First Monday

Tube therapy: Dealing with mental health problems in social video comment threads by Simon Lindgren and Ragnar Lundstrom

In this paper, we examine forms of communication and types of peer support that may arise in relation to mental health problems in peer-produced forums of social video comment threads. We focus specifically on how young fans of young YouTube celebrities (YouTubers) use this social media space to communicate with other YouTubers as well as with other fans about issues related to mental health. Through a semantic analysis, where topical themes in comment discussions are mapped, we address topics concerning mental health conditions in social video fandom discussions. At an overarching level, we are interested in how peer support is realized in the complex relationships between YouTubers and audience members, as well as among audience members. In our analysis, we focus on both the relationships between the individual and the collective, and between cooperation and competition in discussions. In concluding, we reflect upon what hindrances or possibilities for peer support exist in this relatively new and participatory social media setting.


YouTubers and their audience
YouTube as a communication environment
Communicating mental health problems on YouTube
Semantic analysis of YouTube comments
Individual versus collective
Competition versus cooperation
Conclusion: Exchanging social support in a social media setting




In this paper, we take an interest in the forms of communication and types of peer support that may arise in relation to mental health problems in peer-produced forums of social video comment threads. We focus specifically on how young fans of young YouTube celebrities (YouTubers) use this social media space to communicate with other YouTubers as well as with other fans about issues regarding mental health.

Comment fields relating to videos by YouTubers are a relatively important social space for many young people today. Even moderately popular YouTubers may have subscription and viewing numbers that largely surpass those of conventional outlets such as, for example, teen magazines [1]. Furthermore, a striking number of videos by YouTubers address mental health. While the exact number is hard to pinpoint, a YouTube search query for the keyword ‘vlog’, alongside a set of words referring to relevant diagnoses, yields around 10,000 results per month. This suggests that somewhere around 10 such videos are potentially posted every hour on the platform. But the social environments emerging through YouTubers’ output has a specific character, of which we still do not know very much, and which therefore demands to be studied more closely.

Due to the fact that these discussions are carried out within dedicated groups of followers, which presumably share characteristics with so-called fandoms, they can be assumed to follow certain social rules where particular forms of knowledge and appreciation relate to a fandom-specific cultural capital (Fiske, 1992). The complexity by which communication is carried out partly in relation to a YouTuber, partly about the YouTuber, and partly simply in the form of direct communication among audience members, raises some issues. Horton and Wohl (1956) examined this kind of parasocial interaction — the role played by imaginary relationships between a media figure and an audience member.

Through a semantic analysis, where topical themes in comment discussions are mapped, we address a series of issues related to coping with mental health conditions in social video fandom discussions. At the overarching level, we are interested in how peer support can be realised in complex relationships between YouTubers and audience members, as well as among audience members. In the analysis, we focus on both the relationships between the individual and the collective, and between cooperation and competition in discussions. In concluding, we reflect upon hindrances or possibilities for peer support in this participatory social media setting.



YouTubers and their audience

The online video sharing platform YouTube is a social and cultural force and phenomenon to be reckoned with in today’s society. Not only is it the second most visited Web service globally, coming in after Google [2], but its immense user base and enormous traffic makes it a nodal point in the media landscape of the twenty-first century. It is a huge video archive with more than a billion users watching hundreds of millions of hours of video every day, uploading 300 hours of video every minute [3]. While it started out as a simple video repository, it can now be argued that YouTube is now a platform for a range of creative activities (Gauntlett, 2011). Its culture has given rise to new genres of content, as well as new modes of content production, in the form of video blogs, mashups, user-created tutorials, and so on.

Significant research has focused on social practices in YouTube, its different uses and gratifications, and on the role of the platform for identity, economy, and visual culture (Burgess and Green, 2009b; Lovink and Niederer, 2008; Lovink and Somers Miles, 2011; Snickars and Vonderau, 2009; Strangelove, 2010). Some research has examined “vlogging” (video blogging) apparing early in the platform’s existence (Lange, 2009; Lindgren, 2011; Molyneaux, et al., 2008). Video blogs feature YouTubers sharing aspects of their everyday lives, prominently in the form of clips where they speak directly to the camera, or in the form of so-called “follow me around” videos. Much of that research was interested in how personal connections were made possible through a large-scale, and seemingly depersonalised, medium. In recent years, there has been renewed interest in vloggers, particularly “entrepreneurial vloggers” (Burgess and Green, 2009a), or as they are now more commonly called, YouTubers.

While the word “YouTuber” can be used to refer to anyone who uses the platform — anyone engaging in some form of “YouTubing” — the word is generally used to refer to individuals who have been elevated to some degree of fame through self-creating content on YouTube, or who make a living through content by ad revenues, sometimes alongside sponsorship deals and spinoff projects in other media [4]. Beyond megastars such as PewDiePie (55 million subscribers), elrubiusOMG (24 million subscribers), and Jenna Marbles (17 million subscribers), an increasing number of rather moderately sized channels run by self-made YouTube content creators around the world have been able to build, often from scratch, audiences that are surprisingly voluminous and loyal [5]. A survey conducted by Variety in 2014 showed that the top five (and six out of the top ten) most popular public figures among U.S. teens were YouTubers [6].

