Personal and Ubiquitous Computing

, Volume 18, Issue 7, pp 1737–1752 | Cite as

Reflection-through-performance: personal implications of documenting health behaviors for the collective

Original Article


Previous work has examined how technology can support health behavior monitoring in social contexts. These tools incentivize behavior documentation through the promise of virtual rewards, rich visualizations, and improved co-management of disease. Social influence is leveraged to motivate improved behaviors through friendly competition and the sharing of emotional and informational support. Prior work has described how by documenting and sharing behaviors in these tools, people engage in performances of the self. This performance happens as users selectively determine what information to share and hide, crafting a particular portrayal of their identity. Much of the prior work in this area has examined the implications of systems that encourage people to share their behaviors with friends, family, and geographically distributed strangers. In this paper, we report upon the performative nature of behavior sharing in a system created for a different social group: the local neighborhood. We designed Community Mosaic (CM), a system with a collectivistic focus: this tool asks users to document their behaviors using photographs and text, but not for their own benefit—for the benefit of others in their community. Through a 6-week deployment of CM, we evaluated the nature of behavior sharing in this system, including participants’ motivations for sharing, the way in which this sharing happened, and the reflexive impact of sharing. Our findings highlight the performative aspects of photograph staging and textual narration and how sharing this content led participants to become more aware and evaluative of their behaviors, and led them to try to eat more healthfully. We conclude with recommendations for behavior monitoring tools, specifically examining the implications of users’ perceived audience and automated behavioral tracking on opportunities for reflection-through-performance.


Health monitoring Reflection Performance Social computing Community computing Nutrition 



Community Mosaic


Human–computer interaction


Information and communication technologies

1 Introduction

As researchers continue to envision how technology can transform healthcare, one trend has been to design systems that help people become more aware of their current health-related behaviors [3, 12, 34, 45, 48]. Some systems support personal monitoring and reflection1 (e.g., [13]), while many others support these activities in a social setting [3, 4, 8, 26, 32]. Much of the research on social health monitoring has examined how systems can support tracking and reflection in small and large family and friend networks (e.g., [32, 69, 70]) and even among geographically distributed strangers [28, 46]. For example, PatientsLikeMe is a website that allows individuals with serious health problems to document and share detailed information about their conditions, medication regimes and disease progressions with strangers who have the same condition [23]. Users are encouraged to document not only for their own sake, but so that others might learn from their trajectory of experience.

In a similar vein to PatientsLikeMe and motivated by the extensive public health literature showing the importance of community-based interventions, particularly in low-income contexts [52], we developed a system called Community Mosaic (CM). CM aims to help residents work together to overcome local barriers to healthy eating. The novelty of CM is its collectivistic premise: The system encourages users to document how they are trying to eat healthfully specifically and primarily for the benefit of others. As such, CM minimizes individualistic features; for example, the system does not facilitate the personal monitoring of one’s behaviors or the historical visualization of these behaviors as other tools do [12, 29]. While such features are incredibly useful, we were motivated by prior work suggesting the value of collectivism in ethnic minority populations, the primary demographic in our target neighborhoods [36, 37]. As such, designing CM allowed us to explore the implications of an interface that prioritizes visualizing the collective healthy eating successes of the group and group reactions to these successes. Furthermore, motivated by prior public health research that empowers community residents to encourage behavior change [61, 64], our goal was to study how technology can help people care for their communities by sharing their behaviors in a way that hopefully inspires others to eat healthfully.

In our previous work [60], we discussed how sharing messages in CM helped users develop their identities as health advocates. The contribution of the current paper is our analysis of additional benefits of message production and sharing for the content creators. In particular, we discuss how the tasks involved in documenting one’s behaviors for the benefit of others helped users become more aware and analytical of their behaviors and compelled them to make small changes to their own diet. In addition, we discuss the performative nature of content contributors’ interactions with CM and reflect upon the connections between this performance and the personal benefits that users gained from sharing.

This work extends previous research describing how technology can enable social networks to provide support for health management, and research documenting the ways in which identity is constructed and managed in such systems [4, 44, 56]. In particular, our contribution to human–computer interaction (HCI) research on health is twofold. First, we provide an analysis of how and why people contributed to CM and the reflexive impact of those contributions. While prior work has shown how technology can scaffold reflection on health behaviors, our work is unique in its exploration of collectivistic motivators for behavior sharing, the resulting performative documentation that ensues, and the personal impact of this documentation and sharing. Second, we conclude with a discussion of important directions for research on behavior monitoring tools, specifically examining the implications of users’ perceived audience and automated behavioral tracking on opportunities for reflection-through-performance.

2 Related work

In this section, we characterize the variety of ways that information and communication technologies (ICTs) have helped people to monitor their behaviors and the behaviors of others. In addition, we overview prior work analyzing the performative nature of behavior documentation and sharing in online contexts.

2.1 Health monitoring in social contexts

Researchers in HCI and related disciplines have examined how technology can help people monitor and manage their health in a variety of social contexts. Systems have been designed to support aging in place; these technologies help elders reside in their homes longer [17]. This work has explored systems that provide caregivers with information about elders’ well-being through medication monitoring, fall detection, and activity awareness [14, 17, 54]. Monitoring the elder in this way can help facilitate peace of mind for family members, because they have a more persistent awareness of the elder’s health. In addition, early detection of problems is critical for successful intervention [14, 17, 54]. Other research has focused on the technology-supported management of chronic diseases such as diabetes within families and between patients and educators [22, 48, 69]. These systems help users record and reflect upon relevant information such as their diet, physical activity, and blood glucose levels. Visualization and sharing of these data and communication with the caregiving or medical team can help patients learn the relationship between their behaviors and their physiological outcomes, and to identify when changes need to be made [22, 48, 69].

Another body of work has examined how ICTs can help people monitor their behaviors to prevent the onset of disease. Tools have been developed to support both the tracking of behaviors such as healthy eating, physical activity, and sleeping [3, 10, 11, 30, 31, 32, 33]. These systems help users set and visualize progress toward behavioral goals, with the intention of increasing users’ awareness of their behaviors and supporting improved behaviors. To facilitate tracking, researchers have leveraged self-report and automated methods of recording behaviors, such as systems that allow users to take photographs of their meals [4, 22, 57], and those that utilize phone sensors (e.g., accelerometers) and custom platforms for monitoring physical activity [3, 13, 34]. More recent work has moved beyond behavior tracking to biometric data collection; the HeartLink system, for example, allows runners to share their heart rate date with their social networks in real time [15]. This system allowed the social network to cheer on the runner during a run (by pressing a “cheer” button in an online interface); the runner’s phone would then vibrate to alert him or her to this encouragement.

The HeartLink system exemplifies a trend within this health monitoring design space, whereby systems facilitate social influence. A number of tools allow users to share their physical activity levels with others in their social network, typically friends [11, 34, 70]. Still other systems allow users to share their food and drink consumption with others [4, 8, 26]. By making friends and family members aware of one’s behaviors, these systems are designed to encourage healthy behaviors through the provision of accountability mechanisms, friendly competition, and social support through the sharing of encouraging messages [4, 11, 70]. Other work has examined how the wisdom of the crowd can be leveraged through crowdsourcing features that support quick evaluations of food logs [47, 57].

In the projects discussed so far, monitored health information has been shared with close ties: family, friends, and caregivers. Other research has examined the implications of communicating with strangers in online health communities [28, 40, 46, 56]. In these communities, members use discussion boards and chat rooms to share experiential knowledge (i.e., advice grounded in previously gained knowledge or first-hand experiences) and emotional support (e.g., expressions of empathy and encouragement in response to the challenges that community members share) [6, 21, 40, 65]. These communities can facilitate access to a broader network of individuals who are dealing with the same struggles, provide more persistent access to social support, and the anonymity afforded can also make it easier for members to disclose information [40, 41, 46, 49].

