The most important life goals of people with and without social anxiety disorder: Focusing on emotional interference and uncovering meaning in life

Abstract

People with social anxiety disorder (SAD) display maladaptive attitudes towards emotions. In this experience-sampling study, we explored the extent to which people with SAD viewed anxiety and pain as an impediment to pursuing personal strivings and deriving meaning in life. Participants were adults diagnosed with SAD and a control comparison group who completed baseline questionnaires and daily surveys for 14 consecutive days. People with SAD perceived anxiety and pain as interfering with progress towards their strivings to a greater degree than healthy controls. Perception of emotion-related goal interference was inversely associated with daily meaning. This relationship was moderated by diagnostic group such that there was a strong, inverse association with daily meaning in life for people with SAD; for controls, no association was found. Results suggest that negative beliefs about the value of anxiety and pain are pronounced in people with SAD and may impede derivation of meaning in life.

Theoretical models suggest that people with SAD display maladaptive emotion regulation patterns that interfere with positive experiences and meaningful life pursuits (Kashdan, Weeks, & Savostyanova, 2011). People with SAD view emotions as unhelpful and potentially harmful, as they worry that certain emotions may be visible (e.g., anxiety as evidenced by shaking hands) and subject them to unfavorable evaluation (Heimberg, Brozovich, & Rapee, 2010; Spokas, Luterek, & Heimberg, 2009). They hyper-focus on managing and eliminating uncomfortable thoughts, feelings, and behaviors (Kashdan et al., 2013, 2014). The motivation to avoid threat takes precedence over the motivation to pursue potentially rewarding experiences, which may this decrease the probability of deriving rewards from the world unfolding around them (see Richey et al., 2019 for a review). As a result, the experience of anxiety and pain might correlate with decreased progress towards meaningful life pursuits. The current study tested this hypothesis. Using experience-sampling methodology, we examined the extent to which people with and without SAD perceive anxiety and pain as an impediment to deriving meaning in life (MIL) on a daily basis when pursuing important life goals.

Emotion biases and social anxiety disorder

To understand emotional biases in SAD, consider the experience of anxiety. Anxiety can be adaptive. Increased heart rate and sweat prepares the body to take action, visually detecting a threat helps a person make predictions about how best to respond, and so on. Anxiety, within certain levels and when deployed appropriately, can help a person navigate daily life. For people with SAD, however, uncomfortable emotions like anxiety can be crippling and a barrier to pursuing and achieving meaningful life pursuits. This is likely due to a combination of biases in motivational systems, emotion beliefs, and avoidant emotion regulation patterns.

Two motivational systems regulate behavior—the behavioral activation (“promotion”) and behavioral inhibition (“prevention”) systems. The promotion system guides people towards potential rewards and gains with little concern for setbacks or errors, whereas the prevention system aims to prevent exposure to danger, loss, or failure by warning of imminent threat and signaling a need to escape (Carver & Scheier 1990, Strauman, 2017). People with SAD are characterized by an underactive promotion system and an overactive prevention system. For example, they demonstrate biased attention towards threatening cues, especially social threat cues (e.g., Amir, Freshman, & Foa, 2002), and greater neurological reactivity in brain networks associated with avoiding punishment than with those associated with obtaining social rewards (Cremers et al., 2014). Heightened sensitivity to avoiding threat and deficient motivation to pursue rewards may facilitate decision-making based on avoiding failure at the expense of approaching rewards.

In addition to motivational and attentional biases, people with SAD have maladaptive attitudes towards anxiety and anxiety-related sensations. They believe that somatic sensations (e.g., heart racing) could signal the onset of something catastrophic (e.g., heart attack) (Olatunji & Wolitzky-Taylor, 2009). They worry that people will negatively evaluate observable signs of anxiety or nervousness (McWilliams, Stewart, & MacPherson, 2000; Voncken, Alden, & Bögels, 2007). They also underestimate their ability to regulate their emotions (Goldin, Manber-Ball, Werner, Heimberg, & Gross, 2009; Werner et al., 2011). Together, heightened reactivity to threat and negative attitudes about the value of negative emotions like anxiety may facilitate problematic emotion regulation patterns.