While some YouTubers indeed become huge celebrities, they have a larger degree of interaction with audiences than conventional public figures, due to YouTube culture and infrastructure. Indeed, as research has shown, the “attainability”, or communicative availability, of YouTubers plays a large role in why young people are affected by them [7].

YouTubers have in most cases started from the bottom, growing follower by follower. They receive attention and become role models based on their personalities, talents, and creativity. Intimate and personal stories, disclosures, and insights contribute to YouTubers being seen as trustworthy and approachable (Westenberg, 2016). It is important to note however that far from all YouTubers disclose personal and intimate details about their lives and realities. Some of the most popular genres that YouTubers engage within, aside from personality vlogging, are beauty and fashion videos, gameplay videos, and comedy sketches. Thus, there are differences between such thematic categories, as well as among individual YouTubers. However, in general, YouTubers, regardless of genre, have a strong tendency towards playing on a self-referential, direct, and communal sensibility.

YouTube, as a space, encourages audiences to engage in the labour of meaning-making, and in drawing connections between posted content and their lived identities and realities [8]. The YouTube platform allows for such meaning-making and connections being acted out in its comment fields, which sometimes assume functions similar to those of discussion forums. It is common for YouTubers to ask for their viewers’ opinions, and to take them into account when creating videos. Many YouTubers also take part, to varying degrees, in discussions carried out in the comment sections of videos amongst their viewers and fans. This personal interaction contributes to bringing YouTubers into closer communicative contact with their audiences. Being successful on YouTube is just as much about being able to interact with and enthuse one’s audience, as it is about creating good videos.



YouTube as a communication environment

YouTubers’ videos belong to a series of different genres. While many early vloggers were simply sharing bits of their everyday life and reflections, many of today’s YouTubers align with one of the larger genres (beauty/fashion, gameplay, comedy, etc.) while also intermittently posting vlog content in popular categories such as ‘follow me around’, ‘storytime’, ‘get ready with me’, and ‘mukbang’ (eating “together” with viewers). It is part of the appeal of the YouTuber that the ‘real’ person is on display. This is achieved through different techniques, for example by facing the spectator, using the mode of direct address, and talking as if one was conversing privately and personally. The degree to which YouTubers explicitly name or tag their videos with these categories vary. But in sum, YouTubers generally both create and post videos about special interests, as well as about their personal lives. These dimensions often intermingle within a single posted video clip.

Technologically speaking, the vlog output of YouTubers is asynchronous, in the sense that it addresses a non-present audience (Frobenius, 2014) [9]. From the perspective of Goffman, individuals engaging in self-presentation (e.g., YouTubers), have to address imagined recipients and “are under pressure to style their talk as though it were addressed to a single listener”, so that it “involves a conversational mode of address, but, of course, merely a simulated one, the requisite recipients not being there in the flesh to evoke it” [10].

Aside from addressing the camera in a personal and direct manner, YouTubers also use other strategies to create involvement with and among their audiences. Such practices also align with their need to build a following that can help them make money through YouTube views and brand collaborations. One such strategy is explicitly asking for input in the form of comments, likes, and subscriptions. Another is addressing topics that are likely to resonate with the audience for different reasons, such as fads (Pokémon Go, fidget spinners), current events (major sports or music events, seasonal happenings such as summer holidays or Christmas, and — rarely — politics), or themes and issues that are important to the imagined, and/or future, audience (school, parents, leaving home, sex, and relationships).



Communicating mental health problems on YouTube

In this paper, we are particularly interested in the role played by talk about mental health issues as a way for YouTubers to engage with audiences. When approaching the YouTuber phenomenon, trying to acquaint oneself with the types of content and social practices surrounding it, it is in fact rather striking to see how stories, disclosures, and confessionals relating to suffering, or the YouTuber having suffered, from mental health problems, play a quite important part in the output of a substantial number of vloggers. Doing searches for phrases like “anxiety vlog” or “panic attack vlog” on YouTube is a quick way to get an insight into this genre. It entails videos of YouTubers telling stories about their personal experiences of suffering from problems like social anxiety, panic attacks, eating disorders, self-harm, phobias, or even suicidal thoughts. It also entails the reactions and responses in the form of user comments, and comment discussions, following thes videos.

While it is interesting to note that some YouTubers devote their entire channel to telling stories about suffering or recovering from mental health problems, the most striking thing is that YouTubers with other content profiles not only intersperse their gaming, beauty, or other videos with “everyday” content, but also often at one point or another in the history of their channel devote one or several videos to revealing their mental health. It has been widely noted in news media and popular debate that many big YouTubers, regardless of the main focus of their content, are engaged in raising mental health awareness [11]. A notable observation in our research was that one of the most popular Swedish YouTubers seemingly even felt the need to explain to viewers why he is not routinely addressing mental health issues on his channel:

[Video transcript. Youtuber speaking into camera:] I thought I’d take a moment to respond to a question that I often get from viewers, both here on YouTube and on Instagram ... And I know that, this is something that some other YouTubers have been talking about ... And it’s about whether my life is always so great ... “Am I never sad?”, ‘Do I never feel bad?“ ... because I never show stuff like that on YouTube, but of course I do! I am just like everyone else in that I feel bad sometimes, things happen, things have happened with my family, like illnesses, and losing relatives. Life is like that for me too. It’s just that I choose not to show it, but it’s not to try to appear to have a perfect life, or that a perfect life is to be desired. I just choose to not show it, because there is already so many people on YouTube who are sharing about how they feel unwell and that they are crying, but I feel that this is not really my approach. [12]

