The full multilevel model includes affect as an outcome, situation type (i.e., social vs. nonsocial) as a Level 1 predictor, and SAD as a Level 2 dichotomous moderator. The simple slope for each group represents the relationship between situation type and affect for participants with SAD and healthy controls, respectively. The moderation term (SAD * situation type) represents a test of the difference between these simple slopes; a significant moderation term suggests that the magnitude of the difference in affect between social and nonsocial situations differs based on SAD diagnosis. See Table 3.

Table 3.

Within-Person Analyses: Positive and Negative Affect When with Others and Alone

Positive Affect
 Negative Affect
b CI t p b CI t p
Study 1
Social (SAD) 0.08 0.03, 0.14 2.85 .004 −0.03 −0.07, 0.01 −1.41 .158
Social (Control) 0.16 0.11, 0.21 6.12 <.001 −0.06 −0.09, −0.02 −3.06 .002
Social*SAD Group −0.08 −0.16, −0.003 −2.04 .042 0.03 −0.03, 0.08 1.02 .306
Study 2
Social (SAD) 0.18 0.09, 0.27 3.83 <.001 −0.05 −0.11, 0.01 −1.57 .116
Social (Control) 0.23 0.15, 0.32 5.27 <.001 −0.05 −0.11, 0.003 −1.84 .065
Social*SAD Group −0.05 −0.18, 0.07 −0.85 .397 0.01 −0.08, 0.09 0.13 .899
Open in a new tab

Note. CI = confidence interval; t = t-statistic. Socializing is a categorical predictor representing whether participants were with others (1) or alone (0) at the time of the report. Rows with “SAD” represent a test of differences in affect when with others versus alone for participants diagnosed with social anxiety disorder (SAD); rows with “Controls” represent a test of differences in affect when with others versus alone for participants in the healthy control group. Social * SAD Group represents a test of the difference between these simple slopes, where a significant moderation term indicates that the magnitude of the difference in affect between social and nonsocial situations differs based on SAD diagnosis (SAD = 1, healthy controls = 0).

Participants with SAD reported higher PA when with others than when alone (b = 0.08, t = 2.85, 95% CI [0.03, 0.14], p = .004); healthy controls also reported higher PA when with others than alone (b = 0.16, t = 6.12, 95% CI [0.11, 0.21], p < .001). The relationship between being with others and PA was moderated by SAD diagnosis (b = −0.08, t = −2.04, 95% CI [− 0.16, −0.003], p = .042, R2m = .128, R2c = .755), such that the difference in PA between social and non-social situations was smaller for participants with SAD than for controls. Thus, while participants with and without SAD reported higher PA when with others than alone, the magnitude of this difference was larger for controls.

Participants with SAD reported similar levels of NA when with others than alone (b = − 0.03, t = −1.41, 95% CI [−0.07, 0.01], p = .158), whereas healthy controls reported lower NA when with others than alone (b = −0.06, t = −3.06, 95% CI [−0.09, −0.02], p = .002). The relationship between being with others and NA was not moderated by SAD diagnosis (b = 0.03, t = 1.02, 95% CI [−0.03, 0.08], p = .306, R2m = .134, R2c = .616), meaning the magnitude of the difference in NA between social and nonsocial situations was similar for participants with and without SAD.

4. Study 1 Discussion

As predicted, between-person analyses found that participants with SAD experienced higher NA and lower PA across situations than healthy controls. As predicted, within-person analyses found that participants with and without SAD reported higher PA during social interactions than when alone. The magnitude of differences in PA between social and nonsocial situations was smaller for participants with SAD than for controls, suggesting that while people with SAD can and do experience upticks in PA while socializing, they may still exhibit diminished reward responding (Richey et al., 2019). Our third hypothesis was partially supported: healthy adults reported lower NA when with others than alone, whereas participants with SAD reported similar levels of NA across social and nonsocial situations. However, the magnitude of this difference was negligible, suggesting that daily social interactions elicit similar negative emotional experiences as being alone for people with and without SAD.

One limitation of Study 1 is the lack of information about social interactions. People with SAD fear and avoid multiple types of social interactions, but this does not necessarily mean they fear all types or that fear precludes enjoyment. Some research suggests that people with SAD respond differently to different types of social situations (e.g., Thompson & Rapee, 2002). Thus, in Study 2, we sought to replicate findings for PA and NA and then examined PA and NA across different social interactions.

5. Study 2

5.1. Method

5.1.1. Participants and Procedure

As in Study 1, participants were community adults (N = 84) who completed SCIDs. Of the 84 participants, 41 were diagnosed with SAD, 40 were absent of mental illness, and 3 were diagnosed with a primary disorder other than SAD and therefore were excluded. Inter-rater reliability for SAD diagnoses was acceptable (κ = .87). Four eligible participants (2 with SAD; 2 healthy controls) declined to participate in the EMA portion, yielding a final sample of 39 participants with SAD and 38 controls. In the SAD group, three (8%) participants met criteria for generalized anxiety disorder, two (5%) for panic disorder, and one (3%) for agoraphobia; 11 (28%) met criteria for current depressive disorder, two (5%) for substance use disorder, and five (13%) for post-traumatic stress disorder. Average participant age was 28.8 years (SD = 8.74); 63.6% were female; and self-reported race/ethnicity was 51.9% white, 23.4% African American, 3.9% Asian/Asian-American, 10.4% Hispanic/Latino/Mexican-American, 1.3% Middle Eastern, and 9.1% other. Participants with SAD and controls did not differ on age, sex, or ethnicity (ps > .72).