People with SAD tend to display a regulatory profile perhaps best described as experiential avoidance. Experiential avoidance is defined as the negative evaluation of and unwillingness to remain in contact with unwanted internal thoughts, feelings, and sensations (Hayes, Strosahl, & Wilson, 1999). Frequent use of experiential avoidance is a risk factor for the development and maintenance of most anxiety disorders (for reviews see Chawla & Ostafin, 2007; Goodman, Larrazabal, West, & Kashdan, 2019) and specifically, is one of only two daily experiences (the other being low positive emotions) that differentiate people with SAD from psychologically healthy adults (Kashdan et al., 2013). The paradox of deploying experiential avoidance is that the very thoughts and feelings someone is trying to control or get rid of often come back with greater intensity and frequency (Kashdan et al., 2014; Wegner, 1994). When excessive effort and value is placed on controlling emotions, fewer cognitive and emotional resources are available to be fully present and construct meaningful experiences.

Taken together, research suggests that people with SAD perceive anxiety and pain as threatening and/or unhelpful, are motivated to avoid a negative outcome in a given situation (e.g., avoid unfavorable evaluation), deploy rigid strategies to avoid discomfort (e.g., experiential avoidance), all of which takes precedence over potentially rewarding approach-oriented goals (e.g., establish a friendship). This emotion regulation profile might interfere with the ability to behaviorally commit to important life goals. Goal pursuit typically requires the management of anxiety and pain. For example, if a person strives to give a keynote address, she will have to overcome nerves before getting on stage. If a person strives to be an exceptional father, he will have to manage the stress of caring for a sick child. People without anxiety disorders might view some degree of anxiety and pain as helpful (e.g., it helps them “get in the zone”), and adversity might increase motivation to overcome setbacks (e.g., Rothermund & Brandstadter, 2003). For someone with SAD, however, anxiety and pain may be viewed as an impediment to achieving goals. In the present study, we tested this hypothesis by examining the extent to which emotions interfered with the acquisition of MIL while pursuing important life goals, for people with and without SAD.

Emotion biases and meaning in life

Despite divergence in psychological science about what constitutes MIL, most conceptualizations suggest that pursuing and making progress towards important life goals is central to its development (e.g., Martela & Steger, 2016; Reker & Wong, 2012). One way to capture these types of goals empirically is to ask people to generate personal strivings—important life goals that people think about, plan for, and devote resources toward (Emmons, 1986). Strivings act as a compass to organize and guide people towards behaviors that facilitate MIL. Strivings pursuit can represent effort towards a purpose in life (e.g., working towards a striving that reflects a person’s central, fundamental life aim—Kashdan & McKnight, 2013), provide a sense that life makes sense (e.g., working towards strivings that are congruent with one’s values—Sheldon & Elliot, 1999), and increase a sense of significance in the world (e.g., pursuing a goal that can help other people—Sheldon & Kasser, 2001). Sometimes, however, the benefits of working towards strivings are compromised, such as when a person feels poorly equipped to be successful or is pursuing strivings that are incongruent with their values (Sheldon & Elliot, 1999). In the present study, we hypothesize an additional barrier to deriving MIL for people with SAD—the extent that anxiety and pain are perceived as interfering with progress toward personally important strivings.

When people with SAD perceive emotions as impediments, they are likely reinforcing negative beliefs about their (perceived) flawed self (Moscovitch, 2009). People with SAD construct self-concepts that are intertwined with symptoms. For example, they may associate anxiety with representations of who they are (“an anxious person”) rather than symptoms they are experiencing (“a person who feels anxious”) (Gamer, Schmukle, Lua-Krausgrill, & Egloff, 2008; Westberg, Lundh, & Jönsson, 2007). People with SAD are often self-critical and feel upset with themselves when experiencing uncomfortable emotions (Blalock, Kashdan, & Farmer, 2016; Kivity & Huppert, 2016). Thus, perceptions that anxiety and pain are interfering with their strivings might disrupt core components of meaning (Martela & Steger, 2016)—a sense of coherence (emotion-driven behavior is incongruent with their values), significance (they feel inadequate to pursue what matters to them), and purpose (discomfort disrupts progress towards core life goals).

