A New Approach to Understanding Social Anxiety

There’s more to social anxiety disorder than its symptoms, says new research.

Posted Mar 24, 2018

Source: Dimedrol68/Shutterstock

People with an extreme fear of social situations, if this fear is disabling enough, may have the diagnosable condition of social anxiety disorder (SAD). The symptoms that can become the basis for this disorder include extreme anxiety in any of a number of situations, ranging from eating a meal in front of other people to standing up and speaking to a group. The “anxiety” in this condition isn’t an actual fear of people (i.e., it is not a phobia), but instead reflects a fear of any number of social situations. The fear, in turn, leads people to avoid situations in which they’ll become overwhelmingly anxious. However, some situations that might provoke social anxiety are unavoidable. You can hardly go your entire life without having to eat in front of other people, whether it's a formal, work-related event or a rowdy family get-together around the holiday table. Similarly, there will be times you will have to stand up and speak in front of others, even if it's just to offer some words of welcome to a new member of a community group.

From a diagnostic standpoint, Harvard psychologists Alexandre Heeren and Richard McNally (2018) note that psychiatry views SAD as either a discrete categorical entity (meaning that you have it or you don’t) or as a continuum, where the more social situations you fear and avoid, the more impaired you are. They believe that these approaches are flawed, because they fail to take into account the relationships among social anxiety's particular symptoms. Instead, according to the “network” approach they recommend, SAD is “an emergent phenomenon that arises from causal interactions among symptoms." You don't just have one, two, or five situations in which you fear public behavior. The situations cluster together in related ways, forming the nodes of a network. Within such a network might be your fear of, and avoidance of, meeting strangers. This node would be connected to the node of fear of going to a party or calling people you don’t know very well. The fear of new social situations, they maintain, is distinct from the node representing, in their examples, avoidance of test-taking and fear of taking goods back to the store.

You can think of this network model in terms of central joining points connecting related symptoms by straight lines. The more closely connected different nodes are within the system, the greater the severity of the individual’s disorder. If social anxiety nodes are closely connected, you will feel fearful not only of eating in public, but also of speaking in front of your co-workers at that meeting. If fear of eating in public sits off by itself and is relatively distant from any other symptoms of the disorder, then you’ll be less incapacitated overall. If you’d like a metaphor for understanding how this network model works, the authors suggest imagining a diagram showing the intricate relationships among characters of different tribes in Game of Thrones.

The Harvard researchers maintain that “the network approach has ignited an explosion of research on a wide range of disorders," but has received little attention in the area of SAD. In their study, the authors used the network approach to differentiate people with the disorder from those who do not meet the diagnostic criteria for SAD. The analytic approach they used in the current study involved constructing lines among nodes in that elaborate Game of Thrones-like network based on the size of correlations among each of a set of 24 specific fears and their 24 corresponding avoidant behaviors.

The 238 participants receiving a SAD diagnosis ranged in age from 18 to 66 years and were compared with a comparable group of individuals who were free of the diagnosis. A 24-item social anxiety scale asked participants to rate both their fear and avoidance of everyday situations, in addition to those mentioned above, such as telephoning in public, working while being observed, resisting a high-pressure salesperson, urinating in a public bathroom, talking to people in authority, and entering a room where others are already seated. To measure the node centrality of each of the 24 situations, the authors constructed measures of the number of times a node lay on the shortest path between two other nodes, the average distance of a node from all the others in the network, and the sum of all weights showing the correlation of a node with the ones connecting to it. The overall global network strength then became the key variable of interest in determining whether those with the SAD diagnosis had more nodes closely connected than people who did not have the SAD diagnosis.

Again, keep in mind that the premise underlying the study is that people with an actual SAD diagnosis would have a number of interconnected symptoms (not just more), compared to people without the diagnosis, who might have a prominent symptom node or two relatively independent of all the other possible symptom nodes. You might hate having to use a public bathroom to urinate, but enjoy talking to people you don’t know that well or even be an inveterate product-returner at your local department store.

Heeren and McNally summarize their study’s main conclusion as follows: “the chief difference between a person with SAD and a shy person without the disorder is that the probability of fearing (and avoiding) one situation more strongly predicts fearing (and avoiding) another situation” in the person with SAD. People with SAD don’t fear different situations than people without SAD, but their fears are more strongly interconnected. This means, then, that it’s harder to intervene to reduce the maladaptive symptoms of a person with SAD than a person who has an isolated distaste for being in the public eye.

One striking similarity between people with SAD and people who do not have SAD is that the central nodes involve interactions with unfamiliar people, including just looking eye-to-eye at an unfamiliar person. It’s possible, based on this finding, that individuals with SAD had early experiences that reinforced their fear of strangers, and that this core set of fears triggers the others also seen in people with the disorder.

The treatment for people with SAD, according to this network model, should target their most central nodes of fear and avoidance. This means that therapists would first need to build the model of interconnections shown by their particular clients and then delve into the nodes that are most interwoven within the system of symptoms as a whole.

If you’re trying to overcome your own social anxiety, you can start by recognizing that it’s the fear of unfamiliar people that is likely at the core of your symptoms. You could work on improving your public-speaking skills, for example, but this won’t address your primary fear of having to put yourself out there in front of people you don’t know that well. Once you do, you may see positive benefits start to emerge almost on their own. As the authors suggest, “turning off a highly connected node may foster a beneficial cascade of downstream benefits” that could deactivate other, less central, nodes in the system.

The good news is that the network approach suggests a practical set of steps that you can take advantage of in therapy, if you in fact have this diagnosis. Fulfillment in social situations often does involve meeting new people and taking risks in front of them. Teasing apart the source of your fears, and tracing them to their roots, may just be the first step toward overcoming them.

LinkedIn Image Credit: fizkes/Shutterstock


Heeren, A., & McNally, R. J. (2018). Social anxiety disorder as a densely interconnected network of fear and avoidance for social situations. Cognitive Therapy and Research, 42(1), 103-113. doi:10.1007/s10608-017-9876-3