Undoubtedly, mental health is a prominent topic among many YouTubers, and this may not be surprising as their audiences are predominantly young. Mental health problems among young people are on the rise, especially among girls (Bor, et al., 2014; Collishaw, et al., 2010; Hagquist, 2010). As much as 75 percent of all psychological disorders manifest themseelves before the age of 25. Even mild conditions, if not treated, may develop into seriously debilitating problems (Kendall and Kessler, 2002; Kessler, et al., 2003). Many, especially young people, coping with mental illness increasingly turn to online sources for information and support (Oliphant, 2013). Notably, social media sites such as YouTube have become more important in online health searches (Sarasohn-Kahn, 2008; Vance, et al., 2009).

Some young people have difficulties in accessing traditional health services (Gray, et al., 2005; Skinner, et al., 2003). Barriers prohibiting young people from accessing professional help include a preference to self-manage; a belief that no one can help; lack of service knowledge; fear of judgment and stigmatisation; fear of confidentiality breaches; and, lack of anonymity (Gulliver, et al., 2010; Rickwood, et al., 2007). Because of these factors, adolescents with mental health conditions are likely to try to search for informal alternatives in order to obtain care (Sheffield, et al., 2004; Wiist, 2004), where one prominent alternative is to go online (Gowen, 2013). A study by Oliphant (2013) of user engagement with mental health videos on YouTube found that viewers were engaging more frequently with videos that had a personal — as opposed to commercial, educational, or governmental — character.



Semantic analysis of YouTube comments

Data was collected by programmatically querying the YouTube Data API (v3) for all comments to five videos by five different popular Swedish YouTubers. The selection includes videos in which the YouTubers address their personal experiences of mental health more or less exclusively, and for which this is suggested by video titles. The selection also focused on videos to which a significant number of comments had been posted [13]. Furthermore, the selection is focused on young YouTubers, where it can be assumed (and is also sometimes evident from topics discussed in comments) that followers are young. However, we have no means by which to establish that only ”young people” are represented in analysed data.

While providing useful information for researchers, Internet-based data also introduces ethical challenges (Convery and Cox, 2012). Broadly speaking, it introduces issues relating to the role of the researcher as a passive or actively intervening subject in settings where potentially sensitive information about subjects’ situations and intentions may be gathered. More specifically to the research context, asking for and obtaining informed consent from, or otherwise interacting with, participants in contexts such as the one where data for this study were gathered is often not feasible (Sveningsson, 2001; Halvarson and Lilliengren, 2003). It is furthermore the case that the relationship between texts published online — which is the focus of this study — and human subjects, renders the issue of consent ambiguous (Markham and Buchanan, 2012). Nevertheless, our study deals with potentially very sensitive topics (e.g., suicidal thoughts, mental illness, etc.) and very vulnerable subjects (e.g., young people), which means that the collected data must be handled with particular consideration to the issue of identification (Elgesem, 2015) although it is publicly available. All quotes from the collected data included in the sections below in this article have been modified (Markham, 2012) in relation to the original online entries. They were translated into English by the authors, and all references to specific discussion topics, user names, personal names, public institutions, and geographical places were omitted.

To achieve a deeper understanding of patterns of giving and receiving social support, and conditions surrounding interactions in relation to mental health issues on YouTube, we made a closer analysis of the parts of comment threads that in turn were parts of exchanges (sub-threads) that were longer than two posts. The content of these discussions was analysed through semantic mapping, a method for securing a general overview of the topical content of a text corpus. This method comes from the field of computational corpus linguistics, and draws particularly on the strategy of exploiting information that is contained in the co-occurrence statistics of words within a given text. The key assumption is that words are defined by the contexts, of other words, in which they are used. Certain sets of words travel together, in different formations, throughout text streams.

Methods in this field, such as latent semantic analysis (LSA) (Landauer, et al., 1998), have illustrated the usefulness of inducing implicit relationships between contextually similar words for applications such as comparing documents or finding relations between terms, and to expand feature spaces in systems for machine learning and text mining.

Due to the large number of comments, and because we wanted an initial overview of topics in the discussions, this method was a fitting first step. In this study, we performed semantic mapping through the Leximancer system (Smith, 2003; Smith and Humphreys, 2006). This software provides a method for transforming lexical co-occurrence information from natural language into semantic patterns for easy visualisation. It examines a text corpus and derives from it a ranked list of important terms, based on how often they occur, and how often they occur together with other frequently occurring terms. The selected terms are fed into a thesaurus builder, which samples and resamples the text iteratively to learn a set of classifiers from the text. This is done by gradually broadening and extending the definitions of the initial terms. The resulting weighted term classifiers — the concepts — are then used to classify text chunks at a relatively high resolution (every two sentences in the case of this analysis).