The present research: Studying people in context

A challenge in psychological science is to study people “in context.” A contextual science involves understanding individual differences, environmental stimuli that elicit emotions and behaviors, changing situational goals, and varying effects on outcomes of interest (e.g., Aldao, 2013). In this study, we addressed “context” in two ways. First, we examined goal pursuit via participants’ unique strivings. Strivings are idiographic and formed within the context of individualized life experiences. People often construct strivings with similar themes, but their content is unique. For example, one person might strive to “build a trusting relationship with my daughter,” and another person might strive to “be a supportive husband to my wife.” Both strivings contain themes related to developing interpersonal relationships, but they differ in focus (trust vs. support) and relationship target (daughter vs. wife).

Second, we employed experience-sampling methodology to examine strivings pursuit in people’s daily lives. A person’s effort, success, and attitude towards their strivings fluctuate in response to changing environmental contingencies. Compared with global, single time-point trait measures, daily measures assess processes closer to real time, which minimizes recall bias (Reis & Gable, 2000). Even when trait and daily constructs are measured with similar or identical items, people use different information to respond to survey questions. For trait measures, participants aggregate information from salient and/or recent events to give their best guess of what they believe they typically do across situations. For daily/state measures, participants report on what they are doing in that moment (or earlier that day) with less consideration of what they (believe they) typically do. For MIL specifically, trait measures assess MIL as a global judgment (e.g., “I understand my life’s meaning”), whereas state measures reflect the quality of everyday existence (e.g., “In the last two days, I had a sense that I see a reason for me being here”) (see King, Hicks, Krull, & Del Gaiso, 2006). Although global ratings of MIL tend to be stable, experience-sampling studies have demonstrated that daily ratings of MIL can fluctuate on a daily basis (Steger, Frazier, Oishi, & Kaler, 2006; Steger & Kashdan, 2007) and are influenced by daily goal-directed efforts and other growth-oriented behaviors (Steger, Kashdan, & Oishi, 2008). Thus, to best understand how people with and without SAD operate in their everyday lives, we used daily measures for MIL and other study variables across a 14-day period.

Method

Participants

Participants were 84 community adults recruited from northern Virginia and the greater Washington, D.C. metropolitan area. Forty-one participants were diagnosed with Social Anxiety Disorder (SAD), generalized subtype (DSM-IV), and 43 participants were healthy controls currently absent of psychological disorders. Our target sample size was N=40 participants in each group (80 total). Between-group and within-person effect sizes were determined by drawing on meta-analytical and experience-sampling research on social anxiety (e.g., Kashdan, 2007; Kashdan & Wenzel, 2005). For between-group analyses, an effect size of r = .35 would provide power = .88 (Dupont & Plummer, 1990); for within-person analyses, an effect size of r = .20 would provide power ≥ .80 (Nezlek, 2011; Raudenbush & Lui, 2000).

Six participants did not provide end-of-day reports, yielding a final sample of 78 participants (38 with SAD, 40 healthy controls) with an average age of 29 years (SD = 8.7). In terms of comorbid conditions in the SAD group, six participants met DSM-IV criteria for current major depressive disorder, three for generalized anxiety disorder, two for OCD, one for panic disorder, and one for agoraphobia. Fifty participants were female, and self-reported race/ethnicity was as follows: 52.6% Caucasian/white, 21.8% African-American/black, 10.3% Hispanic/Latino, 3.8% Asian/Asian-American, 1.3% Middle Eastern, and 10.3% other. Participants in diagnostic groups did not differ on age, sex, or ethnicity (ps > .7).

Procedure

Study participants were recruited from the community via flyers and online advertisements. Prospective participants completed a preliminary phone screen with trained research assistants. Research assistants first provided informed consent, then conducted a semi-structured interview to assess for symptoms of anxiety, depression, suicidality and psychosis. The initial phone screen streamlined recruitment by identifying participants who endorsed symptoms suggestive of social anxiety (potential SAD group participant) or the absence of psychological disorders (potential healthy control group participant). Potential participants attended an initial assessment at our research laboratory. After research assistants provided a second informed consent, participants completed self-report questionnaires on a laboratory computer. Next, clinical psychology doctoral students administered the Clinical Interview for DSM–IV Axis I Disorders (SCID-IV) to assess for psychological disorders and the SAD module of the Anxiety Disorders Interview Schedule for DSM–IV: Lifetime Version (ADIS-IV) to gather additional information about social anxiety symptoms. Participants in the SAD group met DSM-IV criteria for SAD, generalized subtype. Comorbid diagnoses were allowed, so long as SAD was the primary and/or most severe diagnosis. Exclusionary criteria for all participants were current psychotic or substance use disorders. Participants in the healthy control group did not meet DSM-IV for any current mental health disorder. A random subset of 45 recorded interviews were listened by trained researchers to ensure accurate diagnoses. Reliability analyses indicated acceptable reliability (Cohen’s k = .87).

