Social anxiety has a “dark side” that looks nothing like shyness

Social anxiety is commonly associated with shyness, silence, and a tendency to withdraw from social interactions. However, new research suggests that for some adolescents, this condition manifests through aggression and impulsivity rather than avoidance. This “atypical” presentation appears linked to specific narcissistic traits. The study was published in the journal Personality and Individual Differences.

“There is a prevailing assumption in the popular and professional literature that social anxiety is characterized solely by avoidance of tendencies and behavioral inhibition (i.e., shyness). This is likely a consequence from its formal classification of social phobia, which inadvertently shaped the way we study and understand the clinical phenomena,” explained study author Mollie J. Eriksson, a PhD Candidate in Louis Schmidt’s Child Emotion Lab at McMaster University.

“Nonetheless, this prototypical inhibited presentation does not reflect the lived experience of many individuals with social anxiety symptoms (for a comprehensive review see Kashdan & McKnight). And so, in the current study we aimed to examine the externalizing correlates of social anxiety that are less studied and correspondingly less understood, particularly in a population (i.e., adolescents) in which these dynamics might be especially conspicuous.”

The research team recruited 298 adolescents for the study. The participants ranged in age from 12 to 17 years old. The sample was nearly evenly split between boys and girls. Data was collected through a series of online self-report questionnaires.

Participants answered detailed questions regarding their feelings of social anxiety and their levels of narcissism. The narcissism measure distinguished between vulnerable and grandiose traits. Additional surveys assessed impulsivity and general aggression.

The researchers used a statistical method known as Latent Profile Analysis to group the participants. This technique identifies distinct categories of people based on patterns in their responses. “This is a very robust statistical technique because it uncovers patterns in the data that reflect individual variation in people and not simply associations between data points,” Eriksson said.

The analysis revealed three distinct profiles among the adolescents. The largest group comprised about 46 percent of the sample. These individuals displayed low levels of social anxiety, narcissism, and aggression. This profile appears to represent a well-adjusted or normative group with few social difficulties.

The second group accounted for approximately 30 percent of the participants. Adolescents in this profile reported the highest levels of social anxiety. They also scored high on vulnerable narcissism but low on grandiose narcissism and aggression. This group fits the prototypical description of social anxiety. These teens appear to manage their fear of rejection through inhibition and withdrawal.

The third group made up roughly 25 percent of the sample. This profile was characterized by moderate levels of social anxiety but high levels of impulsivity and aggression. Notably, these adolescents scored the highest on both vulnerable and grandiose narcissism. This combination of traits represents the “atypical” presentation of social anxiety.

“Social anxiety is a broad and heterogeneous mental health problem that is characterized by several features, beyond shyness,” Eriksson told PsyPost. “Recognizing its heterogeneity is the first step in identifying individuals, particularly adolescents, who may be struggling with social anxiety. By providing the tools (i.e., additional features that characterize social anxiety) we can intervene sooner, ideally before symptoms become entrenched, which will ultimately set the adolescent up for greater intra-personal and inter-personal success later in life.”

The researchers also found sex differences in profile membership. Boys were more likely than girls to belong to the third, aggressive profile. This suggests that boys may be more prone to expressing social fears through externalizing behaviors. This aligns with broader socialization norms where boys may be discouraged from showing vulnerability.

“It was exciting that these results replicated previous adult findings, which really underscores the robustness of these findings,” Eriksson said. “Even though this was in line with our a priori hypothesis, it was also interesting that boys were more likely to be in the ‘moderate social anxiety/high externalizing profile.’ It makes me think about how sex/gender influence the expression of social anxiety.”

But the study, like all research, has some limitations. The data was collected at a single point in time. This prevents researchers from establishing a causal relationship between narcissism and the development of aggressive social anxiety. It is unclear if the personality traits precede the anxiety or if they develop concurrently.

“A common misinterpretation we would like to preempt is the assumption that these profiles represent fixed or diagnostic categories,” Eriksson explained. “Rather, they reflect patterns of co-occurring traits and symptoms within a specific developmental window. Additionally, because the data are cross-sectional, we cannot infer developmental pathways or causal mechanisms. Replication (particularly in longitudinal designs) is therefore essential for understanding how these profiles emerge and change over time.”

Tracking these traits from childhood into adolescence could reveal early warning signs. Identifying these patterns early could lead to more effective interventions. Standard treatments for social anxiety may not work for teens who react with aggression rather than fear.

“I hope to examine early childhood antecedents of atypical social anxiety symptomology both behaviorally and biologically,” Eriksson said. “This will really inform treatment and prevention efforts. I also hope to examine in more detail the novel hypothesis we articulated: social anxiety is driven by two divergent self-regulatory pathways. This hypothesis requires a longitudinal study design, which is something we plan to do in the very near future.”

The study, “Characterizing the dark side of social anxiety in adolescence: A replication and extension study,” was authored by Mollie J. Eriksson and Louis A. Schmidt.