The concept index for the corpus, and the matrix with information about co-occurrences between concepts, are used as the basis for calculating a two-dimensional concept map using a clustering algorithm. Finally, the connectedness (summed co-occurrence with all other concepts) of each concept in the semantic network is used to generate a third hierarchical dimension that displays relations between concept-groups — themes. The point with this method, the core elements described by Smith (2003), is to systematically arrive at a visualisation that offers a general overview of the topical content of a text corpus. Steering clear of any initial close reading of the texts, the method can help the researcher disregard any anecdotal content that may be atypical.

In the concept map (see Figure 1), the nodes (dots) represent the identified and machine learned concepts. In other words, a node such as ‘school’ does not merely represent the use of the word ‘school’ in the corpus. Rather, it stands for a set of terms that have been iteratively identified to be travelling together. The node label has been chosen to representatively name the cluster of words that make up the concept. The edges (lines) connecting the nodes represent co-occurrence patterns in text segments throughout the corpus. The connection between ‘friends’ and ‘school’ means that there is a strong relation between those two concepts, in the sense that they often occur together in the same text segments. The layout algorithm displays only the most likely connections and filters away edges that represent less common connections. The circles, with labels in capital letters, represent the conceptual themes. A theme is a cluster of concepts that have a high degree of connectedness. Note that the circles should be interpreted as boundaries — rather than as containers — on the map, meaning that it is the number of nodes (concepts) within a circle (theme), rather than the circle size, which reflects its prevalence and importance in the text.


Concept map describing expressions of mental illness on YouTube
Figure 1: Concept map describing expressions of mental illness on YouTube.
Note: Larger version available here.


Before discussing the semantic analysis in further detail it needs to be pointed out that a significant proportion of the comments in the corpus are in the form of personal and direct messages from individual viewers to a YouTuber. Comments like these are commonly quite brief, employ emojis heavily, and express deep admiration and affection for the YouTuber in question. Quotes 1–4 below show typical examples of such posts:

Quote 1: you’re the best YouTuber I love you.

Quote 2: Goddamn you are so strong started to cry [heart emoji] you’re with me in my heart.

Quote 3: I have an eating disorder right now, this video was really helpful! Thank you for running the world’s best YouTube channel!

Quote 4: Goals! Love the video so very strong.

These kinds of comments are clearly marked by the elevated status of the YouTuber, and by viewers interacting with them as fans. As shown by Quote 1 in particular — which also illustrates one of the most commonly employed phrases in the comments, namely ‘I love you’ — these kinds of posts also tend to not be addressed to other viewers. Furthermore, they do not necessarily relate explicitly to the content of the video, they seldom contain any further information about the viewer making the post, and they are commonly also not responded to by other viewers. It should also be pointed out here that YouTubers only very rarely interact with viewers over comments. These comments could, in other words, be described as expressions through which fans engage with personal, as well as emotionally charged and seemingly intimate, communication practices with an idol who is not responding.

Looking at the semantic map in Figure 1, it shows that three central themes in exchanges between users are ‘emotional support’, ‘lack of understanding’, and ‘positioning’. The theme of ‘emotional support’ relates in part, as discussed above, to viewers expressing support in the form of admiration and affection directly to the YouTuber. In part, this theme also relates to people talking about the role of courage for being able to disclose their experiences of mental ill health, primarily to their friends and family members. The theme ‘lack of understanding’ represents viewers describing themselves as not being understood by their friends, at school or elsewhere, and/or by family members, and feelings of sadness and frustration. Quite often, this is expressed using the phrase ‘no one understands’.

The theme of ‘positioning’ represents how viewers place themselves in relation to experiences of mental health conditions. The interconnected occurrences of the concepts ‘better’, ‘feel’, and ‘bad’ illustrate the common practice among users to describe whether they are currently feeling better, or worse, at the time of posting their comments. The connections between ‘hoping’, ‘better’, and ‘anxiety’ illustrate how users respond to shared experiences of others by posting get-well wishes in comments. Closely related to this, the theme of ‘suicidal thoughts’ — with the concepts ‘living’ and ‘die’, overlapping with the theme ‘positioning’ and the concept of ‘life’ — illustrates that viewers sometimes describe their experiences as being very serious. It is not uncommon for users to claim that they don’t want to live anymore, and that they have had suicidal thoughts. In the following sections we use this conceptual overview of the data as a starting point for closer readings and analysis of some pertinent topics.



Individual versus collective

First, the observations discussed above show that while YouTube comments are in part used by viewers in order to personally show support, admiration, and affect for YouTubers — they are in part used also for communication with other users in the thread, simply using the comment space established by the YouTuber’s video as a peer-to-peer social arena. Comments are used here for sharing information about, sometimes quite serious, experiences of mental health problems among viewers, and how this affects relations to family members and friends, or situations at school. Comments are also used by some viewers to ask for help, and by other viewers for providing support in the form of comforting comments, or by providing more concrete guidance on how and where to seek professional help. These comments are thus characterised by communication between viewers in a setting that could be described as a self-organised community, providing each other with peer support for their experiences.

In order to explain how a seemingly quite elusive and socially distanced context of YouTube comments can provide conditions for sharing experiences of mental health conditions, it could be argued that the frequent employment of emotionally charged and personal language contributes to establish an atmosphere characterised by warmth and intimacy that enables viewers to share experiences of a rather personal character. It could also be argued that the topic of mental ill health allows for a particularly intense form of engagement in these interactions. As YouTubers share experiences of mental health conditions in a media context with functions for talking back to the idol, viewers are invited to interact with them over a highly personal and potentially stigmatised topic on a platform lacking any functionality for private messaging. The ‘illusion of intimacy’ described by Horton and Wohl (1956) is likely both easier to establish, but also rendered less of an illusion, as viewers engage with writing comments on these particular videos.