Following baseline questionnaires and semi-structured interviews, eligible participants completed the experience-sampling portion of the study. Participants completed a one-and-a-half-hour training session on the diary questions and palm pilot technology. Each participant received a hand-held Palm Pilot (X22) that was pre-programmed with diary questions using Purdue Momentary Assessment Tool (PMAT; Weiss, Beal, Lucy, & MacDermid, 2004). Each day for two consecutive weeks, participants received surveys with questions assessing mood, emotion regulation, well-being, and social functioning. Questions were distributed across three types of survey formats: end-of-day, event-contingent, and random prompts (i.e., ecological momentary assessment). For the purposes of the present research, we used only end-of-day reports.

Participants were compensated a minimum of $165. To incentive participant compliance during the diary portion, compensation was structured with participation bonuses. Participants earned $0.50 for each diary prompt they answered and a bonus $10 for each week of uninterrupted completed reports, totaling $50 in bonus payment and a maximum potential payment of $215.

Measures

Strivings assessment

Participants were provided a definition of strivings (“goals or purposes that motivate you; an objective that you are typically trying to accomplish or attain”) and illustrative examples (e.g., “trying to be a good role model to others”). They were instructed that their strivings could be positive or negative, approach- or avoidance-oriented, and within any domain. After reading these instructions, participants generated six strivings in open-ended format. After they generated their list, they identified their two most important strivings and were informed they would answer daily questions about these two strivings during the experience-sampling portion of the study.

Daily measures

Strivings ratings

Participants responded to six items each day about their two most important strivings. For each striving, they rated goal progress interference due to anxiety (“How much did anxiety and pain interfere with progress toward [Striving #1/2] today?”), effort (“How much effort did you put towards [Striving #1/2] today?”), and progress (“How much progress did you put towards [Striving #1/2] today?”). All items used Likert-scale response options from 0=none to 10=extreme amount. Items were retained as single-item measures rather than averaged within dimensions. Strivings differ in their relative influence on each other (e.g., Emmons & King, 1988), and effort, progress, and related outcomes for one striving may be weakly or unrelated to effort, progress, and related outcomes for another.

Meaning in life (MIL)

Daily MIL was measured with two items (“How meaningful did you feel your life was today?” and “How much did you feel your life had purpose today?”). Participants responded to items from 1=very uncharacteristic of me to 7=very characteristic of me today. These items have been validated in prior daily diary studies (Kashdan & McKnight, 2013; Machell, Goodman, & Kashdan, 2015; Steger, Kashdan, & Oishi, 2008). Positive associations with curiosity and life satisfaction demonstrate construct validity, and negligible associations with hedonistic behaviors demonstrate discriminant validity.

Emotion suppression

Daily use of suppression was measured with four items from the Emotion Regulation Questionnaire (ERQ; Gross & John, 2003), modified for daily surveys—“I kept my emotions to myself,” “When I was feeling positive emotions, I was careful not to express them,” “I controlled my emotions by not expressing them,” and “When I was feeling negative emotions, I made sure not to express them.” Items were rated on a Likert scale from 1=strongly disagree to 7=strongly agree. These items have been validated in prior daily diary studies on emotion regulation (Kashdan & Steger, 2006; Brockman, Ciarrochi, Parker, & Kashdan, 2017). Positive associations with negative affect demonstrate construct validity, positive associations with trait measures of emotion suppression demonstrate convergent validity, and negligible associations with trait cognitive reappraisal demonstrate discriminant validity (Kashdan & Steger, 2006; Brockman, et al., 2017).

Affect

Participants rated their affective experiences each day using twelve items. Six items measured positive affect (content, relaxed, enthusiastic, joyful, proud, and interested) and six items measured negative affect (anxious, angry, sluggish, sad, irritable, distressed). Participants used a 5-point Likert scale from 1=very slightly/not at all to 5=extremely. These emotion adjectives include representation from each of the four dimensions of the circumplex model of emotions (high and low arousal, positive and negative valence—Barrett, 1998) and reflect items from the Positive and Negative Affect Schedule (PANAS; Watson, Clark, & Tellegen, 1988) used in prior experience-sampling studies (e.g., Nezlek & Kuppens, 2008).