Procedures were similar to Study 1 with the following exceptions. Participants in Study 2 earned higher compensation ($165 minimum and up to $50 in compliance bonuses), and surveys were completed via Purdue Momentary Assessment Tool (PMAT).

5.2. Measures

5.2.1. Baseline Questionnaires

Social Anxiety Symptoms.

The 17-item SIAS was again used to assess general social and evaluation fears and avoidance. Reliability was acceptable (α = .88).

5.2.2. Ecological Momentary Assessment Measures

Momentary PA and NA.

Momentary affect was assessed with four positive emotion items (content, enthusiastic, joyful, relaxed) and four negative emotion items (angry, sluggish, sad, anxious/nervous) from the PANAS (Watson et al., 1988).3 Participants reported the degree to which each emotion reflected the way they felt in that moment using a 5-point Likert scale from 1=very slightly or not at all to 5=extremely. Items of each subscale were averaged to create total scores of PA and NA, respectively, at each prompt.

Social Interactions.

Participants reported whether they were alone or with someone else (i.e., romantic partner, family member, close friend, superior [boss/teacher], coworker/neighbor, acquaintance, or stranger) at the time of prompts. Each social interaction was coded as 1=yes or 0=no.

6. Results

6.1. Analytical Plan

Analytical plan is identical to Study 1. Simulation power analyses again confirmed that our sample size of 77 with up to 70 responses per participant is sufficiently powered to detect small effects, even at low levels of compliance.4

6.2. Descriptive Statistics

Participants submitted 2,684 reports, with an average of 34.9 reports per participant. Participants completed at least three of five surveys on 57.6% of study days. Most participants (63.6%) completed at least one survey each study day; only 10 participants completed surveys on less than 10 days. Descriptive statistics are presented in Table 1.

Participants with SAD reported much higher social anxiety symptoms at baseline (M = 38.56, SD = 7.36) than controls (M = 15.80, SD = 5.68; Cohen’s d = 0.87, t = −14.61, p < .001). As in Study 1, participants with and without SAD were equally likely to be with others when prompted (OR = 0.73, t = −1.19, 95% CI [0.44, 1.23], p = .236, R2m = .005, R2c = .308).

6.3. Primary Analyses

6.3.1. Between-Person Analyses

As in Study 1, we conducted between-person analyses to first examine if participants with SAD and controls differed in mean affect in social compared to nonsocial situations. As in Study 1, when with others, participants with SAD reported lower PA (b = −0.94, t = −6.47, 95% CI [−1.22, −.65], p < .001) and higher NA (b = 0.40, t = 4.67, 95% CI [0.23, 0.56], p < .001) than controls. As in Study 1, when alone, participants with SAD reported lower PA (b = −0.88, t = − 6.22, 95% CI [−1.16, −.60], p < .001) and higher NA (b = 0.39, t = 4.73, 95% CI [0.23, 0.55], p < .001) than controls. See Table 2.

6.3.2. Within-Person Analyses

As in Study 1, we conducted within-person analyses to examine if participants with SAD and healthy controls differed in levels of affect when with others than when alone, and then examined if the magnitude of these differences differed between diagnostic groups.

As in Study 1, participants with SAD reported higher PA when with others than when alone (b = 0.18, t = 3.83, 95% CI [0.09, 0.27], p < .001); the same was true for controls (b = 0.23, t = 5.27, 95% CI [0.15, 0.32], p < .001). SAD diagnosis did not moderate the relationship between being with others and PA (b = −0.05, t = −0.85, 95% CI [−0.18, 0.07], p = .397, R2m = .189, R2c = .630), meaning that the magnitude of the difference in PA between social and nonsocial situations was similar for participants with and without SAD.

Participants with SAD reported similar levels of NA when with others than when alone (b = −0.05, t = −1.57, 95% CI [−0.11, 0.01], p = .116); controls also reported similar levels of NA when with others than when alone (b = −0.05, t = −1.84, 95% CI [−0.11, 0.003], p = .065). SAD did not moderate the relationship between being with others and NA (b = 0.01, t = 0.13, 95% CI [−0.08, 0.09], p = .899, R2m = .098, R2c = .575), meaning that the magnitude of the difference in NA between social and nonsocial situations was similar for participants with and without SAD. See Table 3.

6.4. Exploratory Analyses

Participants socialized with the following types of social partners: (cumulative percentage is >100% because participants could select multiple types of social partners during each social interaction): romantic partner (31.5%), family member (26.7%), coworker/neighbor (17.3%), close friend (18.7%), superior (6.8%), acquaintance (4.9%), and stranger (3.5%).

To examine whether participants with SAD were more likely to be with certain types of interaction partners than controls, we conducted a series of multilevel logistic models, where SAD was entered as a dichotomous Level 2 predictor, and social interaction partner type was entered as a dichotomous outcome.5 Participants with SAD were less likely to be with close friends (OR = 0.34, t = −3.17, 95% CI [0.17, 0.66], p = .002, R2m = .052, R2c = .407) than controls; controls were with close friends 13% of time, compared to 5.7% of the time for participants with SAD. All other tests were nonsignificant (ps > .41), meaning that participants with SAD were no more or less likely to be with other social interaction partners.

Next, we examined how PA and NA differed when participants were with different social interaction partners. We ran a series of multilevel models in a subset of the data when participants were with other people. In these models, affect (PA or NA) was entered as a continuous outcome; social interaction partner type was entered as a Level 1 dichotomous predictor that represented differences in affect between the specified partner type and all other partner types; and SAD was entered as a Level 2 moderator whose interaction term represented a test of diagnostic group differences in how partner type predicted affect. Participants with and without SAD reported similar affect patterns across social interaction partners, characterized by lower PA when with superiors and coworkers or neighbors and higher PA when with romantic partners and close friends compared to when with other social interaction partners. All analyses for NA were nonsignificant, meaning that participants’ reported level of NA did not differ across social interaction partners. Full statistics are presented in Table 4.