PatientsLikeMe is an online community where individuals with serious health conditions go beyond sharing narratives to recording and sharing very detailed quantitative data about their health conditions, such as comprehensive logs of diagnoses, medication usage, and disease symptom and progression information [23]. By sharing this information, patients help others to learn from what has and has not worked for them, which can be of tremendous help to individuals battling life-altering diseases.

In summary, technology has been used in several ways to support wellness and disease management in social contexts. This work has ranged from connecting close ties to connecting strangers all over the world. However, little work has examined the implications of health behavior and information sharing with weak ties that one shares an affiliation with as a result of living in the same geographic area. This is a surprising gap in research given the vast body of research showing the value of community-based participatory interventions for those living in the same neighborhood [1, 18, 24]. These interventions are important because the neighborhood environment can powerfully influence individuals’ desire and ability to pursue wellness [9, 18, 24, 27, 53], as such connecting individuals who are facing the same neighborhood challenges can be both motivating and encouraging. Researchers have designed community-based interventions to help residents learn to overcome barriers such as limited access to healthy foods, increase their self-efficacy for making behavioral changes, and strengthen relationships among neighborhood residents, which can have a protective effect by establishing connections that facilitate the sharing of social support [18, 19, 43, 61].

One example of such neighborhood-focused research in the field of HCI is our EatWell system [26], which allowed users to record and share audio stories about how they were trying to eat healthfully in their neighborhood. This system helped facilitate a sense of empowerment among users as they were encouraged to see others in their community succeeding at eating well, and caring enough about the health of the community to share their strategies to help others. The work presented in this paper builds upon this prior work, specifically examining the personal implications of sharing behaviors with others in one’s community (through photographic and textual documentation), and the performative nature of this sharing.

2.2 Performance in digital behavior sharing

Sharing information about one’s health-related activities is reflective of a larger trend in which people are sharing personal information in a variety of online contexts (e.g., [55]). Researchers have characterized this sharing in terms of its performative nature. In a broad sense, Goffman [25] describes a performance as the activities that people engage in, in the presence of others. These activities are done with the intention of influencing the observers to perceive the projected self in a particular way. Observers need not be physically present—researchers have described the invisible and imagined audiences that people perform for online [71].

In social networking systems, users construct a desired self-image as they make their activities and interests known to others [16]. In these environments, performance of self can take the form of composing text (e.g., in blogs and status updates) in a particular way, staging photographs, and selectively choosing which photographs to show others [5, 51, 56, 71]. For example, prior work has examined how Facebook users post-curated collections of photographs as a way to convey a certain identity to others [51]. Posting very specific textual and photographic depictions of one’s activity are performative because they are not simply un-orchestrated depictions of behavior. Instead, they are structured to emphasize certain aspects of one’s identity and deemphasize or hide others [51].

This is not a new phenomenon; in 1980, Boerdam and Martinius [5] discussed how, when taking photographs, people make adjustments for the fact that an audience will eventually view a photograph. As such, attempts to control how one is viewed through photographic and textual representations of the self are common endeavors; social networking systems and other social ICTs simply facilitate new ways of constructing and projecting one’s identity. In this paper, we examine how people engage in performance not through photographs of themselves, but rather through photographs and texts of what are they doing to try to eat healthfully. While the photographer is not always the documented subject in this content, he or she can be understood as the implied subject as the photographs and texts represent a perspective or idea that the content contributor wants to convey [66].

Various researchers have studied the sociology of disclosure and identity management in technologies that facilitate health monitoring, such as the systems described earlier in this paper. Much of this work echoes more general studies of information sharing online. For example, researchers have found that people will often selectively post-behavioral information in online environments (e.g., their physical activity levels) that projects a positive image (e.g., that they are in control of their weight management) [4, 56]. Other work has shown that people are more restrictive in the information they share on general-purpose tools like Facebook, preferring to share detailed information (such as how they were struggling with their health goals) on dedicated online health communities [29, 56]. The latter platform allows users to share with people who are more likely to be dealing with the same challenges and in an environment that allows for more anonymity than a platform like Facebook.

Our research extends this prior work on the performative nature of sharing information about one’s health behaviors in computer-mediated contexts. We examine how participants constructed the representations of their behaviors in CM, particularly how this was done with an eye toward how the content would be received by their audience. In addition, we examine how these efforts had a reflexive impact on content contributors’ assessment of their eating behaviors and their subsequent consumption of food.

3 Community Mosaic

In [60], the design of CM is discussed in detail. Here, we focus on how we designed the tool to help users care for the well-being of those in their community, as opposed to incorporating more individualistic features. CM is a system that helps residents share the steps they are taking to eat healthfully, with the goal of inspiring others to do the same. The idea is that if users share how they are able to successfully eat healthfully, others might learn from this experiential knowledge. CM users take photographs of foods they are eating, preparing, and purchasing, food establishments, or anything else they feel is helping them to pursue a nutritious diet. They can also write text messages describing these experiences.

Users then send these messages to the CM phone number. At this point, our SMS/MMS gateway processes these messages and the photographs and text are displayed on the CM visualization, which runs on a large touch screen monitor (see Fig. 1). The main screen on the CM visualization shows a series of buildings; windows in the buildings illuminate as new messages arrive (Fig. 2). Users can press any building window to expand it and reveal the details of the message. In this detail screen, the message’s photograph and text are displayed (Fig. 3).
Fig. 1

Community Mosaic display

Fig. 2

Community Mosaic visualization, main screen

Fig. 3

Community Mosaic visualization, detail screen

Messages are shown anonymously but are grouped by users; this means that in the detail screen for a message, viewers can scroll through all other messages shared by this anonymous user. This feature helps viewers get a sense of content contributors’ history of experience. Finally, in the detail screen for each message, there is a community response panel that allows viewers to press buttons to share their reactions to the message. Users can press a button to say that (1) they are inspired to try this message, (2) they want to learn more about the healthy eating strategy shown in the message, and (3) they hope others in the community will try this strategy. This panel was designed to help convey the collective sentiment regarding each message. As such, we included the “I hope others will try this” button to help users think not only about themselves, but the wellbeing of the community. Furthermore, it is framed as the “community response” rather than a series of individual reactions.

Our system design incorporated other collectivistic features as well. For example, the CM visualization main screen shows the number of times that users were inspired to try the messages. We deliberately crafted the language to read “We were inspired to try ideas shared X times.” By framing the tally in terms of “we,” our goal was to emphasize the collectivistic focus of the system. Similarly, the system prompt in the CM visualization main screen asks, “How are we trying to eat healthfully today?” Again the use of “we” instead of “you” highlights CM’s collectivistic focus. The prompt goes onto encourage users to “inspire others,” which further emphasizes that users are asked to share messages for the benefit of others.

CM does not provide explicit individualistic incentives for users to document their behaviors. For example, users cannot see visualizations of their eating history on their phones. The system does contain some implicitly individualistic motivators. In the CM visualization main screen, messages that have been interacted with more are shown with a higher opacity (i.e., those messages that have been viewed more and for which the community response panel buttons have been pressed more). The purpose of this was to help viewers quickly scan the main screen and understand which messages the community finds most compelling (in terms of provoking views and community response panel button presses). As such, participants may have been motivated to send messages that would be well received so that they would stand out among the other messages.

In addition, recall that the detail screen for each message has a community response panel that allows viewers to press one of three buttons. This panel also displays the number of times users have pressed each button, indicating that they were inspired to try the strategy represented in the photograph or text, wanted to learn more about this strategy, or hoped others will try it. As such, there is a differentiation of messages—users can see that some messages have been more compelling than others. This feature may have been motivating to competitive (and individualistic) users who hoped that their messages would be viewed more.

In summary, these implicitly individualistic elements were likely motivating to users; indeed, prior research has shown the motivating quality of competitive elements in health behavior monitoring systems [11, 42, 70]. However, as we will discuss later in this paper, we found that the collectivistic system motivators had important personal benefits for message creators.