Results

Analytic plan and descriptive statistics

Data were conceptualized as hierarchically nested, with days (Level 1) nested within people (Level 2). Analyses were conducted using a series of multilevel models using Hierarchical Linear Modeling (HLM) Version 7.01. Continuous Level 1 variables were entered group-mean centered, continuous Level 2 variables were entered grand-mean centered, and dichotomous variables were entered uncentered.

Reliability estimates for multi-item daily measures were calculated using unconditional three-level models with items nested within days within people (Nezlek, 2011). Reliability for positive affect (.64), negative affect (.60), and MIL (.88) was acceptable; reliability for emotion suppression was low (.37). We also examined the variance partitioning for each measure by examining unconditional models and determining variance attributable to between-person variability and the variance attributable to within-person variability. Daily measures demonstrated sufficient within-person variability (relative to total variability) for daily positive affect (35%), negative affect (59%), MIL (38%), and emotion suppression (34%), as well as all strivings measures: effort (56%), progress (54%), and goal progress interference due to anxiety (43%) for participants’ most important striving, and effort (50%), progress (45%), and goal progress interference due to anxiety (48%) for participants’ second most important striving. These results suggest that construct vary considerably within-person and provide additional support for conducting analyses using intensive repeated assessments.

Primary analyses

Group differences in daily meaning and goal progress interference due to anxiety

Compared with healthy adults, people with SAD reported lower daily MIL (b = −1.44, t = −5.36, p < .001) and greater goal progress interference due to anxiety for their most important (b = .92, t = 4.26, p < .001) and second most important striving (b = .53, t = 2.57, p < .05).

Goal progress interference due to anxiety and meaning in life

Our primary analyses reflect “slopes-as-outcomes” in which we examined the effect of Level 2 SAD diagnosis on the Level 1 slope (i.e., association between daily goal progress interference due to anxiety and MIL). We first confirmed that daily goal progress interference due to anxiety predicted daily MIL, then examined the cross-level interaction of SAD diagnostic group. Significant interactions were probed by examining simple slopes.

Daily goal progress interference due to anxiety was inversely related to daily MIL for participants’ most important striving (b = −.21, t = −4.08, p < .001) and second most important striving (b = −.11, t = −1.96, p = .05), although this was result was just above traditional significance value cutoffs. SAD diagnosis significantly moderated the effect of daily anxiety on MIL for participants’ most important striving (b = −.11, t = −2.40, p < .05), but not for their second most important striving (b = −.08, t = −1.73, p = .08). Analyses of simple slopes found that for people with SAD, goal progress interference due to anxiety was strongly, negatively associated with daily MIL (β = −.28, t = 4.26, p < .001). For healthy controls, goal progress interference due to anxiety was unrelated to daily MIL (β = −.07, t = 1.05, p = .29).

To rule out the possibility that this relationship is driven by energy devoted towards one’s goals, we ran separate models controlling for daily effort and progress made towards strivings. Daily goal progress interference due to anxiety remained inversely associated with daily MIL after controlling for effort towards participants’ most important (b = −.24, t = −4.85, p < .001) and second most important striving (b = −.15, t = −2.86, p < .01), as well as after controlling for progress toward participants’ most important (b = −.21, t = −4.50, p < .001) and second most important striving (b = −11, t = −2.20, p < .05). Moreover, SAD remained a significant moderator of the relationship between goal progress interference due to anxiety and MIL after controlling for daily effort towards participants’ most important (b = −.10, t = −2.28, p < .05) and second most important striving (b = −.09, t = −2.06, p < .05), as well as progress toward participants’ most important (b = −.10, t = −2.34, p < .05) and second most important striving (b = −.09, t = −2.08 p < .05).

Secondary analyses

Construct specificity: Controlling for affect

MIL judgments tend to be influenced by a person’s mood at the time they provide reports (Hicks & King, 2007, 2009; King et al., 2006). To account for potential mood effects on MIL, we ran a series of models examining relationships between MIL and affect. In addition to broader negative affect, we conducted supplementary analyses using a single-item measure of anxiety to determine if results were attributable to anxiety in particular.