As viewers express support for a YouTuber and share experiences of mental health problems online, they are also engaging in processes by which an informal but nevertheless networked viewer collective is organised. On the one hand, as this collective is defined largely by engaging with expressing understanding and support in relation to experiences of mental ill health, it may improve the conditions for viewers to share such experiences. It could also be the case that the presence of other viewers intensifies the experience of interacting with a YouTuber idol ‘as if they were in the circle of one’s peers’. On the other hand, however, it could also be the case that the illusion of intimacy between the YouTuber and the viewer is challenged by the presence of other viewers.

In the comments analysed here, tensions and conflicts between individual and collective aspects of the interaction sometimes emerge. The following quote, showing a discussion consisting of a comment and responses to this from three different viewers, illustrates how such conflicts can take shape. In this case it is seemingly about one user demonstrating their status by making statements claiming to have more knowledge about the YouTuber in question:

Quote 5

(user a) Ask your doctor for Sertralin! Since I started with that [...] I haven’t had any anxiety attacks for several years! ‘Like’ so Therese sees this! [heart emoji]

(user b) She said she took Sertralin for four years..., but she quit.

(user c) I’m sure she and her doctor have already talked about that, or maybe the doctor ruled it out

(user d) She doesn’t want to be on medication!

The communication in the comments are indeed marked to some extent by traits described in the original theory about parasocial interaction. In their classic paper about such interactions, Horton and Wohl [14] argued that “the new media” — at the time, radio, television, and film — had the capacity to render an illusion for an audience member that there is a personal relationship with a media figure (a presenter, actor, celebrity, or fictional character), that is analogous to the relationships in a primary group. The audience member sees the performer “as if they were in the circle of one’s peers” and responds as though in a typical social relationship. This “seeming face-to-face relationship” is what they define as a “para-social relationship”. The related “simulacrum of conversational give and take may be called para-social interaction”. Note that this is not the same as the idea of identification with a media figure, but rather about interacting with the figure (Rosengren and Windahl, 1972). Referring to a type of content that they call “personality programmes”, Horton and Wohl describe how “the greatest pains are taken by the persona to create an illusion of intimacy”. Further, they explain that it can be called “an illusion because the relationship between the persona and any member of his audience is inevitably one-sided, and reciprocity between the two can only be suggested” [15].

But the communication in the comments above are also significantly shaped by co-viewing, and interactions among fans. As the Quote 5 shows, fans sometimes ask other fans for assistance when trying to communicate with their idol, and they sometimes relate to the loosely connected network of viewers as a resource for such needs. They sometimes also intervene in the interaction of other viewers. This collective experience can thus be disruptive for the individual, and the interaction between fans is, at least in part, characterised by strategies for managing the illusion of intimacy and exclusivity with the idol. The observations made here show that while the social environment of YouTube comment threads sometimes produce conditions under which people are able to share experiences of mental health conditions collectively, the presence of an elevated subject, and individual viewers’ will to connect with the idol, may also produce tensions between viewers that undermine the conditions for realising peer support.



Competition versus cooperation

Apart from the tensions between individual and collective aspects discussed above, another type of tension appears in the relationship between the conceptual themes of ‘emotional support’/‘strength’ on the one hand and ‘diagnosis’/‘affirmation’ on the other (cf. Figure 1). This is a tension common to Internet culture between competition on the one hand and cooperation on the other. Fuchs [16] is among those who have argued that the Internet has given rise to “novel methods and qualities of domination and competition”, while it has at the same time created “new opportunities for cooperation and participation that question domination”. This paradox sits at the centre of “the ambivalent Internet” marked by “divergent responses” in “divergent audiences” [17]. Indeed, the Internet, according to Fuchs, “is an antagonistic space that by producing new networks of domination also produces potential networks of liberation” [18]. In our analysis, user discussions about diagnoses express significant levels of competition, while ‘emotional support’ discussions counterbalance this through their emphasis on cooperation.

As the map shows, the theme of diagnosis (representing terms relating to psychiatric vocabulary) is strongly related to the theme of ‘affirmation’, and comments by which viewers acknowledge that shared experiences must indeed be troublesome and stressful. Discussions between viewers here are about whether certain experiences correspond with specific diagnoses, and about the correct definitions of particular diagnoses. The following three quotes are examples of comments in these discussions:

Quote 6: It’s not necessarily the case that you have a diagnosis, but you might. Go see a doctor.

Quote 7: Get yourself some knowledge. Depression is in the DSM 5 which means that it is a psychiatric diagnosis/illness.

Quote 8: Diagnoses sometimes overlap, I have Asperger’s and I recognised myself in almost everything about ADHD but that doesn’t mean that I have ADHD.

As the analysed comments are related to videos in which YouTubers discuss their experiences of mental health problems, using quite detailed psychiatric terminology, referring to one or several specific diagnoses, the centrality of the theme of diagnosis in the map thus in part illustrates viewers reiterating terms used by YouTubers in videos. What Quotes 6–8, and also Quote 5 above, furthermore illustrate however, is that viewers also sometimes use psychiatric terminology in order to achieve legitimacy and status in relation to other commenting viewers.