Community Mosaic is designed to be used by individuals within a constrained geographic area to facilitate the sharing of information that is locally relevant and meaningful. In our prior work, we have shown the benefit of such deeply local systems, which bring together individuals within a shared geographic locale to pursue healthy behaviors [26, 59, 60]. These systems have the potential to facilitate a sense of empowerment, for example, as individuals see that others like them are able to overcome local barriers to healthy eating (e.g., the prevalence of fast food restaurants) and that residents care enough about the health of the community to take the time to share content that benefits others [26, 59].

4 Method

We conducted a study with 43 participants to evaluate user engagement with CM. We installed the CM display at a YMCA in a low-income Atlanta, GA neighborhood; the display was accessible for 6 weeks. The YMCA is a community-focused organization that provides a variety of health and family services to local neighborhoods, including exercise facilities and childcare. In this paper, we focus on the portion of our evaluation that examined user contributions: the nature of sharing, the reasons why participants contributed, and the impact of sharing on contributors. To gain insight into these topics, we conducted semi-structured interviews with participants during the deployment and focus groups at the conclusion. Twenty-nine people participated in either one or two interviews, and 26 participated in the closing focus groups. We used an iterative, inductive approach to analyze the interview data [68]. All interview audio recordings were transcribed and inductively coded to label phenomena in the data. These codes were then iteratively clustered to arrive at higher-level themes.

We complemented our qualitative analysis with pre- and post-deployment surveys in which participants answered demographic questions as well as questions about their reactions to and experiences using CM. In addition, we conducted a deductive content analysis of the CM messages [38], examining to what extent messages were personal, that is, clearly showing how the user was trying to eat healthfully themselves (as opposed to making generic recommendations that others should try). Messages were coded as personal if they clearly showed that the message creator had tried or is currently implementing the healthy eating strategy, not just making a generic suggestion.

Most participants were female (83 %). They ranged in age; most were 18–47 (70 %), with the remainder 48–57 (22.5 %) and 58–67 (7.5 %). As participants attended the same YMCA, some likely knew one another; however, CM messages were displayed anonymously. The Transtheoretical Model of Behavior Change is a widely used health behavior theory that characterizes the stages that people move through as they go from not thinking about engaging in healthier behaviors, to preparing to make changes, to maintaining behavior change [63]. At the start of the study, participants completed surveys that assessed what stage of change they were in [39]. Most participants (56 %) were in the preparation stage (intended to eat less fat in the near future or had recently started to), many (26 %) were in the action stage (for the past 1–5 months they had been eating a low-fat diet), and some (13 %) were in the maintenance stage (had been eating a low-fat diet for 6 months or more). Only two participants were either not thinking about making changes or just beginning to contemplate change. (We did not measure stage of change at the conclusion of the study as we assumed the relatively brief period of use would not be enough time to cause full shifts between stages. Instead, we were interested in the ways in which CM was used, an initial indication of its impact on attitudes and behaviors.) Participants received a $30 gift card for basic participation (agreeing to use CM and completing the surveys), and an additional $30 gift card for participating in a follow-up interview and focus group.

5 Results

In designing CM, our goal was that as users shared messages, the individuals viewing them would be inspired to try out those ideas themselves. Surprisingly, we found that the process of sending messages encouraged message creators to engage in healthy eating practices. Specifically, as participants engaged in the production of messages, they became more aware and analytical of their eating habits. Furthermore, as participants modeled positive behaviors for others (through eating healthfully and documenting these behaviors in their messages), they were themselves motivated to live up to the positive example that they were trying to set.

Our results highlight how the creation of messages in CM can be seen as a kind of performance. In the sections that follow, we begin with an overview of the content shared and participants’ motivations for sharing. We then discuss the performative nature of message production. Next, we describe the personal implications of creating messages for the collective, including increased self-awareness and self-analysis, and users’ desire to adhere to the positive behaviors they were modeling for others. Our focus on the impact of reflection-through-performance (i.e., thinking back on one’s behaviors as one carefully depicts them for an audience) on self-understanding and behavioral intention is critical for understanding how systems like CM may be useful for helping people move toward healthier habits.

5.1 CM messages: What and why?

In total, 278 CM messages were shared during the study. As participants sent messages, they showed or described a variety of foods that they eat to try to be healthy, such as sushi, smoothies, kale salad, water, baked sweet potatoes, couscous, turkey bacon, grilled salmon, a tuna sandwich, collard greens, trail mix, and soy ice cream (see Fig. 4, Table 1). Occasionally they shared messages about other topics (e.g., exercising) or sent a photograph of a food establishment (e.g., Subway restaurant). Most users shared their own behaviors as opposed to impersonal advice, giving viewers a real example of how to eat healthy: In our content analysis, only 20 % of messages were classified as “impersonal,” meaning that they contained no obvious indication that the message sender had tried the healthy eating strategy shared.
Fig. 4

Photographs shared during our CM deployment

Table 1

Text messages shared during our CM deployment (reproduced verbatim)

Lunch today consisted of strawberry mango soup, a scoop of chicken salad on top of mixed greens sprinkled vinegarette dressing and half of a croissant

This strawberry smoothy is a nutritious way to start the day! It can also be a wonderful meal replacement with added protein powder

Chili’s at camp creek has a new soup with chicken and avocados. No diary in the soup. Really good! Will power

After a half hour of cardio box and an hour of zumba this evening at the “Y” I’m still not hungry so I snacked on the tastiest tangerine I’ve ever had that I found at this little side street market in Decatur, and some blueberries

Just had a footlong veggie sub in wheat bread from Subway! It was great!

Sunday morning brunch. Scrambled brown eggs with bell pepper, spinach and onions. Hot tea: blend of green and black teas, guava and strawberry. Fruit: papaya, pineapple and orange (DeKalb Farmer’s Market) … and a good book to nourish my mind and spirit

Desire a summer treat? Then try soy icecream as opposed to dairy icecream. The Purely Decadent brand is one of my top picks!

Pearl soymilk makes a green tea soymilk. I picked it up on sale at Kroger. Green tea is loaded with antioxidants. Give it a try!

Participants sent messages before, during, and after eating the food they discussed in the message. Sometimes they shared foods that were examples of times when they chose a healthier alternative to what they could have otherwise been eating. For example, P8 discussed a message she sent about eating out at lunchtime:

I had a hard decision… Lunchtime is always kind of tricky and [in my message] I was just saying I was choosing what type of bread or what I was going to have. And I think I mentioned I was going to have a tuna on wheat, which was pretty much as opposed to having anything extra like condiments, like mayo and stuff. Just trying to keep it light on that particular day.

In a prior report [60], we discussed how participants viewed CM as a useful tool for counteracting the local draws to unhealthy foods. Participants described the prevalence of unhealthy foods in their neighborhoods, a well-documented characteristic of low-income communities [20, 27, 53]. In the face of such challenges, they discussed how the consistent and public presence of CM helped them draw residents toward healthier options. For example, P12 described how CM gave residents a new way of encouraging healthy eating:

It gives another avenue to use that we didn’t have to us previously… We only had word of mouth and after awhile that kind of dies down… [CM is a] new, you know, way to get information across for others… The more options we have, the better we tend to do in our quest for eating healthier.

This quote exemplifies a sentiment expressed by several participants—that by sharing their experiences in CM, they were doing their part in encouraging and supporting others in eating healthfully. As P12 describes, CM was not conceived of as a typical health monitoring tool, in which much of the motivation for documenting one’s behaviors is improving one’s own health. Instead, CM users were excited about documenting their behaviors such that they could help others. For example, some participants described how they specifically tried to share the most creative ways that they were trying to eat healthfully (as opposed to exhaustive logs of all their meals) [60]. This approach was taken as participants tried to determine how they could best expand residents’ understanding of potential options for eating healthfully. For example, P4 described how she shared messages to showcase vegetarian meal options:

I grew up as a vegetarian and now I’m trying incorporate the same thing with my kids, my kids are vegetarians as well… I figure, you know most people in the South, they eat a lot of soul food, they eat a lot of fried food. So, I wanted to display something that was a little bit different from what they were accustom to.