Table 4.

Study 2: Positive and Negative Affect Across Social Interaction Partners

Positive Affect
 Negative Affect
b CI t p b CI t p
Romantic partner 0.37 0.18, 0.57 3.83 <.001 −0.03 −0.14, 0.08 −0.53 .597
Close friend 0.23 0.06, 0.40 2.61 .009 −0.04 −0.14, 0.06 −0.83 .407
Family member 0.11 −0.10, 0.32 1.05 .293 −0.01 −0.13, 0.11 −0.13 .900
Superior −0.50 −0.76, −0.24 −3.77 <.001 0.02 −0.13, 0.17 0.27 .788
Coworkers/Neighbors −0.33 −0.55, −0.11 −2.94 .003 −0.05 −0.18, 0.08 −0.81 .421
Acquaintance −0.11 −0.43, 0.22 −0.64 .522 −0.06 −0.25, 0.13 −0.62 .538
Stranger −0.20 −0.56, 0.15 −1.12 .262 0.07 −0.14, 0.28 0.62 .538
Open in a new tab

Note. CI = confidence interval; t = t-statistic. Mean differences in positive and negative affect across social interaction partners did not differ between participants with and without SAD (i.e., no moderation; ps > .05).

7. Study 2 Discussion

In Study 2, we first sought to replicate Study 1 findings in a second sample of adults diagnosed with SAD and psychologically healthy controls. As in Study 1, participants with and without SAD were equally likely to be with others when prompted. We replicated all between-person effects found in Study 1: participants with SAD reported lower PA and higher NA in social situations than nonsocial situations, as did healthy controls.

We replicated some, but not all, within-person effects. As in Study 1, participants with SAD reported higher PA when in social than nonsocial situations and no differences in NA across situations. As in Study 1, healthy controls also reported higher PA in social situations, but they also reported no differences in NA across situations; this NA finding differs from Study 1, in which controls reported lower NA in social situations. We replicated moderation effects for NA but not PA: the magnitude of difference in NA between social and nonsocial situations was similar for participants with and without SAD (as in Study 1), as was the magnitude of difference in PA across situations (this finding differs from Study 1, in which people with SAD reported smaller increases in PA from social to nonsocial situations than controls).

We conducted an additional set of analyses examining group differences in frequency and emotional experiences across social interaction partners. Participants with and without SAD reported largely similar affect patterns, characterized by relatively higher PA when with romantic partners and close friends and lower PA with superiors and coworkers or neighbors. In terms of frequency of social contact with partners, the only difference observed was that participants with SAD were less likely than controls to be with close friends when prompted.

8. General Discussion

Across two EMA studies, we examined the emotional experiences of participants with and without SAD during naturally occurring social interactions. First, we conducted traditional between-person analyses and found in both studies that participants with SAD experienced higher NA and lower PA across situations than healthy controls. These findings replicate prior work on the affective profile of people with SAD (Brown et al., 1998; Kashdan, 2007; Kashdan & Collins, 2010). Second, we conducted within-person analyses and found that in both studies, participants with SAD experienced higher PA and similar NA when with others than alone—an affect pattern largely similar to that of controls. These findings suggest that participants with SAD can and do experience positive emotions during social interactions, and that socializing is more pleasurable than spending time alone. Third, in both studies, participants with SAD were as likely as healthy controls to be with other people when prompted. These findings are consistent with prior research that differentiates social anxiety from social anhedonia (Brown et al., 2007). Fourth, exploratory analyses in Study 2 found that participants reported lower PA when with superiors and coworkers or neighbors and higher PA when with romantic partners and close friends. These findings suggest that emotional benefits of social interactions likely vary based on the type and closeness of one’s relationship with whom they interact. Importantly, the significance and interpretation of all effects did not change after controlling for depression diagnoses, suggesting that our results are not better explained by depression.

8.1. Patterns of Emotional Deficits in Social Anxiety

Using traditional between-person comparisons, we found that participants with SAD experienced lower PA and higher NA in social situations than controls. We then assessed if people with SAD experienced different levels of PA and NA in social situations than they did when they were alone (i.e., within-person comparisons). As predicted, people with SAD reported higher PA when with other people than when alone. One parsimonious explanation is that social interactions partially and momentarily satisfy the basic human desire for social connection (Baumeister & Leary, 1995). Our results suggest that despite elevated fear of negative evaluation and potential rejection, people with SAD still benefit from and strive for social connection. Nonetheless, we did not assess mechanisms and therefore offer speculative conclusions that need to be tested in future research.

Contrary to prediction, participants with SAD reported similar levels of NA when with others than when alone. Although we did not measure the temporal sequence of emotions before, during, or after social interactions, one possible explanation is that negative emotions are also—or even more strongly evoked—after the social interaction. For people with SAD, negative emotions often occur in anticipation of interactions (negative pre-event rumination) and after they have occurred (negative post-event rumination) (see Modini et al., 2018). In considering between-person analyses that found participants with SAD reported higher NA than healthy controls in both social and nonsocial interactions, it is possible that participants with SAD maintained a relatively high, immutable level of NA. It is also possible that social interactions more strongly target reward systems that dictate PA than threat systems that dictate NA (Kawamichi et al., 2015, 2016). Healthy controls in Study 2 also reported similar NA in social and nonsocial situations. Thus, although contrary to prediction, three of four tests across samples suggest that participants with and without SAD report negligible differences in NA in social versus nonsocial situations. In contrast, four of four tests across samples suggest that participants with and without SAD reported higher PA in social situations than when alone. Nonetheless, these results were not replicated across studies; controls in Study 1 reported lower NA in social situations (as expected). More intensive longitudinal research is needed to better understand emotional trajectories before, during, and after social interactions.