Main effects of SAD predicting MIL remained significant when controlling for positive affect (b = −1.44, t = −5.36, p < .001), negative affect (b = −1.44, t = −5.35, p < .001), and anxiety (b = −1.43, t = −5.21, p < .001). Similarly, the moderation effect of SAD on goal progress interference due to anxiety and MIL for participants’ most important strivings remained significant when controlling for positive affect (b = −.10, t = −2.68, p < .01), negative affect (b = −.10, t = −2.47, p < .05), and anxiety (b = −.10 t = −2.42 p < .05). For participants’ second most important striving, SAD diagnosis moderated the relationship between goal progress interference due to anxiety and MIL after controlling for positive affect (b = −.08, t = −2.08, p < .05) and negative affect (b = −.09, t = −2.11, p < .05), but not for anxiety (b = −.09, t = −1.93, p = .054).

Construct specificity

Controlling for emotion suppression

It is possible that results are a function of broader emotion suppression, a form of avoidant coping that is not necessarily linked to a specific goal. People with SAD frequently use emotion suppression, a strategy that tends to be ineffective at reducing distress and inversely associated with positive emotions and well-being (Ehring, Tuschen-Caffier, Schnülle, Fischer, & Gross, 2010; Kashdan & Steger, 2006). We ran a series of models testing all primary analyses controlling for emotion suppression. The main effect of SAD on meaning remained negatively correlated (b = −1.44, t = −5.36, p < .001), and the effect of goal progress interference due to anxiety moderating the effect of SAD on MIL for participants’ most important strivings remained significant (b = −.11, t = −2.45, p < .05). The moderation term remained non-significant for participants’ second most important striving (b = −.08, t = −1.73, p = .08). Thus, after accounting for the degree of emotion suppression that participants deploy on a given day, goal progress interference due to anxiety still predicts daily MIL, and SAD diagnosis still moderates this relationship.

Discussion

Theoretical models suggest that people with SAD have difficulty deriving meaningful experiences because they hold negative attitudes toward emotions and disproportionately focus on trying to control and/or get rid of unwanted emotions (Kashdan et al., 2011). As a result of excess focus on emotion regulation, people with SAD may be less equipped to attend to potentially meaningful experiences as they unfold and have fewer available resources to pursue valued goals. In the present study, results suggest that over the course of 14 days, people with SAD perceived anxiety and pain as interfering with progress towards important life strivings to a greater degree than healthy controls. For people with SAD, perceptions of goal progress interference due to anxiety predicted less MIL on a given day; this was true regardless of how much effort or progress they made towards their strivings. The impact of goal progress interference due to anxiety appears specific to SAD; for participants in the healthy control comparison group, daily goal progress interference due to anxiety was unrelated to daily MIL.

Emotion biases in SAD

People with SAD tend to evaluate anxiety and other putatively negative emotions as harmful, bad, and/or not useful. Empirical support for emotion biases has been demonstrated in self-report research (e.g., McWilliams et al., 2000; Olatunji & Wolitzky-Taylor, 2009; Voncken, et al., 2007), experimental research on attentional biases to internal and external threat cues (e.g., Garner, Mogg, & Bradley, 2006, Pines & Mineka, 2005) and neurological research on reward sensitivity (e.g., Mueller et al., 2009; Wieser, McTeague, & Keil, 2011). Findings from the present research build off this work by demonstrating that these biases have some degree of specificity, in that people with SAD believe anxiety and pain are interfering with their most important strivings. This was true regardless of how much negative affect they experienced on a given day.

To be clear, although adults in the control group did not meet criteria for mental health diagnoses, this does not mean they are devoid of anxiety or emotions interfering with goal pursuit. A “healthy” control group should be interpreted as people who can experience undesirable negative emotions, setbacks, dysfunction or impairment, or other forms of distress. Likewise, the experience of anxiety or impairment (e.g., disruptions in goal pursuit) alone is insufficient to warrant a diagnosis of social anxiety disorder or any mental health disorder. As such, prior to empirical testing as done in this study, we cannot assume that people with and without SAD experience different levels of goal progress interference due to anxiety, broadly or within specific domains such as goal pursuit. Our work did not simply show that people with anxiety disorders believe anxiety is a problem, as we provided direct evidence of the effects of these beliefs on day-to-day MIL deficits. This within-person association extends prior work on the unique positivity deficits linked to SAD compared with other emotional disturbances (e.g., Kashdan, 2007)