These examples highlight the orientation of our users, that they documented their eating behaviors with a primary focus on, and a keen eye toward how sharing this content would help fellow residents. A post-deployment survey showed that all participants who had viewed messages in the CM visualization felt that they were at least somewhat useful. And in interviews, participants described learning new or useful healthy eating ideas as they viewed messages in CM. They discussed having their interest peaked by discussions of various fruits and vegetables, organic products, and new places to eat. Furthermore, participants discussed learning new ways to prepare foods, seeing meals that they would have never thought of making and even being exposed to dishes that they had never heard of. For example, P13 said,

I like the community approach, because everybody has their own ideas and I saw some things on there I had never even heard of… Like the stuffed grape leaves….I didn’t know what that was when I looked at it, so, you know you never know what you can gain from other people.

P3 also discussed having her thinking expanded in terms of what she can eat to be healthy:

Some of them fruits that they had were different than what I normally would have tried… And then some of the Caribbean dishes, tropical dishes, I like different types of foods, and it gave me an idea of oh yeah I can eat that. That’s not [too full of] fat and calories or a lot of grease [isn’t] involved because it’s grilled or it’s baked or smoked or whatever.

Thus, our findings suggest that content contributors’ attempts to expand viewers’ understanding of how they can eat healthfully were successful in helping many users learn practical ideas that they could try themselves.

5.2 Production and performance

As participants crafted their messages, we found that while some took photographs without setting up the shot in a particular way (28 %; n = 8) or wrote text messages as a stream of consciousness (21 %; n = 6), most participants (62 %; n = 18) engaged in some level of orchestration by staging photographs or investing care and consideration in the development of their text messages. For example, multiple people staged the foods they were eating before taking a shot. This involved organizing foods and containers in a way that the message sender was satisfied with. For example, P16 said:

I’d first take like a couple pictures and I would see which one’s best. Like I would usually use my oven, it’s got like a little white, white backdrop… So, I was like okay, I will sit it there and I tried to zoom in as far as I [could] to the actual item.

P4 specifically tried to arrange the shot in a way that would be persuasive:

I set [the photos] up how I think it would [be most] persuasive to someone that’s trying to become more healthy and [trying to eat] healthy things.

Some participants were primarily concerned with viewers being able to clearly see what was in the photo and the point that they were trying to get across, such that the message was actually useful. For example, P5 was trying to be “like a little movie producer… I wanted to, you know, be clear and simple to understand.” P9 had this to say:

I definitely tried to make sure the lighting was good and I definitely staged it a little bit. I mean it wasn’t as if I wasn’t eating what I took a picture of but, you know, like when I was showing the protein shake or if I had like a roll up on there and I wanted to show the hummus so if people went to the store they could visually have a picture of what they were looking for. So I definitely—whatever I took a picture of… I was eating, but I definitely tried to set it up so that you could actually see what you were looking at.

Other participants were concerned less with clarity and more with the esthetics of the photograph. P23 described trying to make her photograph attractive to viewers:

I did stage the picture so that it could have a good presence when other people viewed it cause I was trying to make sure that [the food was] neat on my plate, make sure that I didn’t eat any of the food before I took the picture or things of that nature.

Participants also described thinking about the text that they were going to send. Sometimes this simply meant reviewing their messages for typos and grammatical issues. Other participants took the time to think about the content they wanted to share in the message before typing something “off the cuff.” For example, P10 said,

It wasn’t something that I just was like, ‘Oh, eat two apples and a banana and you’ll be healthy.’ You know, it was like, I… thought about it because it was gonna be sent to people. They may take it to heart, say ‘Oh, I’m gonna use this.’

P10 was compelled to think seriously about what she was going to share because she knew that others would be viewing her messages. As with the preparation of photographs, some participants were also concerned with the clarity of their messages or making them appealing to the viewer. For example, P38 said,

I want to convey the message. I want to convey it as explicit as I can. So I’m looking from [the viewer’s] point of view, from their understanding too. To say, how can I make them, you know, have them see exactly what I see, what I’m saying through the words that I’m using. So I try to really be articulate.

P3 was concerned with drawing the viewers in by writing text that was tantalizing:

I tried to think… about a menu, like if I was at a restaurant and I tried to design the words that I used in that sense… You know when you go to a restaurant, you look at a menu and they make the food look so tantalizing and savory and just appealing to the eye, I tried to think of it in that way, to appeal to an audience like if they were at a restaurant. Now I didn’t always do that, but I tried to.

Thus, for many participants, sharing messages in CM was a thoughtful process that involved simple evaluative thinking as they prepared their foods to be photographed and as they wrote their text messages. Through this process, they assessed the way that foods were setup according to a number of criteria, including clarity, usefulness, persuasiveness, esthetics, and appeal. This suggests that participants did not share their experiences without any forethought as to their benefit for, and impact on others. To the contrary, the process of sending a message caused many to take a closer look at the foods they were sharing. In this way, CM helped people take an extra beat—even if only for a second—to reflect on the foods they were eating. As we will discuss later, this was one of the several ways that CM led users to become reflective about their eating practices.

Our findings stand in contrast to what one might expect people to do when compiling a personal food log. With a personal log, users may be less concerned with orchestrating the message because the log may be seen as a way to jog their memory of what they have eaten, as opposed to an attempt to persuade or compel others. The latter requires more thought and deliberation, which in turn requires content creators to spend more time evaluating the foods they are eating and what they want to say about those foods.

We offer that this process of constructing messages can be seen as a form of performance. Participants did not just directly report how they were trying to eat healthfully. Instead, they took care to construct particular representations of their behaviors, knowing they would be consumed by an invisible audience. This finding is reinforced by prior work documenting how photographers anticipate their future audience as they take photographs [5], and research showing the careful ways in which people manage the self-reflective photographs and texts that they share in computer-mediated contexts such as Facebook [51, 56, 71].

Interestingly, while previous work has mainly discussed performativity in online environments as a means of self-presentation [4, 56, 71], our participants spoke about their performance with a more outward focus: helping users understand their message, and persuading users to try new things by making foods look appealing. This outward focus is likely a result of the system prompts that encourage users to focus on how they can help others. Another explanation is that messages in CM were displayed anonymously, potentially reducing the desire to construct a particular identity. In either case, a direction for future work would be to further examine why users were less vocal about how messages reflected their own identity than in prior studies of online environments.

5.3 Behavioral awareness

As participants examined their messages to make sure that they were clear, appealing, and useful, they spent a bit of extra time reflecting on their eating habits. Indeed, the majority of interview participants (69 %; n = 20) said that the process of having to document their habits meant that they became more aware of their eating practices. As P34 mentioned, she started to see, “what I really eat.” P7 described how sharing messages made her think about what she was eating:

I can say honestly [using CM] has pushed me to be more conscious as to what we are eating… thinking about what I am going to eat helped determine what I would send.

This quote highlights how being a contributor in CM necessitated that users spend even a small amount of time reflecting upon what they were eating. This reflection meant that their food choices were on their mind more than they were previously, even if only for a brief moment. This process of consciousness raising is similar to the growing self-awareness that individuals gain when keeping personal diet and exercise logs [12]. However, unlike existing monitoring systems, CM persuades users to document their health-related behaviors using a motivator that does not primarily focus on the benefit for the documenter.

Indeed, as has been reported previously [60], many participants were motivated to record their behaviors to inspire others to eat better and to show them that it is possible to do so. And yet, CM users also gained a personal benefit from sharing, specifically, a greater awareness of their eating habits. And, as previous research has shown, increasing awareness of one’s diet-related behaviors is an important step toward improving these habits [72]. Thus, an important implication for future work is that there may be value in designing systems that incentivize users to document their behaviors by using motivators beyond those that are focused solely on the individual. Such approaches may be particularly helpful in encouraging behavioral reflection by those who are not intrinsically motivated to do so.