We failed to find support for the “mood brightening” hypothesis. In Study 1, the pattern of results was opposite as expected; participants with SAD reported smaller differences in PA between social and nonsocial settings than controls. In Study 2, participants with and without SAD reported similar differences in PA between social and nonsocial situations. Results from Study 1 are consistent with a recent study that found while participants high in social anxiety experienced increased positive emotions during social interaction, this increase in PA was attenuated relative to participants low in social anxiety (Barber et al., 2021). These findings might suggest that people with SAD have social reward processing deficits that attenuate PA (Richey et al., 2019). Taken together, although participants with SAD reported higher PA when they were with others than when they were alone, their capacity to experience positive emotions from social interactions may be inhibited. Nonetheless, this finding was not replicated in Study 2 and should be interpreted accordingly.

8.2. Social Interaction Frequency and Partners

Participants with and without SAD in both studies did not differ in their frequency of socializing across social partners (as found in Oren-Yagoda et al., 2021 and similar to findings in undiagnosed samples using trait social anxiety symptoms: Brown et al., 2007; Daros et al., 2019). This finding is somewhat paradoxical, as SAD is in part defined by avoidance of social interactions. Nonetheless, people with SAD do not necessarily fear or avoid all social situations. Moreover, the DSM-5 avoidance criterion for SAD includes avoidance of social situations or enduring them with intense fear/anxiety (APA, 2013). People with SAD often engage in internal avoidance behaviors in the context of social interactions (Goodman, Larrazabal et al., 2019; Kashdan, Goodman et al., 2014). The likelihood that someone with SAD does not avoid entering social situations is very low, but this criterion does broaden the ways in which someone with SAD can try to avoid social scrutiny. Nonetheless, in this study, we assessed whether and with whom participants were socializing, and we cannot infer if they were also avoiding socializing.

As for who people with SAD socialized with, exploratory analyses in Study 2 found that people with SAD were less likely than controls to be with close friends. This finding is consistent with one study in which participants with higher social anxiety spent less time with close companions than those with lower social anxiety (Hur et al., 2020). Given that people with SAD have fewer close friends than less socially anxious peers (Schneier et al., 1994) and sometimes report having no close friends at all (Whisman et al., 2000), it is unsurprising that they reported fewer interactions with close friends than controls. Nonetheless, having fewer interactions with close friends does not preclude the possibility of experiencing positive emotions when those interactions do occur.

We also explored differences in emotional experiences across social partners and found that social interactions with close others (i.e., romantic partners, close friends) facilitated greater positive emotional experiences relative to socializing with other interaction partners; this was true for participants with and without SAD. When interacting with close social partners, people may have less intense evaluative fears and perceive social situations as less ambiguous, thus providing greater opportunity for meaningful social connection and positive emotions (e.g., Allan, 1998). The demands of uncertain situations where social performance expectations are unclear contribute to social evaluative fears and emotionally unsatisfying social experiences (Whiting et al., 2013). Although more work is needed to examine the intricacies of social interactions, our results provide initial evidence that emotional experiences vary across social settings in similar ways for people with and without SAD.

8.3. Clinical Considerations

Our findings offer several clinical considerations. First, given that participants with SAD reported higher PA in social situations than when alone, traditional cognitive-behavioral therapies focused on reducing social avoidance can be augmented with interventions to increase social approach behaviors. Approach and avoidance systems are related but independent systems, and the approach system largely regulates positive emotions (Eisenberger & Cole, 2012). SAD is characterized by overactive avoidance motivation and deficient approach motivation (Richey et al., 2019). Social approach processes are modifiable and may alleviate social anxiety symptoms (e.g., Alden et al., 2018). Second, we found evidence of positivity deficits, such that participants with SAD reported lower PA on average than controls and smaller increases in PA from nonsocial to social situations. The classification of SAD as an anxiety disorder certainly warrants evidence-based exposure therapy; nonetheless, SAD is unique among the anxiety disorders in that it is characterized by PA deficits. Treatment focused on symptom reduction (e.g., decreasing anxiety in social situations) can be augmented with interventions that increase positive emotions (e.g., behavioral activation). Clinicians can also review ways to address cognitive processes and safety behaviors that interfere with enjoyment of social situations (e.g., self-focused attention: Clark & Wells, 1995; averted eye contact, restricted self-disclosure, reassurance seeking; Alden & Taylor, 2004).

8.4. Limitations and Future Directions

Several study limitations offer opportunities for future research. First, future research can gather detailed information about the nature of social interactions. Most participants were interacting with someone they knew, and people with SAD may experience different emotions when interacting with strangers. On the one hand, they might have more difficulty interacting with strangers because of the high degree of uncertainty (Boelen & Reijntjes, 2009); on the other hand, if they are interacting with someone who they are unlikely to see again (e.g., passenger on an airplane), they might have less difficulty because the consequences of feared rejection are lower (Epley & Schroeder, 2014). Second, future research can examine emotion regulation strategies during and after social interactions. Although participants with SAD experienced elevated PA in social situations relative to being alone, they may have difficulty maintaining positive emotions (Eisner et al., 2009; Kashdan & Steger, 2006). Future research can show how to improve regulatory deficits, including strategies that increase positive emotions (e.g., savoring; Jose et al., 2012) and reducing use of strategies that may increase negative emotions (e.g., rumination; Abbott & Rapee, 2004). Third, two additional positive and two additional negative emotion adjectives were used in Study 1 than in Study 2. Although we replicated most effects across studies, two were inconsistent (healthy controls’ difference in NA between social and nonsocial situations, and SAD diagnosis moderating social situation type and PA), which might be attributable to slight differences in affect measures. Selection of emotion items is highly variable, and the debate about what “counts” as an emotion is long standing, including whether certain emotions are universal. ESM research in particular would benefit from greater uniformity in affect measurement. Fourth, research can examine precursors to social experiences that make it more or less likely someone will enter and enjoy an interaction. Unexplored territory includes whether people with SAD require a greater degree of trust with social interaction partners before experiencing the reciprocal self-disclosures that produce intimacy and healthy relationships, and whether people with SAD show strength-related deficits when socializing such as inhibited expressions of courage, humor, compassion, and forgiveness. A greater attention to behavior, as an adjunct to reported experiences, will allow for a greater understanding of the barriers to rewarding social interactions for people suffering from SAD.