In relation to our discussions in previous sections, the ways in which users employ psychiatric vocabulary could also be described as expressing a form of parasocial relationship. Portraying oneself as having psychiatric knowledge in this context is also about claiming a special closeness, through having experiences and knowledge, with a YouTuber. The employment of psychiatric terminology thus introduces hierarchies into the social relationships between viewers, as regards their knowledge of psychiatric terminology, which in turn may undermine the openness and mutuality of peer relations. Support exchanges that draw on the semantic theme of ‘diagnosis’ can often be described as scrutinizing, or even disciplinary, in character. They are often focused on discussing the correctness of the ways in which users describe experiences of mental ill health, or the psychiatric authenticity of mental health claims.

These types of discussions are contrasted in the comments by those relating to the conceptual themes of ‘emotional support’ and ‘strength’ where experiences of mental ill health are construed in terms of individual agency and ability. The theme of ‘strength’ in particular, is crucial for understanding such support. As Quotes 2 and 4 above both illustrate, the concept of ‘strength’ is sometimes employed by viewers to describe a YouTuber, and for expressing admiration for a YouTuber and/or video in question. It is sometimes used alongside the concept of ‘courage’ to emphasize that the YouTuber is courageous for making a given video, and for talking about mental health issues publicly. It is also used for describing how a YouTuber has been able to manage their life, i.e., cope, with mental ill health in everyday situations. However, the modes of expression within the theme of ‘strength’ are not only characteristic for expressing admiration towards YouTubers, they are also used similarly for supporting other viewers. As exemplified by Quote 9 below, strength can be related to having the courage to tell others about experiences of mental ill health, and to challenge the lack of understanding which is sometimes found among friends and family members. As Quote 10 shows, strength in this context can also be about finding strategies for everyday coping, and challenging the consequences of living with mental ill health (an eating disorder, in this case):

Quote 9: Talk to your family, talk to friends, call the hospital or something, until someone understands, until someone takes you seriously — fight and be strong!

Quote 10: Be strong! Try to eat as much as you can!

As mental ill health is by definition very stressful and difficult for an individual, and is linked also to disabilities and significant stigma, it is perhaps not surprising that the notion of ‘strength’ is important as a supportive concept in this context. In the comments, ‘strength’ not only provides form and content to phrases expressing support, it is also used to describe both the act of asking for help, as well as that of providing support in relation to experiences of mental ill health. It is important to note however that strength is employed by users here also in order to describe what it means to live with mental ill health also on a more general level. Importantly, this illustrates how mental ill health is not constructed exclusively in negative terms in this context, but also in ways that direct attention to the abilities and agency of the individual. ‘Empowering’ discourse like this has been observed also in previous research, most notably in studies looking at so called ‘pro-ana’ (anorexia) forums (Giles, 2006) but also sites dedicated to Asperger’s Syndrome (Clarke and van Amerom, 2007). It has provoked academic interest as well as controversy, as seemingly extreme expressions of the ways in which the hegemonic perspectives of health professionals can be challenged and resisted online, precisely as illustrated here by discursively re-interpreting previously stigmatised health categories in positive, and sometimes even celebratory, ways. On the one hand raising significant concern about the risks potentially associated with unsupervised organisation of mental health peer support online, it has on the other hand also “turned the idea of mental-illness-as-stigma on its head” [19], and directed attention to analysing how ‘identity work’ relates to the ways in which people participate and find meaning in these communities.



Conclusion: Exchanging social support in a social media setting

The observations discussed here show how online communities and social video comment threads can provide contexts in which processes of re-negotiating mental health problems in empowering terms can take form. While young people suffering from mental ill health may be able to find comfort in identifying their own experiences with those of a popular YouTuber, and take part in practices of admiring and idolising their lives, the theme of strength here allows users also to identify and renegotiate experiences of mental ill health as being markers of a decidedly able, and even successful, position.

Another interesting finding to note here is that the discussions that draw on the theme of ‘diagnosis’ commonly include recommendations to exclusively seek ‘professional help’, or to ‘go see a doctor’ as Quote 6 illustrates. Conversely, support that focuses on reinterpreting experiences of mental ill health in terms of ‘strength’ commonly does not limit recommendations to seeking out established health care institutions and psychiatric professionals, as illustrated by Quotes 9 and 10.

In terms of hindrances for peer support, there are considerable challenges for organising democratic peer support in contexts where central high-status figures, and a culture of fandom, are key features. In addition to challenges introduced by tensions between individual and collective needs, further challenges are also created due to tensions between competitive and cooperative interaction in comments. Our analysis shows that exchanges that emerge around the topics of diagnoses and psychiatric terminology, are especially marked by competition.

In terms of possibilities for peer support, our analysis suggests that interaction in social video comment threads may produce conditions under which young people are able to share experiences of mental health problems. Our analysis has shown that in those exchanges that are focusing on ‘emotional support’, and on renegotiating mental health problems in terms of ‘strength’, there is evidence of communication that is characterised by community and cooperation. Importantly, our findings indicate that certain forms of cooperative interaction may produce spaces for mutual exchanges even in otherwise hierarchical and competitive settings.