5.4 Self-analysis

As participants decided what to send, some also became more self-analytical as they assessed the healthiness of their behaviors. For example, P32 said that as she was preparing to share a message, she examined the foods she was eating and asked herself,

What is the sodium content, you know, how many calories am I eating?

Self-analysis occurred, in part, because participants had to evaluate their eating habits to see whether they were healthy enough to share. For example, P11 said,

A lot of times I’m like, ‘Oh, I probably couldn’t send this today,’ or you know, ‘Could I send this today into the, you know, into the Community Mosaic?’ So yeah, it makes you think about what you’re eating.

Similarly, P16 realized that some of the foods she was eating were not very healthy:

I was going [to] take a picture and then I was like ‘No, wait a second… That is maybe not very healthy… maybe I should not take a picture of that… Maybe, if I don’t want to send it, I probably shouldn’t be sitting eating it or buying it in the first place.’

As these quotes illustrate, the process of assessing their eating habits meant that many participants realized how unhealthily they were eating. Even participants who generally consider themselves to be healthy benefited from this process of self-analysis. P28 said that while she is normally health-conscious, having to decide whether or not her meal was healthy enough to send made her take a second look at her habits:

It made me realize, like if I was eating something that was not as healthy as I thought it should be, I thought about it—I thought about it a little bit more… Even though I’m normally very conscious of what I’m eating, I would think about it and say, ‘Hm, I don’t think this is one that I would want to send.’

Similarly, preparing to send messages gave P9 the chance to look more closely at the foods she was eating. In doing so, she realized she was not doing as well as she thought:

It made me think of how badly I was eating… Like, typically I think of myself as a healthy eater but it made me realize that I was, like, on the party train and had left my good eating habits behind me… Because I realized I wasn’t sending as many messages as I wanted to and I realized that it was… at first I was like, I don’t have time to do this, you know, to do the text… And then two, I realized the other reason was I wasn’t eating things that I really wanted to take pictures of.

Assessing meals to decide whether they were healthy enough to recommend to others led users to evaluate what they were eating in ways that they may not have done previously. Again, given the literature stating that reflection on one’s eating habits can help people engage in positive behavior change [72], this is an exciting finding.

5.5 Impact on behaviors

In addition to helping participants reflect upon and analyze their behaviors, sharing messages compelled many participants (n = 11) to eat more healthfully. Some wanted to eat better because they knew that others would be looking at their habits. For example, P4 said that using CM,

…has made me more conscientious about what I’m eating because, you know, I felt like there’s somebody watching me or I’d have an audience and I can’t talk about nutrition and health and low fat if I’m not doing it. I can’t demonstrate it if I’m not doing it.

This finding echoes prior work that has shown the accountability that social health monitoring tools afford [4]. Other participants were motivated to live up to the example they were putting forth in the messages they shared. As P10 discussed, by sending messages, message creators were reminded of how they should be eating. These participants saw dissonance in sharing CM messages that encourage others to eat well and yet not “practicing what they were preaching.” For example, P6 said,

I could not send a message with a clear conscience you know, if I had eaten a box of chocolates that day.

P11 further discussed this feeling of dissonance,

You’re thinking about, about things that you can send in your everyday life so you try—for me, I try—tomake better decisions. And it works… I think it helped because, you know, you’re thinking about it and thinking about what you can send… You felt kind of weird to send out the messages and then you’re not following it yourself.

P13 described how sending messages helped encourage her to get back into her healthy eating routine:

It would tell me that the things I was sending pictures of—I need to eat more of that. ‘Cause like I said sometimes I fall off the boat, I might have a food attack or something, or a junk attack and, you know, the pictures would show me that I need to eat more of what I was focusing on [in the messages I sent]… It showed me that, you know, the pictures, the messages I was sending, I need to be eating more of that. Not just, you know, for the study… but I need to involve them more in my daily diet and my daily eating habits.

P33 also discussed how sharing messages led him to make healthier eating decisions:

It actually, you know, put me [in the mindset of] ‘Well, okay, I need to make a better choice than, you know, stopping somewhere and getting some McDonald’s or something.’ Or, if I did, maybe I need to get a salad or something different. So it, it actually kind of helped me in the sense I would make more healthier choices in eating. I actually think I might have lost a couple of pounds doing it that way too… Just a little bit, you know, nothing major… But I think it is something that I’ll continue to do even after the fact like, you know, really pay attention to what I’m eating.

In summary, sharing messages had both collective and personal behavioral implications for the message senders. With a focus on the collective, participants were inspired to share their positive behaviors to encourage others that it is possible to eat healthfully and to inspire them to want to eat well [60]. Yet, while the motivation to share was a collective one for many participants, many reaped a very personal behavioral benefit, as they were encouraged to continue eating nutritiously themselves.

5.6 Summary

Previous work has shown that keeping a fitness log helps people reach their wellness goals because they gain an increased awareness of their behaviors and begin to more critically assess them [72]. The benefit of personal logging has motivated a number of systems that support users in both manually and automatically keeping digital records of their diet and exercise [4, 10, 12, 32, 48, 69, 70].

CM is interesting and distinct from this previous work, however, because it asks users to document their behaviors not for their own sake, but for the benefit of others in their community. And yet we found that users still received similar benefits to individually focused monitoring applications as well as social applications that incentivize documentation through competitive mechanisms and the potential of receiving social support [10, 12, 42, 48, 70]. In contrast, CM motivates users to record their behaviors using a cooperative and externally focused incentive, but still affords the personal benefits of self-awareness, self-analysis, and early signs of healthier decision-making. These findings suggest that future work should further explore such external mechanisms for compelling users to monitor their behaviors and the extent to which individuals who were not otherwise interested in monitoring their behaviors are persuaded to do so.

6 Discussion

By encouraging participants to share their eating experiences for the benefit of the community, CM yielded positive benefits for participants. We suggest that future work further examines the value and limitations of such collectivistic motivators. Previous work on health monitoring systems has often examined how people are influenced by seeing others’ behaviors, and how people modify their behaviors based upon social pressure and support [11, 70]. To complement this prior work, our research examined the more fundamental question of how information about one’s behaviors is documented and shared—and the performative elements therein. We argue that just as we perform our identities, the act of documenting and sharing health behaviors is performative and that examining it as such yields questions and opportunities for design.

Previous work has often described the ways in which social networking systems such as Twitter are platforms for digital performances [58]. It follows, then, that we must consider how social systems for health are similarly platforms for performance and what this means for design and for wellness. We use our findings to highlight open areas for future research, adding to the recent body of work in this space [4, 56]. We specifically focus on the intersection of performativity and collectivistic motivators in health monitoring tools by considering two topics: perceived audience and automation in monitoring tools. In our discussion, we examine the implications of each for reflection-through-performance, that is, the way in which crafting the representation of one’s behaviors in social monitoring tools can help facilitate reflection on these behaviors. We conclude with broader reflections on research that attempts to understand how health monitoring tools are used by individuals with varying perspectives on the individual and the collective.

6.1 Perceived audience

In computer-mediated environments such as social networking sites, people often balance multiple audiences, tailoring and managing their presentation of self for each [50]. As in real-life interactions, people shift how they present themselves online, depending on the audience that they are engaged with. Our participants discussed how they tried to inspire others by sharing how they were trying to eat healthfully. Furthermore, they spoke about how CM helped them to counteract the draws to unhealthy eating by local food vendors, by providing community residents with alternative options [60]. As such, it is reasonable to assume that at the very least, participants perceived their audience to be local residents who could benefit from new ideas for how to eat healthfully. While this perspective might at first glance seem condescending, our findings also show that as they tried to help others, participants frequently realized just how unhealthily they were eating and how they could themselves benefit from behavior change.

