Social comparison opportunities in digital health interventions: Are we using theory and evidence effectively?


By Dr. Dani Arigo, Rowan University and Rowan University School of Osteopathic Medicine (reposted from UCL’s Center for Behavior Change Digi-Hub blog, published May 24, 2018)

As a clinical psychologist who studies the role of social process and perceptions in health behavior change, I’ve been both delighted and overwhelmed by our increasing ability to use digital tools that connect users to each other. So much of our daily lives are spent engaged in social interaction (or pondering social experiences), and there is huge potential to use the social environment to promote health. But it seems that we still have a long way to go before we can capitalize on this potential.

For instance, literature on intervention components of digital health tools – particularly work that describes mobile apps for physical activity and weight loss – indicates that a significant proportion of these tools uses “social comparison” to promote behavior change. (See here for an example.) Yet “social comparison” can mean a lot of different approaches or techniques. When researchers say that social comparison is used in an intervention, what does this mean in practice? Moreover, is this work maximizing the vast body of knowledge on social comparison theory and applications to health contexts?

Social comparison refers to the process of self-evaluation relative to another person or group. The ability to size ourselves up next to potential allies or competitors is thought to be innate and protective; it often happens quickly and implicitly, whenever we’re exposed to relevant information about other people. As a result, any digital health intervention that offers the opportunity to learn about others’ health behaviors could facilitate social comparison. This means that features such as message boards, direct messages, profiles/feeds, or anything else that lets users share text or photos with each other could be categorized as social comparison.

But simply providing information about others may not actually induce social comparison, as people (and therefore, users of digital health tools) differ in their tendencies to make comparisons or value the evaluations they provide. Here, concluding that comparison is a primary mechanism of action could lead us astray; other mechanisms such as social support and social reinforcement may be more powerful in these situations, or may interact with social comparison to produce behavior change. Features such as challenges (competitions) and leaderboards, which directly compare a user’s progress to that of other users, are more likely to induce the types of comparisons that could lead to behavior change.

Unfortunately, many existing reviews of digital health interventions do not specify what they categorize as “social comparison,” making it difficult to determine which features of a tool are most beneficial. To make things even more confusing, these reviews rarely indicate how users’ information is communicated to each other – via raw numbers (pounds lost or steps taken), data visualization (graphs), photos, videos, or avatars? Or some combination? Which of these is most effective for promoting behavior change?

But perhaps the most significant limitation of existing work in this space is the assumption that all opportunities for social comparison are created equal – specifically, that they will always have the desired positive effect. Social psychology theory and empirical work show that this is not accurate. As noted, people differ in their affinity for social comparison, meaning that facilitating or inducing this process via digital health tool won’t be very effective for some people. Even more importantly, for some people, comparison has negative effects. Anxiety, despondence, and frustration all are fairly common responses to comparison, and only some people are able to channel this negative affect into motivation for behavior change; many of us simply give up. (This could explain some users’ disengagement from digital health tools.)

Finally, differences in affinity for comparison or positive/negative response to a comparison opportunity do not exist just between people. Each of us fluctuates in our interest and response over time, depending on factors such as mood, stress level, and progress toward our behavioral goals. We’re continually improving the capacity for digital health tools to assess users’ dynamic preferences and needs, and respond to these in real time (e.g., JITAI designs), but this capacity has not yet been applied to tools’ social comparison features. We now have enormous potential to tailor social comparison features, which may provide more engaging and effective digital health interventions.


  • What is the best approach to assessing social comparison preferences and needs, in order to inform social comparison tailoring in a digital health tool?
  • What steps are necessary in order to tease apart which social comparison intervention features and formats (e.g., text vs. photos) are most beneficial, for whom, and when?

@UofSHealthPsych at Healthier U Day (University of Scranton)


healthieruf161On Friday, September 16th, the Clinical Health Psychology Lab took part in the University of Scranton’s Healthier U Day event from 1:00-4:00 pm. Seven lab members introduced fellow students to the concept of health psychology, demonstrated its usefulness for behavior change, and provided information about ways to stay healthy on campus.

At 1:00, the line to enter Healthier U Day stretched the entire length of the Dionne Green. (If you’ve never been to our campus, this is about the size of a soccer field.) We were pleased to see so many students interested in learning more about health and wellness on campus. Our table was greeted with groups of 5 to 10 students at a time, who were eager to learn. To keep up with the flow of people, students were directed to start with our survey question and work their way through the rest of our table from there.