9. Conclusion

SAD is defined by interpersonal distress, avoidance, and impairment. Yet, despite having fewer, less satisfying social relationships, our results suggest people with SAD enjoy social interactions when they occur. Contrary to lay belief, we found that people with SAD were happier when with others than alone. Feeling anxious or concerned about socializing does not preclude experiencing pleasure while socializing. Our replicated findings across two clinical samples offer promise that people with SAD can and do experience positive emotions during social interactions.

Supplementary Material

1

Highlights.

  • People with SAD display positive affect deficits in daily life.

  • People with SAD report higher positive affect when with others than when alone.

  • People with SAD may experience diminished reward responding when socializing.

  • People with and without SAD report similar affect across interaction partners.

  • Anxiety about socializing does not preclude positive emotions while socializing.

Acknowledgments

We thank Niels van Berkel for his instrumental help with data preparation.

Funding:

Study 1 was supported by a grant from the National Institute on Alcohol Abuse and Alcoholism (NIAAA) awarded to Fallon Goodman (F31-AA024372); Study 2 was supported by a grant from the National Institute of Mental Health awarded to Todd Kashdan (R21-MH073937).

Footnotes

Declarations of interest: none.

1

Given that SAD and depressive disorders are highly comorbid (Langer & Rodebaugh, 2014), we reran all models with depression diagnosis as a Level 3 (i.e., person level) covariate. The significance and interpretation of focal effects did not change when controlling for depression. Full models are provided in Supplementary Tables S1 and S2.

2

SIAS scores were not recorded for one participant with SAD due to a technical error; therefore, we report group differences for 86 participants (41 participants with SAD, 45 controls).

3

In Study 2, we used the same four adjectives in Study 1 to measure positive and negative emotion (PA: enthusiastic, joyful, relaxed, content; NA: angry, anxious, sad, sluggish), which reflect the four quadrants of the emotion circumplex (i.e., high-low arousal x positive-negative valence; Russell, 1980). We removed two emotion items from each subscale that occur less frequently in daily life (proud; guilty) or are conceptually similar to other emotions (interested; irritable) (Trampe et al., 2015). Nonetheless, Study 1 items may capture more breadth in NA and PA than those in Study 2 and should be interpreted accordingly.

4

As in Study 1, we reran all models controlling for depression diagnosis. As in Study 1, the significance and interpretation of focal effects did not change. Full models are provided in Supplementary Tables S1 and S2.