A more detailed analysis of how our participants viewed their audience would help to further illuminate the ways in which they framed and shared their eating experiences. For example, if participants perceived the CM audience to be individuals who are resistant to change, they might take more time to craft their messages and stage photographs such that they are persuasive, or they might be less motivated to share if they feel they are unlikely to impact their audience. We previously reported that CM users rarely shared ethnic dishes: only 8 % contained depictions or descriptions of traditional African American foods, though in our formative work participants were most excited about learning healthier versions of such foods [59]. If participants imagined CM users to be culturally dissimilar, that might explain decisions to share or not share ethnic dishes.

In sum, in health monitoring tools that ask users to document their behaviors for the benefit of others, the way that users view their audience may change the message production process, which in turn may impact the effect of creating a message on the sender. For example, if a user’s perception of her audience leads her to spend less effort composing messages (e.g., because she doubts her ability to influence others), this means she will spend less time reflecting on her behaviors, which may mean less time considering the healthiness of her behaviors or opportunities for improving her habits. Future work should examine how users’ perceptions of their audience impact behavior documentation and reflection in social monitoring tools.

6.2 Automation

Prior work has yielded tools that help users self-report their behaviors (e.g., what meals they have eaten) and tools in which the system automatically tracks behaviors [2, 3, 4, 7, 33, 34]. Automating behavior monitoring is beneficial because it helps ensure more consistent tracking (whereas users may forget or not feel like self-reporting their behaviors), enables more objective measures of behaviors (e.g., physical activity levels), and can better facilitate visualizations of large amounts of data through the provision of a more exhaustive data set.

At the same time, automated tracking can remove the reflection-through-performance that is enabled in systems such as CM. Our participants became more aware and analytical of their behaviors through the process of deciding how to frame and share their behaviors. It was through the process of considering what meals to take a photograph of, how to stage that photograph in a way that is persuasive, and through considering how their audience may react to and benefit from their message that participants engaged in beneficial self-reflective thinking. As such, future work on automated behavior tracking tools should consider how such tools can be designed to still afford such opportunities for reflection-through-performance.

For example, there are many commercial2 and research tools that help users automatically track and share their physical activity levels (e.g., [3, 12, 34]). Future work could examine how such systems might allow users to more thoughtfully customize how that data are shared with other system users. Beyond obvious features like allowing users to manage the group of people that such data are shared with, these systems could support richer performative control, such as allowing users to add evocative photographs to convey how they felt after completing an activity or customizing how tracked information is displayed to different audiences.

For example, in a system that encourages people to share their physical activity experiences to inspire others to be more active, users might want to frame their activity differently for users who are more or less motivated to increase their activity levels. The system could (privately) assess user stages of change [63] and give content contributors the option of appending customized messages to their shared activity data. For individuals who are contemplative (just starting to think about engaging in physical activity), content contributors might emphasize how they overcame barriers to engage in the depicted activity, and for individuals who are in action (have been engaging in physical activity for a few months), contributors might instead emphasize how they are maintaining their activity levels over time. System scaffolds could help users identify the kinds of message framings that may be helpful for individuals at various stages of behavior change. As users determine what about their physical activity levels they can share with these different audiences, they would engage in increased reflective thinking, which may be useful in helping them to assess to what extent they are achieving their desired levels of physical activity, barriers they need to overcome, and ideas for how they can overcome such barriers.

While our work suggests that performativity in behavior sharing can lead to important benefits for users, longitudinal experimental evaluations are necessary to confirm these findings. Future work should also systematically evaluate whether the addition of performative elements in automated behavior tracking tools, such as those suggested here, increase awareness and ultimately facilitate behavior change for content contributors.

6.3 Collectivism and individualism: understanding use

As reported previously [59], we used a validated measure of collectivism and individualism to assess participants’ values. While most expressed agreement with collectivist values, most also expressed agreement with individualistic values. One explanation for these dual values is that researchers have begun to argue that collectivism and individualism are not polar opposites, but rather values that can be held to varying degrees depending on the situation [35]. For example, someone may feel that interdependence is important in the context of their family but not when considering their coworkers. Another potential explanation is that because the USA is often characterized as an individualistic nation overall, subcultures within the USA are likely influenced by individualistic values such as personal achievement. Whatever the reason, our findings highlight the importance of carefully understanding the nuance of users’ orientations toward the individual and the collective, and what this might mean for systems that support reflection and behavior sharing with others.

For example, the presence of these dual values of individualism and collectivism may be one reason why our participants embraced the idea of sharing their behaviors to benefit others—they may have valued contributing to the well-being of the group (a collectivistic perspective) and also felt a personal sense of pride that their unique ideas were being showcased (an individualistic value). If we were to involve participants who were more purely collectivistic or individualistic, this may have affected the use, impact of, and reaction to CM. For example, more purely individualistic users may not have been as motivated or interested in modeling positive behaviors for others. These users may also have been more inwardly focused in their performativity, for example, less focused on helping others through their performance and more interested in the management of how their identity is presented. People who are more purely collectivistic might have focused more on how their contributions are helping others. Pure collectivists may also have been less reflective on how they could change their own behaviors, or if they were self-reflective, they may be more motivated by considering how their well-being contributes to the wellbeing of the collective, as opposed to gaining a personal satisfaction from improving their health.

As discussed in this paper and in prior reports on this research [26, 59, 60], collectivistic systems have the potential to yield positive impact, facilitating feelings of empowerment, a shared pursuit of community goals, and even greater self-reflective thinking. An important question is to what extent collectivistic systems face any potential unintended consequences or challenges. One concern is that no population is homogenous, so there will certainly be people with varying levels of collectivistic values using the system together. More work should systematically examine the impact of this mixed participation. For example, we found previously that individuals whose dominant value is individualism tended to contribute less to CM than those for whom collectivism was the dominant value [59]. Another concern is that for individuals who are not naturally inclined to be reflective, a collectivistic system like CM might not produce the kinds of self-awareness and analysis that we saw many of our participants engage in. Such users may require further scaffolding to achieve these kinds of personal benefits.

7 Conclusion

Researchers and companies are becoming increasingly interested in how technology can help people monitor their health-related behaviors. While some of this research has focused on personal tools, increasingly, this work is incorporating social elements. Prior work has examined systems that facilitate social support and influence in a variety of contexts; typically these systems have included individualistic motivators for behavior documentation (e.g., by monitoring their behaviors, users receive system-enabled personal benefits, such as rich visualizations of their data and social support from friends). Our work explored the design of a health monitoring tool with a collectivistic motivator: CM prompted users to share their behaviors primarily and specifically for the benefit of others.

In our evaluation of CM, we found that users often engaged in a kind of performance as they staged photographs and carefully crafted text messages with their imagined audience in mind. By engaging in this performance, they spent a bit of extra time reflecting on their behaviors, and this reflection led to self-awareness, self-analysis, and even small changes to the content contributors’ behaviors. We encourage future work that examines how collectivistic motivators in health monitoring tools can uniquely encourage people to document their behaviors and receive reflexive benefits such as increased awareness of their behaviors.


  1. 1.

    In this paper, I use the term reflection to refer to the process of considering one’s present, past, or planned health-related behaviors, attitudes, and experiences. This cognitive processing may take the form of reflection-in-action (during the unfolding of an event), or reflection-on-action (after the completion of an event) [62, 67].

  2. 2.



The author thanks the study participants, the YMCA of Atlanta and Humana for their support for this research. Beki Grinter, Vasudhara Kantroo, Hee Rin Lee, Miguel Osornio, and Mansi Sharma provided invaluable help with this work.