Of key importance for future research is thus to investigate further how and when users — perhaps especially young users — of social media platforms are able to engage in such cooperative interaction, and under what conditions such spaces can be used for the democratic exchange of peer support. End of article


About the authors

Simon Lindgren is Professor of Sociology, and director of the Centre for Digital Social Research (DIGSUM), at Umeå University, Sweden. His research is about social interaction, participation, power, and resistance in networked online media. He also works with developing methodological tools and strategies for analysing discursive and social network aspects of the evolving digital media landscape. He is the author of New noise: A cultural sociology of digital disruption (New York: Peter Lang, 2013) and Digital media & society (London: Sage, 2017). More information can be found at
E-mail: simon [dot] lindgren [at] umu [dot] se

Ragnar Lundström is Associate Professor of Sociology at Umeå University, Sweden. He has previously worked primarily with public discourse about welfare cheating and news media portrayals of crime victims. He currently researches the topics of climate change, political activism online, and social support online.
E-mail: ragnar [dot] lundstrom [at] umu [dot] se



1. The largest Swedish teen magazines have total circulations (and much smaller numbers of subscribers) such as: Frida (8,000), Julia (35,000), Okej (36,000), and Kamratposten (50,000). The most popular Swedish YouTuber, pewdiepie, has 58,000,000 subscribers. The tenth most popular Swedish YouTuber, Tova Helgesson, has 395,000 subscribers. See,,,,, and



4. See, for example, and

5. Head, 2015, p. 2.


7. Wilma & Emil, “Vi mâr inte alltid bra” [“We do not always feel well”], at

8. Jenkins, et al., 2013, p. 187.

9. This, then, means that we define the increasingly common live broadcasts through YouTube, and other platforms such as Periscope and Facebook, as being something other than vlogs.

10. Goffman, 1979, p. 12.

11. See, for example,;; and,

12. Wilma & Emil, “Vi mâr inte alltid bra” [“We do not always feel well”], at

13. The selection includes comments made to the following five YouTube videos:

  1. Therese Lindgren, “Frågestund: Min panikångest | Varför, hur och när får jag panikattacker” [“Q&A: My anxiety attacks | Why, how, and when do I have panic attacks”], at (623 comments).
  2. Tova Helgesson, “VARFÖR JAG ÄR SOM JAG ÄR” [“WHY I AM THE WAY I AM”], at (2,400 comments).
  4. Keela, “Deep Talk | Om att leva med en fobi.” [“About living with a phobia”] at (815 comments).
  5. Moa Murderess SWE, “Jag ville inte leva längre” [“I didn”t want to live anymore”], at (1,114 comments).

14. Horton and Wohl, 1956, p. 215.

15. Horton and Wohl, 1956, p. 217.

16. Fuchs, 2008, pp. 119–120.

17. Phillips and Milner, 2017, p. 10.

18. Fuchs, 2008, pp. 119–120.

19. Giles and Newbold, 2013, p. 477.



W. Bor, A.J. Dean, J. Najman, and R. Hayatbakhsh, 2014. “Are child and adolescent mental health problems increasing in the 21st century? A systematic review,” Australian & New Zealand Journal of Psychiatry, volume 48, number 7, pp. 606–616.
doi:, accessed 15 April 2019.

J. Burgess and J. Green, 2009a. “The entrepreneurial vlogger: Participatory culture beyond the professional-amateur divide,” In: P. Snickars and P. Vonderau (editors). The YouTube reader. Stockholm: National Library of Sweden, pp. 89–107.

J. Burgess and J. Green, 2009b. YouTube: Online video and participatory culture. Cambridge: Polity.

J. Clarke and G. van Amerom, 2007. “‘Surplus suffering’: Differences between organizational understandings of Asperger’s syndrome and those people who claim the ‘disorder’,” Disability & Society, volume 22, number 7, pp. 761–776.
doi:, accessed 15 April 2019.

S. Collishaw, B. Maughan, L. Natarajan, and A. Pickles, 2010. “Trends in adolescent emotional problems in England: A comparison of two national cohorts twenty years apart,” Journal of Child Psychology and Psychiatry, volume 51, number 8, pp. 885–894.
doi:, accessed 15 April 2019.

I. Convery and D. Cox, 2012. “A review of research ethics in Internet-based research,” Practitioner Research in Higher Education, volume 6, number 1, pp. 50–57.

D. Elgesem, 2015. “Consent and information — Ethical considerations when conducting research on social media,” In: H. Fossheim and H. Ingierd (editors). Internet research ethics. Oslo: Cappelen Damm Akademisk, pp. 14–34, and at, accessed 15 April 2019.

J. Fiske, 1992. “The cultural economy of fandom,” In: L.A. Lewis (editor). The adoring audience: Fan culture and popular media. London: Routledge, pp. 30–49.

M. Frobenius, 2014. “Audience design in monologues: How vloggers involve their viewers,” Journal of Pragmatics, volume 72, pp. 59–72.
doi:, accessed 15 April 2019.

C. Fuchs, 2008. Internet and society: Social theory in the information age. London: Routledge.

D. Gauntlett, 2011. Making is connecting: The social meaning of creativity from DIY and knitting to YouTube and Web 2.0. Cambridge: Polity Press.