5

These analyses do not account for the fact that participants could report multiple types of interaction partners in a given prompt.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Abbott MJ, & Rapee RM (2004). Post-event rumination and negative self-appraisal in social phobia before and after treatment. Journal of Abnormal Psychology, 113(1), 136–144. [DOI] [PubMed] [Google Scholar]
  2. Alden LE, Buhr K, Robichaud M, Trew JL, & Plasencia ML (2018). Treatment of social approach processes in adults with social anxiety disorder. Journal of Consulting and Clinical Psychology, 86(6), 505–517. [DOI] [PubMed] [Google Scholar]
  3. Alden LE, & Taylor CT (2004). Interpersonal processes in social phobia. Clinical Psychology Review, 24(7), 857–882. [DOI] [PubMed] [Google Scholar]
  4. Alden LE, & Taylor CT (2010). Interpersonal processes in social anxiety disorder. Interpersonal Processes in the Anxiety Disorders: Implications for understanding psychopathology and treatment. (pp. 125–152). American Psychological Association. [Google Scholar]
  5. Allan G (1998). Friendship, sociology and social structure. Journal of Social and Personal Relationships, 15(5), 685–702. [Google Scholar]
  6. American Psychiatric Association. (2013). Diagnostic and Statistical Manual of Mental Disorders, 5th Edition: DSM-5 (5th ed.). American Psychiatric Publishing. [Google Scholar]
  7. Barber KC, Michaelis MA, & Moscovitch DA (2021). Social anxiety and the generation of positivity during dyadic interaction: Curiosity and authenticity are the keys to success. Behavior Therapy. [DOI] [PubMed] [Google Scholar]
  8. Bates D, Sarkar D, Bates MD, & Matrix L (2007). The lme4 package. R package version, 2(1), 1–29. [Google Scholar]
  9. Baumeister RF, & Leary MR (1995). The need to belong: Desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin, 117(3), 497–529. [PubMed] [Google Scholar]
  10. Berry DS, & Hansen JS (1996). Positive affect, negative affect, and social interaction. Journal of Personality and Social Psychology, 71(4), 796–809. [Google Scholar]
  11. Boelen PA, & Reijntjes A (2009). Intolerance of uncertainty and social anxiety. Journal of Anxiety Disorders, 23(1), 130–135. [DOI] [PubMed] [Google Scholar]
  12. Brown TA, Chorpita BF, & Barlow DH (1998). Structural relationships among dimensions of the DSM-IV anxiety and mood disorders and dimensions of negative affect, positive affect, and autonomic arousal. Journal of Abnormal Psychology, 107(2), 179–192. [DOI] [PubMed] [Google Scholar]
  13. Brown LH, Silvia PJ, Myin-Germeys I, & Kwapil TR (2007). When the need to belong goes wrong: The expression of social anhedonia and social anxiety in daily life. Psychological Science, 18(9), 778–782. [DOI] [PubMed] [Google Scholar]
  14. Bylsma LM, Taylor-Clift A, & Rottenberg J (2011). Emotional reactivity to daily events in major and minor depression. Journal of Abnormal Psychology, 120(1), 155–167. [DOI] [PubMed] [Google Scholar]
  15. Chorpita BF, Yim L, Moffitt C, Umemoto LA, & Francis SE (2000). Assessment of symptoms of DSM-IV anxiety and depression in children: a revised child anxiety and depression scale. Behaviour Research and Therapy, 38(8), 835–855. [DOI] [PubMed] [Google Scholar]
  16. Clark DM, & Wells A (1995). A cognitive model of social phobia. In Heimberg R, Leibowitz M, Hope DA, & Schneier FR (Eds.), Social phobia: Diagnosis, assessment, and treatment (pp.69–93). New York: Guilford Press. [Google Scholar]
  17. Daros AR, Daniel KE, Meyer MJ, Chow PI, Barnes LE, & Teachman BA (2019). Impact of social anxiety and social context on college students’ emotion regulation strategy use: An experience sampling study. Motivation and Emotion, 43(5), 844–855. [Google Scholar]
  18. Diener E, Seligman ME, Choi H, & Oishi S (2018). Happiest people revisited. Perspectives on Psychological Science, 13(2), 176–184. [DOI] [PubMed] [Google Scholar]
  19. Doorley JD, Goodman FR, Disabato DJ, Kashdan TB, Weinstein JS, & Shackman AJ (2020). The momentary benefits of positive events for individuals with elevated social anxiety. Emotion, 21(3), 595–606. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Dunn EW, Biesanz JC, Human LJ, & Finn S (2007). Misunderstanding the affective consequences of everyday social interactions: The hidden benefits of putting one’s best face forward. Journal of Personality and Social Psychology, 92(6), 990–1005. [DOI] [PubMed] [Google Scholar]
  21. Edwards SL, Rapee RM, & Franklin J (2003). Post-event rumination and recall bias for a social performance event in high and low socially anxious individuals. Cognitive Therapy and Research, 27(6), 603–617. [Google Scholar]
  22. Eisenberger NI, & Cole SW (2012). Social neuroscience and health: Neurophysiological mechanisms linking social ties with physical health. Nature Neuroscience, 15(5), 669–674. [DOI] [PubMed] [Google Scholar]
  23. Eisner LR, Johnson SL, & Carver CS (2009). Positive affect regulation in anxiety disorders. Journal of Anxiety Disorders, 23(5), 645–649. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Epley N, & Schroeder J (2014). Mistakenly seeking solitude. Journal of Experimental Psychology: General, 143(5), 1980–1999. [DOI] [PubMed] [Google Scholar]
  25. Geyer EC, Fua KC, Daniel KE, Chow PI, Bonelli W, Huang Y, Barnes LE, & Teachman BA (2018). I did OK, but did I like it? Using ecological momentary assessment to examine perceptions of social interactions associated with severity of social anxiety and depression. Behavior Therapy, 49(6), 866–880. [DOI] [PubMed] [Google Scholar]
  26. Goodman FR, Kashdan TB, Stiksma MC, & Blalock DV (2019). Personal strivings to understand anxiety disorders: Social anxiety as an exemplar. Clinical Psychological Science, 7(2), 283–301. [Google Scholar]
  27. Goodman FR, Larrazabal M, West JT, & Kashdan TB (2019). Experiential avoidance across anxiety disorders. In Olatunji BO (Ed). Cambridge Handbook of Anxiety and Related Disorders (pp. 255–281). Cambridge, UK: Cambridge University Press. [Google Scholar]