Our discussion began with the question, “What is your most common barrier to exercise?” We offered four options, and over 50 students responded: 64% said “I’m busy/have no time,” 18% said “I have no one to go with, ”18% said “The gym is too crowded/ I fear being judged,” and less than 1% said “I don’t know where to go.” These results demonstrate that time management seems to be the largest barrier to physical activity for college students. However, we observed the majority of students who reported fear of judgement or did not have anyone to go with were female, with the exception of males who were freshman. Such observations could lead us to new research questions about social support for exercise in these subgroups.

Based on the responses, students were directed to a visual web of solution stems, printed on a poster (pictured below). Solutions were recommended by lab members as methods that work for us in everyday life, so students got some insight into how we overcome the psychological barriers presented on the poster.

psychological-barriers-to-pa-healthier-u-day-2016We also introduced students to the types of studies and research questions that are conducted by our lab. We tried to make sure that the female students knew about Project CHASE, as we are recruiting for that study. We continued by giving students an overview of the field of health psychology. Students were given handouts, including exercise resources on and off campus and tips for healthy eating behaviors.

Our exercise resource sheet included information about off-campus resources and on-campus options other than the university’s gym. It included: The Jewish Community Center’s Group Exercise Classes, Yoga for Grief Relief, and Nay Aug Park, as well as  The Byron Center’s Open Swim and Intramural Teams. Students were surprised to see some of the options they had for physical activity in the area, and many seemed excited to take home a copy of the sheet. Some examples included in the healthy eating sheet included advice like “don’t eat and work” and “don’t completely take your favorite foods out of your diet”. These handouts were meant to increase convenience and thereby increase the likelihood of positive health behaviors.


From Left: Team Members Zuhri Outland, Marissa DeStefano, Kerri Mazur, & Sabrina DiBisceglie

After guiding students through the survey and suggestions for positive health behaviors, several people were interested in taking the Health Psychology course offered in the spring semester (PSYC 228). Many students were unfamiliar with the concept of health psychology beforehand, and were curious to learn more after visiting our table. Overall, we were pleased with the feedback we received at the event, and we hope our presentation will allow students to make healthier choices!

Contributors to this post: Marissa DeStefano, Zuhri Outland, Kristen Pasko, Sabrina DiBisceglie, Kerri Mazur, and Dr. Arigo.


Our Social Brains: What’s In It for Health?


Originally posted on July 13, 2014 by Dr. Arigo

One of my recent projects was a small pilot study to test the feasibility of a new physical activity program. Although there are multiple components of my physical activity program, the novelty is its emphasis on social connections. As you can see from previous posts, understanding the effects of the social environment on health and health behavior change is central to my work; I believe that it’s central to overcoming ubiquitous barriers to healthy behaviors such as eating well and quitting smoking. We have decades of evidence to support the relationship between a person’s social environment and his or her health. For example, one of the most elegant and powerful studies on this relationship happened in the 1980s, when Sheldon Cohen and colleagues demonstrated that social support can protect a person from catching the common cold.

Advancements in statistical modeling techniques (as well as access to large data sets) led to findings on social network effects – in essence, we become more like close others in appearance (weight) and habits (smoking) over time. Although the math in the observational (i.e., no experimental manipulation, an thus no inference of causation) network studies has been questioned, FaceBook’s recent experimental manipulation of our News Feed content has delivered the same conclusion: we are affected by others, and we become more like “close” others over time. (For the record: ethical problems noted here, but important work and clever execution. I may return to this in another post.)

Neuroscientist Mattew Lieberman’s Social: Why Our Brains are Wired to Connect takes this concept several steps further. The goal of Lieberman’s work at UCLA is to determine the neural underpinnings of our social behaviors. In Social, he describes his (and other) research at length, integrates it with personal stories, and makes suggestions for how we might be able to use it to improve societal problems. The book is repetitive at times, but it hammers home a few critical points:

(1) The brain’s “default” network – what happens when we’re not actively focusing on anything in particular – is social. When we don’t have something to do, we default to thinking about our relationships and trying to figure out others’ motivations.

(2) The brain’s “mentalizing system” likely is responsible for our ability to connect with and understand others.