  1. 1.
    Agrawal T, Hoffman JA, Ahl M, Bhaumik U, Healey C, Carter S, Dickerson D, Nethersole S, Griffin D, Castaneda-Sceppa C (2012) Collaborating for impact: a multilevel early childhood obesity prevention initiative. Family Community Health 35:192–202CrossRefGoogle Scholar
  2. 2.
    Amft O, Tröster G (2008) Recognition of dietary activity events using on-body sensors. Artif Intell Med 42(2):121–136CrossRefGoogle Scholar
  3. 3.
    Anderson I, Maitland J, Sherwood S, Barkhuus L, Chalmers M, Hall M, Brown B, Muller H (2007) Shakra: tracking and sharing daily activity levels with unaugmented mobile phones. Mob Netw Appl 12(2–3):185–199CrossRefGoogle Scholar
  4. 4.
    Baumer EP, Katz SJ, Freeman JE, Adams P, Gonzales AL, Pollak J, Retelny D, Niederdeppe J, Olson CM, Gay GK (2012) Prescriptive persuasion and open-ended social awareness: expanding the design space of mobile health. In: CSCW’12. ACM, pp 475–484Google Scholar
  5. 5.
    Boerdam J, Martinius WO (1980) Family photographs: a sociological approach. Neth J Sociol 16(2):95–119Google Scholar
  6. 6.
    Braithwaite DO, Waldron VR, Finn J (1999) Communication of social support in computer-mediated groups for people with disabilities. Health Commun 11(2):123–151CrossRefGoogle Scholar
  7. 7.
    Chi P, Chen J, Chu H, Chen B-Y (2007) Enabling nutrition-aware cooking in a smart kitchen. In: CHI ‘07 extended abstracts. ACM Press, pp 2333–2338Google Scholar
  8. 8.
    Chiu M-C, Chang S-P, Chang Y-C, Chu H-H, Chen CC-H, Hsiao F-H, Ko J-C (2009) Playful bottle: a mobile social persuasion system to motivate healthy water intake. In: Ubicomp’09, pp 185–194Google Scholar
  9. 9.
    Cohen S (2004) Social relationships and health. Am Psychol 59(8):676CrossRefGoogle Scholar
  10. 10.
    Connelly KH, Faber AM, Rogers Y, Siek KA, Toscos T (2006) Mobile applications that empower people to monitor their personal health. E & I Elektrotech Informationstech 123(4):124–128CrossRefGoogle Scholar
  11. 11.
    Consolvo S, Everitt K, Landay JA (2006) Design requirements for technologies that encourage physical activity. In: CHI’06. ACM, pp 457–466Google Scholar
  12. 12.
    Consolvo S, Klasnja P, McDonald DW, Avrahami D, Froehlich J, LeGrand L, Libby R, Mosher K, Landay JA (2008) Flowers or a robot army? Encouraging awareness and activity with personal, mobile displays. In: Ubicomp’08, pp 54–63Google Scholar
  13. 13.
    Consolvo S, McDonald DW, Toscos T, Chen MY, Froehlich J, Harrison B, Klasnja P, LaMarca A, LeGrand L, Libby R, Smith I, Landay JA (2008) Activity sensing in the wild: a field trial of UbiFit garden. In: CHI’08, pp 1797–1806Google Scholar
  14. 14.
    Consolvo S, Roessler P, Shelton BE, LaMarca A, Schilit B, Bly S (2004) Technology for care networks of elders. Pervasive Comput 3(2):22–29CrossRefGoogle Scholar
  15. 15.
    Curmi F, Ferrario MA, Southern J, Whittle J (2013) HeartLink: open broadcast of live biometric data to social networks. In: Proceedings of the 2013 ACM annual conference on human factors in computing systems. ACM, pp 1749–1758Google Scholar
  16. 16.
    DiMicco JM, Millen DR (2007) Identity management: multiple presentations of self in Facebook. In: Proceedings of the 2007 international ACM conference on supporting group work. ACM, pp 383–386Google Scholar
  17. 17.
    Dishman E (2004) Inventing wellness systems for aging in place. Computer 37(5):34–41 Google Scholar
  18. 18.
    Economos CD, Irish-Hauser SA (2007) Community interventions: a brief overview and their application to the obesity epidemic. J Law Med Ethics 35(1):131–137CrossRefGoogle Scholar
  19. 19.
    Feathers JT, Kieffer EC, Palmisano G, Anderson M, Sinco B, Janz N, Heisler M, Spencer M, Guzman R, Thompson J, Wisdom K, James SA (2005) Racial and ethnic approaches to community health (REACH) detroit partnership: improving diabetes-related outcomes among African American and Latino adults. Am J Public Health 95(9):1552–1560CrossRefGoogle Scholar
  20. 20.
    Freedman DS (2011) CDC Health Disparities and Inequalities Report, USAGoogle Scholar
  21. 21.
    Frey LR (2003) Group communication in context: studies in bona fide groups. Lawrence Erlbaum Associates, LondonGoogle Scholar
  22. 22.
    Frost J, Smith BK (2003) Visualizing health: imagery in diabetes education. In: DUX’03. ACM, pp 1–14Google Scholar
  23. 23.
    Frost JH, Massagli MP (2008) Social uses of personal health information within PatientsLikeMe, an online patient community: what can happen when patients have access to one another’s data. J Med Internet Res 10:3CrossRefGoogle Scholar
  24. 24.
    Glanz K, Rimer BK, Viswanath K (eds) (2008) Health behavior and health education. Jossey-Bass, San Francisco, CAGoogle Scholar
  25. 25.
    Goffman E (1959) The presentation of self in everyday life. Anchor, New YorkGoogle Scholar
  26. 26.
    Grimes A, Bednar M, Bolter JD, Grinter RE (2008) EatWell: sharing nutrition-related memories in a low-income community. In: Proceedings of CSCW’08. ACM, pp 87–96Google Scholar
  27. 27.
    Horowitz CR, Colson KA, Hebert PL, Lancaster K (2004) Barriers to buying healthy foods for people with diabetes: evidence of environmental disparities. Am J Public Health 94(9):1549–1554CrossRefGoogle Scholar
  28. 28.
    Johnson GJ, Ambrose PJ (2006) Neo-tribes: the power and potential of online communities in health care. Commun ACM 49(1):107–113CrossRefGoogle Scholar
  29. 29.
    Kamal N, Fels S, McGrenere J, Nance K (2013) Helping me helping you: designing to influence health behaviour through social connections. In: Kotzé P, Marsden G, Lindgaard G, Wesson J, Winckler M (eds) Human–computer interaction—INTERACT 2013. Springer, Berlin, pp 708–725CrossRefGoogle Scholar
  30. 30.
    Kim S, Kientz JA, Patel SN, Abowd GD (2008) Are you sleeping? Sharing portrayed sleeping status within a social network. In: Proceedings of the 2008 ACM conference on Computer supported cooperative work. ACM, pp 619–628Google Scholar
  31. 31.
    Kim S, Schap T, Bosch M, Maciejewski R, Delp EJ, Ebert DS, Boushey CJ (2010) Development of a mobile user interface for image-based dietary assessment. In: Proceedings of the 9th international conference on mobile and ubiquitous multimedia. ACM, Limassol, CyprusGoogle Scholar
  32. 32.
    Kimani S, Berkovsky S, Smith G, Freyne J, Baghaei N, Bhandari D (2010) Activity awareness in family-based healthy living online social networks. In: IUI’2010. ACM, pp 337–340Google Scholar
  33. 33.
    King AC, Ahn DK, Oliveira BM, Atienza AA, Castro CM, Gardner CD (2008) Promoting physical activity through hand-held computer technology. Am J Prev Med 34(2):138–142CrossRefGoogle Scholar
  34. 34.
    Klasnja P, Pratt W (2012) Healthcare in the pocket: mapping the space of mobile-phone health interventions. J Biomed Inform 45(1):184–198CrossRefGoogle Scholar
  35. 35.
    Komarraju M, Cokley KO (2008) Horizontal and vertical dimensions of individualism–collectivism: a comparison of African Americans and European Americans. Cultur Divers Ethnic Minor Psychol 14(4):336–343CrossRefGoogle Scholar
  36. 36.
    Kreuter MW, Haughton LT (2006) Integrating culture into health information for African American women. Am Behav Sci 49(6):794–811CrossRefGoogle Scholar
  37. 37.