  28. Goodman FR, Stiksma MC, & Kashdan TB (2018). Social anxiety and the quality of everyday social interactions: The moderating influence of alcohol consumption. Behavior Therapy, 49(3), 373–387. [DOI] [PubMed] [Google Scholar]
  29. Hughes AA, Heimberg RG, Coles ME, Gibb BE, Liebowitz MR, & Schneier FR (2006). Relations of the factors of the tripartite model of anxiety and depression to types of social anxiety. Behaviour Research and Therapy, 44(11), 1629–1641. [DOI] [PubMed] [Google Scholar]
  30. Hur J, DeYoung KA, Islam S, Anderson AS, Barstead MG, & Shackman AJ (2020). Social context and the real-world consequences of social anxiety. Psychological Medicine, 50(12), 1989–2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Jose PE, Lim BT, & Bryant FB (2012). Does savoring increase happiness? A daily diary study. The Journal of Positive Psychology, 7(3), 176–187. [Google Scholar]
  32. Kashdan TB (2007). Social anxiety spectrum and diminished positive experiences: Theoretical synthesis and meta-analysis. Clinical Psychology Review, 27(3), 348–365. [DOI] [PubMed] [Google Scholar]
  33. Kashdan TB, Adams LM, Farmer AS, Ferssizidis P, McKnight PE, & Nezlek JB (2014). Sexual healing: Daily diary investigation of the benefits of intimate and pleasurable sexual activity in socially anxious adults. Archives of Sexual Behavior, 43(7), 1417–1429. [DOI] [PubMed] [Google Scholar]
  34. Kashdan TB, & Collins RL (2010). Social anxiety and the experience of positive emotion and anger in everyday life: An ecological momentary assessment approach. Anxiety, Stress, & Coping, 23(3), 259–272. [DOI] [PubMed] [Google Scholar]
  35. Kashdan TB, Goodman FR, Machell KA, Kleiman EM, Monfort SS, & Ciarrochi J, & Nezlek J (2014). A contextual approach to experiential avoidance and social anxiety: Evidence from an experimental interaction and daily interactions of people with social anxiety disorder. Emotion, 14(4), 769–781. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Kashdan TB, & Steger MF (2006). Expanding the topography of social anxiety: An experience-sampling assessment of positive emotions, positive events, and emotion suppression. Psychological Science, 17(2), 120–128. [DOI] [PubMed] [Google Scholar]
  37. Kashdan TB, Weeks JW, & Savostyanova AA (2011). Whether, how, and when social anxiety shapes positive experiences and events: A self-regulatory framework and treatment implications. Clinical Psychology Review, 31(5), 786–799. [DOI] [PubMed] [Google Scholar]
  38. Kawamichi H, Sugawara SK, Hamano YH, Makita K, Kochiyama T, & Sadato N (2016). Increased frequency of social interaction is associated with enjoyment enhancement and reward system activation. Scientific Reports, 6(1), 1–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Kawamichi H, Yoshihara K, Sasaki AT, Sugawara SK, Tanabe HC, Shinohara R, Sugisawa Y, Tokutake K, Mochozuki Y, Anme T, & Sadato N (2015). Perceiving active listening activates the reward system and improves the impression of relevant experiences. Social Neuroscience, 10(1), 16–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Kleiman EM (2017). EMAtools: Data management tools for real-time monitoring/ecological momentary assessment data. Retrieved from https://CRAN.Rproject.org/package=EMAtools
  41. Langer JK, & Rodebaugh TL (2014). Comorbidity of social anxiety disorder and depression. In The Oxford Handbook of Depression and Comorbidity (pp. 111–128). Oxford University Press. [Google Scholar]
  42. Lavigne GL, Vallerand RJ, & Crevier-Braud L (2011). The fundamental need to belong: On the distinction between growth and deficit-reduction orientations. Personality and Social Psychology Bulletin, 37(9), 1185–1201. [DOI] [PubMed] [Google Scholar]
  43. Lazarus G, & Shahar B (2018). The role of shame and self-criticism in social anxiety: A daily-diary study in a nonclinical sample. Journal of Social and Clinical Psychology, 37(2), 107–127. [Google Scholar]
  44. Liebowitz MR (1987). Social phobia. Modern Problems of Pharmacopsychiatry, 22, 141–173. [DOI] [PubMed] [Google Scholar]
  45. Lüdecke D, & Lüdecke MD (2017). Package ‘sjPlot’.
  46. Mattick RP, & Clarke JC (1998). Development and validation of measures of social phobia scrutiny fear and social interaction anxiety. Behaviour Research and Therapy, 36(4), 455–470. [DOI] [PubMed] [Google Scholar]
  47. Maxwell SE, Lau MY, & Howard GS (2015). Is psychology suffering from a replication crisis? What does “failure to replicate” really mean?. American Psychologist, 70(6), 487–498. [DOI] [PubMed] [Google Scholar]
  48. Mehl MR, Vazire S, Holleran SE, & Clark CS (2010). Eavesdropping on happiness: Well-being is related to having less small talk and more substantive conversations. Psychological Science, 21(4), 539–541. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Mellings TMB, & Alden LE (2000). Cognitive processes in social anxiety: The effects of self-focus, rumination and anticipatory processing. Behaviour Research and Therapy, 38(3), 243–257. [DOI] [PubMed] [Google Scholar]
  50. Mueller S, Ram N, Conroy DE, Pincus AL, Gerstorf D, & Wagner J (2019). Happy like a fish in water? The role of personality–situation fit for momentary happiness in social interactions across the adult lifespan. European Journal of Personality, 33(3), 298–316. [Google Scholar]
  51. Modini M, Rapee RM, & Abbott MJ (2018). Processes and pathways mediating the experience of social anxiety and negative rumination. Behaviour Research and Therapy, 103, 24–32. [DOI] [PubMed] [Google Scholar]
  52. Morgan JK, Lee GE, Wright AG, Gilchrist DE, Forbes EE, McMakin DL, Dahl RE, Ladouceur CD, Ryan ND, & Silk JS (2017). Altered positive affect in clinically anxious youth: The role of social context and anxiety subtype. Journal of Abnormal Child Psychology, 45(7), 1461–1472. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Moscovitch DA (2009). What is the core fear in social phobia? A new model to facilitate individualized case conceptualization and treatment. Cognitive and Behavioral Practice, 16(2), 123–134. [Google Scholar]