(3) We are hard-wired to prioritize our social connections – they likely kept us safe from predators and starvation as we evolved. Accordingly, we can improve failing institutions like the U.S. public education system by increasing the social connectivity of students around learning (rather than restricting their interactions in the classroom and allowing their social energies to be used elsewhere).

The third point is most interesting to me, as the same could be said for health promotion. But there are several ways to facilitate social connections – which is/are most effective remains an open and compelling question.

For me, reading this book not only underscored the importance of the work I do, but it encouraged me to return to blogging. Sharing my thoughts and my work with others is the only way to make them useful.

The Body Problem


Originally posted on March 8, 2014 by Dr. Arigo

Around 2005, I decided to focus my work on disordered eating behavior and body image. Like many college women, I saw this broad topic everywhere. In addition to concerns among family and friends, body image, eating, and weight were plastered all over the media (and critiques of the media). During my gap year, I worked on an eating disorders unit at a psychiatric hospital, and I doubt that I could have been any more immersed in these issues.

Given the chasm between treatment and research in eating disorders, I was fortunate that the cultural interest in eating and weight included in the empirical literature. Professional journals do succumb to trends, and it seemed that body image research was fashionable at the time. Some of the classics include Crandall’s examination of social contagion of binge eating and Becker’s description of introducing disordered behavior to a population by giving them televisions. To me, some of the most interesting work was in mapping the temporal relationships between mood state, social and self-perceptions, and eating behavior, using electronic daily diaries. (I have yet to contribute to this area, though it is on the agenda for my first few years in my new position.)

Although there still is much to learn about these relationships, the fervor for body-related topics seems to have died down in the past few years. The first clue regarding this change came from my students in Abnormal Psychology, around 2010. Until then, I had been very much aware of the negative stereotype of young women with eating disorders (see Wasted and Skinny Bitch, for examples). When I asked my students to describe the stereotype, they surprised me: many of them seemed not to share this perception (including males), and expressed compassionate, considerate attitudes toward all people with these disorders. I suspect there were several reasons for this disconnect, including a high SES student population. But it seemed that times were a’changin, and I was behind in my own field.

Much has changed in the past ten years. Obesity is a bigger problem than ever, and the innovative work of Brian Wansink and Kelly Brownell has shifted attention to macro-level problems such as the food environment and public health policy. My own work has broadened to include the full spectrum of eating and weight disorders, with a heavier emphasis on obesity, diabetes, and weight loss than on body image and disordered eating. I’m also much more interested in the science of eating behavior – and how eating can be manipulated – than in collections of symptoms. But as I continue to write and publish in body image and disordered eating, I realize that I am farther from the pulse of these topics than ever. For example, I’m currently working with a student to revise her Master’s thesis for publication – a clever approach to help young women combat their immediate negative responses to thin media figures. Yet the Introduction to the “problem” of negative body image features a preponderance of references to literature from the 1980s and 90s, with virtually nothing newer than 2004.

Have we solved the body problem while I wasn’t looking? Or, have we come to accept it as such a part of our culture that we’re no longer doing interesting, novel research on it? Perhaps neither, or a little of both. In retrospect, part of the appeal of this research in the late 90s and early 2000s was its novelty; women were first starting to speak out about their experiences and disclose the severity of the consequences. We do have better treatments than ever (e.g., the Body Project, CBT for bulimia nervosa), though there is more to be done to optimize them. Members of our society are aware of the problem and talk more openly about it than they used to. And as obesity wreaks physical and psychological havoc, more of our resources go to understanding problems of excess weight and overeating (e.g., binge eating disorder).

I’m pleased to see the increasing unification of the eating disorders and obesity fields. From both research and clinical perspectives, what makes eating so interesting to me is that we cannot live without it (unlike heroin), but there are so many ways that it can go awry. And it’s entirely possible that each of the distinct problems I’ve noted stem from the same central disturbance. For example, a recent paper in Psychosomatic Medicine demonstrated that, in both healthy weight and overweight women, exposure to degrading treatment of overweight individuals in media clips led to increases in salivary cortisol (a stress hormone). Interpretations can and should vary, until further research is shared. It suggests to me that among women, there is something universally anxiety-provoking about the negative overweight stereotype. Some of us avoid the immediate distress of it by eating more, and some of us take drastic measures to prevent it from becoming reality. At present, too little is known about the development of truly healthy mindsets and habits. Let’s hope that’s the next trend.