    Kreuter MW, McClure SM (2004) The role of culture in health communication. Annu Rev Public Health 25:439–455CrossRefGoogle Scholar
  38. 38.
    Krippendorff K (2004) Content analysis: an introduction to its methodology. Sage, Thousand OaksGoogle Scholar
  39. 39.
    Kristal AR, Glanz K, Curry SJ, Patterson RE (1999) How can stages of change be best used in dietary interventions? J Am Diet Assoc 99(6):679–684CrossRefGoogle Scholar
  40. 40.
    Kummervold PE, Gammon D, Bergvik S, Johnsen JA, Hasvold T, Rosenvinge JH (2002) Social support in a wired world: use of online mental health forums in Norway. Nord J Psychiatry 56(1):59–65CrossRefGoogle Scholar
  41. 41.
    Leimeister JM, Krcmar H (2005) Evaluation of a systematic design for a virtual patient community. J Comput Mediat Commun 10:4CrossRefGoogle Scholar
  42. 42.
    Lin JJ, Mamykina L, Lindtner S, Delajoux G, Strub HB (2006) Fish’n’Steps: encouraging physical activity with an interactive computer game. In: Ubicomp’06, pp 261–278Google Scholar
  43. 43.
    Lisovicz N, Johnson RE, Higginbotham J, Downey JA, Hardy CM, Fouad MN, Hinton AW, Partridge EE (2006) The Deep South network for cancer control. Cancer 107(S8):1971–1979CrossRefGoogle Scholar
  44. 44.
    Maitland J, Chalmers M (2011) Designing for peer involvement in weight management. In: Proceedings of the 2011 annual conference on Human factors in computing systems. ACM, pp 315–324Google Scholar
  45. 45.
    Maitland J, Sherwood S, Barkhuus L, Anderson I, Hall M, Brown B, Chalmers M, Muller H (2006) Increasing the awareness of daily activity levels with pervasive computing. In: Pervasive Health’06. IEEE, pp 1–9Google Scholar
  46. 46.
    Maloney-Krichmar D, Preece J (2005) A multilevel analysis of sociability, usability, and community dynamics in an online health community. ACM Trans Comput Hum Interact 12(2):201–232CrossRefGoogle Scholar
  47. 47.
    Mamykina L, Miller AD, Grevet C, Medynskiy Y, Terry MA, Mynatt ED, Davidson PR (2011) Examining the impact of collaborative tagging on sensemaking in nutrition management. In: Proceedings of the 2011 annual conference on Human factors in computing systems. ACM, Vancouver, BC, Canada, pp 657–666Google Scholar
  48. 48.
    Mamykina L, Mynatt E, Davidson P, Greenblatt D (2008) MAHI: investigation of social scaffolding for reflective thinking in diabetes management. In: CHI’08. ACM, pp 477–486Google Scholar
  49. 49.
    Mankoff J, Kuksenok K, Kiesler S, Rode JA, Waldman K (2011) Competing online viewpoints and models of chronic illness. In: Proceedings of the SIGCHI conference on human factors in computing systems. ACM, pp 589–598Google Scholar
  50. 50.
    Marwick AE (2011) I tweet honestly, I tweet passionately: twitter users, context collapse, and the imagined audience. New Media Soc 13(1):114–133CrossRefGoogle Scholar
  51. 51.
    Mendelson AP, Papacharissi Z (2010) Look at us: collective narcissism in college student Facebook photo galleries. The networked self. In: Papacharissi Z (ed) The networked self: identity, community and culture on social network sites. Routledge, LondonGoogle Scholar
  52. 52.
    Minkler M, Wallerstein N (2008) Community-based participatory research for health: from process to outcomes. Wiley, San FranciscoGoogle Scholar
  53. 53.
    Morland K, Wing S, Diez Roux A, Poole C (2002) Neighborhood characteristics associated with the location of food stores and food service places. Am J Prev Med 22(1):23–29CrossRefGoogle Scholar
  54. 54.
    Mynatt ED, Rowan J, Jacobs A, Craighill S (2001) Digital family portraits: supporting peace of mind for extended family members, pp 333–340Google Scholar
  55. 55.
    Naaman M, Boase J, Lai C-H (2010) Is it really about me? Message content in social awareness streams. In: Proceedings of the 2010 ACM conference on computer supported cooperative work. ACM, pp 189–192Google Scholar
  56. 56.
    Newman MW, Lauterbach D, Munson SA, Resnick P, Morris ME (2011) It’s not that i don’t have problems, I’m just not putting them on Facebook: challenges and opportunities in using online social networks for health. In: CSCW’11. ACM, pp 341–350Google Scholar
  57. 57.
    Noronha J, Hysen E, Zhang H, Gajos KZ (2011) Platemate: crowdsourcing nutritional analysis from food photographs. In: UIST’11. ACM, Santa Barbara, CA, USA pp 1–12Google Scholar
  58. 58.
    Papacharissi Z (2012) Without you, I’m nothing: performances of the Self on Twitter. Int J Commun 6:18Google Scholar
  59. 59.
    Parker AG, Grinter RE (2014) Collectivistic health promotion tools: accounting for the relationship between culture, food and nutrition. Int J Hum Comput Stud 72(2):185–206CrossRefGoogle Scholar
  60. 60.
    Parker AG, Kantroo V, Lee H, Osornio M, Sharma M, Grinter RE (2012) Health promotion as activism: building community capacity to affect social change. In: Proceedings of CHI’12, pp 99–108Google Scholar
  61. 61.
    Plescia M, Groblewski M, Chavis L (2008) A lay health advisor program to promote community capacity and change among change agents. Health Promot Pract 9(4):434–439CrossRefGoogle Scholar
  62. 62.
    Ploderer B, Reitberger W, Oinas-Kukkonen H, van Gemert-Pijnen J (this issue) Editorial: Social interaction and reflection for behaviour change. Pers Ubiquitous ComputGoogle Scholar
  63. 63.
    Prochaska JO, Redding CA, Evers KE (2008) The transtheoretical model and stages of change. In: Glanz K, Rimer BK, Viswanath K (eds) Health behavior and health education. Jossey-Bass, San Francisco, pp 97–121Google Scholar
  64. 64.
    Russell KM, Champion VL, Monahan PO, Millon-Underwood S, Zhao Q, Spacey N, Rush NL, Paskett ED (2010) Randomized trial of a lay health advisor and computer intervention to increase mammography screening in African American women. Cancer Epidemiol Biomark Prev 19(1):201–210CrossRefGoogle Scholar
  65. 65.
    Salem DA, Bogat GA, Reid C (1997) Mutual help goes on-line. J Community Psychol 25(2):189–207CrossRefGoogle Scholar
  66. 66.
    Samsell M (2010) Mediated: the image as a performative interface in the photographic relationship. In: Coohill PT (ed) Art inspiring transmutations of life. Springer, Netherlands, pp 371–382CrossRefGoogle Scholar
  67. 67.
    Schön DA (1983) The reflective practitioner: how professionals think in action. Basic Books, New YorkGoogle Scholar
  68. 68.
    Thomas DR (2006) A general inductive approach for qualitative data analysis. Am J Eval 27(2):237–246CrossRefGoogle Scholar
  69. 69.
    Toscos T, Connelly K, Rogers Y (2012) Best intentions: health monitoring technology and children. In: CHI’12. ACM, Austin, TX, USA, pp 1431–1440Google Scholar
  70. 70.
    Toscos T, Faber A, Connelly K, Upoma AM (2008) Encouraging physical activity in teens. In: Proceedings of pervasive health’08, pp 218–221Google Scholar
  71. 71.
    Westlake E (2008) Friend me if you Facebook: generation Y and performative surveillance. Drama Rev 52(4):21–40CrossRefGoogle Scholar
  72. 72.
    Wing RR, Hill JO (2001) Successful weight loss maintenance. Annu Rev Nutr 21(1):323–341CrossRefGoogle Scholar

Copyright information

© Springer-Verlag London 2014

Authors and Affiliations

  1. 1.College of Computer and Information ScienceNortheastern UniversityBostonUSA
  2. 2.College of Health SciencesNortheastern UniversityBostonUSA

Personalised recommendations