  54. Moscovitch DA, Rowa K, Paulitzki JR, Ierullo MD, Chiang B, Antony MM, & McCabe RE (2013). Self-portrayal concerns and their relation to safety behaviors and negative affect in social anxiety disorder. Behaviour Research and Therapy, 51(8), 476–486. [DOI] [PubMed] [Google Scholar]
  55. Nakagawa S, Johnson PC, & Schielzeth H (2017). The coefficient of determination R 2 and intra-class correlation coefficient from generalized linear mixed-effects models revisited and expanded. Journal of the Royal Society Interface, 14(134), 20170213. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Nezlek JB (2008). An introduction to multilevel modeling for social and personality psychology. Social and Personality Psychology Compass, 2(2), 842–860. [Google Scholar]
  57. Nezlek JB (2012). Introduction. In Diary methods for social and personality psychology. 1–9. SAGE Publications Ltd, 10.4135/9781446287903 [DOI] [Google Scholar]
  58. Oren-Yagoda R, Schwartz M, & Aderka IM (2021). The grass is always greener: Envy in Social Anxiety Disorder. Journal of Anxiety Disorders, 102445. [DOI] [PubMed] [Google Scholar]
  59. Peeters F, Nicolson NA, Berkhof J, Delespaul P, & deVries M (2003). Effects of daily events on mood states in major depressive disorder. Journal of Abnormal Psychology, 112(2), 203–211. [DOI] [PubMed] [Google Scholar]
  60. R Core Team (2019). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/ [Google Scholar]
  61. Raudenbush SW, & Liu X (2000). Statistical power and optimal design for multisite randomized trials. Psychological Methods, 5(2), 199–213. [DOI] [PubMed] [Google Scholar]
  62. Revelle W (2020). Package ‘psych v2.0.12’.
  63. Richey JA, Brewer JA, Sullivan-Toole H, Strege MV, Kim-Spoon J, White SW, & Ollendick TH (2019). Sensitivity shift theory: A developmental model of positive affect and motivational deficits in social anxiety disorder. Clinical Psychology Review, 72, 101756. [DOI] [PubMed] [Google Scholar]
  64. Rodebaugh TL, Heimberg RG, Brown PJ, Fernandez KC, Blanco C, Schneier FR, & Liebowitz MR (2011). More reasons to be straightforward: Findings and norms for two scales relevant to social anxiety. Journal of Anxiety Disorders, 25(5), 623–630. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Rodebaugh TL, Woods CM, & Heimberg RG (2007). The reverse of social anxiety is not always the opposite: The reverse-scored items of the Social Interaction Anxiety Scale do not belong. Behavior Therapy, 38(2), 192–206. [DOI] [PubMed] [Google Scholar]
  66. Rottenberg J (2017). Emotions in depression: What do we really know?. Annual Review of Clinical Psychology, 13, 241–263. [DOI] [PubMed] [Google Scholar]
  67. Russell JA (1980). A circumplex model of affect. Journal of Personality and Social Psychology, 39, 1161–1178. [DOI] [PubMed] [Google Scholar]
  68. Russell JJ, Moskowitz DS, Zuroff DC, Bleau P, Pinard G, & Young SN (2011). Anxiety, emotional security and the interpersonal behavior of individuals with social anxiety disorder. Psychological Medicine, 41(3), 545–554. [DOI] [PubMed] [Google Scholar]
  69. Sandstrom GM, & Dunn EW (2014). Is efficiency overrated? Minimal social interactions lead to belonging and positive affect. Social Psychological and Personality Science, 5(4), 437–442. [Google Scholar]
  70. Schneier FR, Heckelman LR, Garfinkel R, Campeas R, Fallon BA, Gitow A, Street L, Bene DD, & Liebowitz MR (1994). Functional impairment in social phobia. The Journal of Clinical Psychiatry. 55(8), 322–331. [PubMed] [Google Scholar]
  71. Silvia PJ & Kwapil TR (2010). Aberrant asociality: How individual differences in social anhedonia illuminate the need to belong. Journal of Personality, 79(6), 1315–1332. [DOI] [PubMed] [Google Scholar]
  72. Stone AA, & Shiffman S (2002). Capturing momentary, self-report data: A proposal for reporting guidelines. Annals of Behavioral Medicine, 24(3), 236–243. [DOI] [PubMed] [Google Scholar]
  73. Sun J, Harris K, & Vazire S (2019). Is well-being associated with the quantity and quality of social interactions? Journal of Personality and Social Psychology, 119(6), 1478–1496. [DOI] [PubMed] [Google Scholar]
  74. Tellegen A, Watson D, & Clark LA (1999). On the dimensional and hierarchical structure of affect. Psychological Science, 10(4), 297–303. [Google Scholar]
  75. Thompson S, & Rapee RM (2002). The effect of situational structure on the social performance of socially anxious and non-anxious participants. Journal of Behavior Therapy and Experimental Psychiatry, 33(2), 91–102. [DOI] [PubMed] [Google Scholar]
  76. Trampe D, Quoidbach J, & Taquet M (2015). Emotions in everyday life. PloS One, 10(12), e0145450. [DOI] [PMC free article] [PubMed] [Google Scholar]
  77. Venaglia RB, & Lemay EP (2017). Hedonic benefits of close and distant interaction partners: The mediating roles of social approval and authenticity. Personality and Social Psychology Bulletin, 43(9), 1255–1267. [DOI] [PubMed] [Google Scholar]
  78. Vittengl JR, & Holt CS (1998). Positive and negative affect in social interactions as a function of partner familiarity, quality of communication, and social anxiety. Journal of Social and Clinical Psychology, 17(2), 196–208. [Google Scholar]
  79. Watson D, Clark LA, & Tellegen A (1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54(6), 1063–1070. [DOI] [PubMed] [Google Scholar]
  80. Whisman MA, Sheldon CT, & Goering P (2000). Psychiatric disorders and dissatisfaction with social relationships: Does type of relationship matter?. Journal of Abnormal Psychology, 109(4), 803–808. [DOI] [PubMed] [Google Scholar]
  81. Whiting SE, Jenkins WS, May AC, Rudy BM, Davis TE, & Reuther ET (2013). The role of intolerance of uncertainty in social anxiety subtypes. Journal of Clinical Psychology, 70(3), 260–272. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS1739920-supplement-1.docx (21.8KB, docx)

RESOURCES