People with and without SAD differed in the relationship between goal progress interference due to anxiety and MIL extracted during the pursuit of personally derived life goals. Specifically, daily goal progress interference due to anxiety was inversely related to daily MIL for people with SAD, but unrelated to MIL for healthy controls. One potential explanation is that when people with SAD believe anxiety is getting in the way of pursuing what is most important to them, they are reinforcing negative self-beliefs; they are given further evidence that anxiety is a fixed part of who they are, a person who is inherently flawed (Moscovitch, 2009). In contrast, a healthy adult may be able to bounce back from setbacks with little to no lasting effect on their well-being. They might be better able to respond with flexibility (Kashdan & Rottenberg, 2010), such as generating multiple strategies to continuing making progress towards their goals (Goodman, Disabato, Kashdan, & Machell, 2017). For people with SAD, goal progress interference due to anxiety may offer external evidence for biased beliefs about who they are; for healthy adults, goal progress interference due to anxiety may be view an expected, benign part of goal pursuit.

It is interesting to note that for people with SAD, goal progress interference due to anxiety was linked with decreased MIL regardless of how much anxiety or total negative (or positive) affect they experienced. Of note, however, when controlling for anxiety, the relationship between goal progress interference due to anxiety and MIL for their second most important striving was no longer significant. As such, we offer tentative explanations but caution against over-interpretation of results. One explanation is that people with SAD hold fixed, negative beliefs about emotions across diverse contexts. Research on experiential avoidance offers indirect support, as people with SAD use experiential avoidance more frequently than non-anxious peers (e.g., Kashdan et al., 2013, 2014). Alternatively, people with SAD may pursue strivings that are more likely to be infringed upon by anxiety. There is some research to suggest that people with SAD construct strivings around managing their emotions to a greater degree than non-anxious peers (Goodman, Kashdan, Stiksma, & Blalock, 2019). If a person strives to regulate their emotions (e.g., “feel less anxious around other people”), then emotions may be viewed as a direct barrier to achieving that goal. Nonetheless, the linguistic content of participants’ strivings was not explored in the present study, and more research is needed to understand the complexity of these relationships.

Limitations and future directions

A challenge for social anxiety research is disentangling perceived versus actual judgments about oneself and one’s functioning/performance. People with SAD form overly negative judgments about themselves and overestimate the probability of negative social outcomes (e.g., rejection), but they may also elicit actual negative outcomes (e.g., Creed & Funder, 1998; Heerey & Kring, 2007; Taylor & Alden, 2011). In this study, our measure of goal interference is based on subjective judgments, not objective and/or performance ratings. Although this distinction plagues many self-report measures, items with greater specificity could minimize the influence of self-perception biases (e.g. “I avoided tasks related to [my striving] because I felt too anxious to work on them”). Similarly, future research can incorporate measures specific to social anxiety (e.g., “My social anxiety is out of control” from the Multidimensional Assessment of Social Anxiety—Gros, Simms, Antony, & McCabe, 2012). In addition to incorporating different types of self-report measures, future research can include informant reports from close friends, romantic partners, and/or strangers. By gathering other people’s perspectives, researchers can more precisely distinguish (potentially biased) self-perceptions from objective impairment.

In this study, MIL was the primary outcome of interest given its strong link with strivings and goal pursuit. Given that people with SAD are characterized by deficits in other facets of well-being beyond MIL (Goodman, Doorley, & Kashdan, 2018), future research can explore the influence of goal progress interference due to anxiety on other well-being facets, such as satisfaction with life (Diener, Emmons, Larson, & Griffin, 1985), psychological needs satisfaction (Sheldon & Hilpert, 2012), and quality of life (Frisch, Cornell, Villanueva, & Retzlaff, 1992).

Conclusion

To better understand how and why people with SAD experience difficulty managing their emotions and achieving valued goals, we need a precise understanding of the beliefs, values, and mindsets held about adversity that arises during goal pursuit. Findings from the present study extend prior research on attitudes towards anxiety, suggesting that people with SAD perceive anxiety as an impediment to making progress towards their most important life goals, i.e., strivings. Findings from the present study also extend prior research on positivity deficits to show that people with SAD derive fewer psychological benefits each day when pursuing their goals. Ultimately, the beliefs that arise during the course of SAD appear to be costly, impeding the development of meaning and purpose in life and perhaps other fundamental elements of well-being.