D. Giles, 2006. “Constructing identities in cyberspace: The case of eating disorders,” British Journal of Social Psychology, volume 45, number 3, pp. 463–477.
doi:, accessed 15 April 2019.

D.C. Giles and J. Newbold, 2013. “‘Is this normal?’ The role of category predicates in constructing mental illness online,” Journal of Computer-Mediated Communication, volume 18, number 4, pp. 476–490.
doi:, accessed 15 April 2019.

E. Goffman, 1979. “Footing,” Semiotica, volume 25, numbers 1–2, pp. 1–30.
doi:, accessed 15 April 2019.

L.K. Gowen, 2013. “Online mental health information seeking in young adults with mental health challenges,” Journal of Technology in Human Services, volume 31, number 2, pp. 97–111.
doi:, accessed 15 April 2019.

N.J. Gray, J.D. Klein, P.R. Noyce, T.S. Sesselberg, and J.A. Cantrill, 2005. “Health information-seeking behaviour in adolescence: The place of the Internet,” Social Science & Medicine, volume 60, number 7, pp. 1,467–1,478.
doi:, accessed 15 April 2019.

A. Gulliver, K.M. Griffiths, and H. Christensen, 2010. “Perceived barriers and facilitators to mental health help-seeking in young people: A systematic review,” BMC Psychiatry, volume 10, number 113.
doi:, accessed 15 April 2019.

C. Hagquist, 2010. “Discrepant trends in mental health complaints among younger and older adolescents in Sweden: An analysis of WHO data 1985–2005,” Journal of Adolescent Health, volume 46, number 3, pp. 258–264.
doi:, accessed 15 April 2019.

C. Halvarson and P. Lilliengren, 2003. “Private explanatory systems and informed consent online,” In: M. Thorseth (editor). Applied ethics in Internet research. Trondheim, Norway: Programme for Applied Ethics, Norwegian University of Science and Technology, pp. 125–139.

J.L. Head, 2015. “Just click play: A content analysis of YouTubers,” M.A. thesis, University of South Alabama.

D. Horton and R.R. Wohl, 1956. “Mass communication and para-social interaction: Observations on intimacy at a distance,” Psychiatry, volume 19, number 3, pp. 215–229.
doi:, accessed 15 April 2019.

H. Jenkins, S. Ford, and J. Green, 2013. Spreadable media: Creating value and meaning in a networked culture. New York: New York University Press.

P.C. Kendall and R.C. Kessler, 2002. “The impact of childhood psychopathology interventions on subsequent substance abuse: Policy implications, comments, and recommendations,” Journal of Consulting and Clinical Psychology, volume 70, number 6, pp. 1,303–1,306.
doi:, accessed 15 April 2019.

R.C. Kessler, K.R. Merikangas, P. Berglund, W.W. Eaton, D.S. Koretz, E.E. Walters, 2003. “Mild disorders should not be eliminated from the DSM-V,” Archives of General Psychiatry, volume 60, number 11, pp. 1,117–1,122.
doi:, accessed 15 April 2019.

T.K. Landauer, P.W. Foltz, and D. Laham, 1998. “An introduction to latent semantic analysis,” Discourse Processes, volume 25, numbers 2–3, pp. 259–284.
doi:, accessed 15 April 2019.

P. Lange, 2009. “Videos of affinity on YouTube,” In: P. Snickars and P. Vonderau (editors). The YouTube reader. Stockholm: National Library of Sweden, pp. 70–88.

S. Lindgren, 2011. “Collective problem-solving and informal learning in networked publics: Reading vlogging networks on YouTube as knowledge communities,” In: E. Dunkels, and G.-M. Frânberg, C. Hällgren (editors). Interactive media use and youth: Learning, knowledge exchange and behavior. Hershey, Pa.: Information Science Reference, pp. 50–64.
doi:, accessed 15 April 2019.

G. Lovink and R. Somers Miles (editors), 2011. Video vortex reader II: Moving images beyond YouTube. Amsterdam: Institute of Network Cultures, and at, accessed 15 April 2019.

G. Lovink and S. Niederer (editors), 2008. Video vortex reader: Responses to Youtube. Amsterdam: Institute of Network Cultures, and at, accessed 15 April 2019.

A. Markham, 2012. “Fabrication as ethical practice: Qualitative inquiry in ambiguous internet contexts,” Information, Communication & Society, volume 15, number 3, pp. 334–353.
doi:, accessed 15 April 2019.

A. Markham and E. Buchanan. 2012. “Ethical decision-making and Internet research: Recommendations from the AoIR Ethics Working Committee (version 2.0),” Association of Internet Researchers, and at, accessed 15 April 2019.

H. Molyneaux, S. O’Donnell, K. Gibson, and J. Singer, 2008. “Exploring the gender divide on YouTube: An analysis of the creation and reception of vlogs,” American Communication Journal, volume 10, number 2, pp. 1–14.

T. Oliphant, 2013. “User engagement with mental health videos on YouTube,” Journal of the Canadian Health Libraries Association/Journal de l’Association des Bibliothèques de La Santé du Canada, volume 34, number 3, pp. 153–158.
doi:, accessed 15 April 2019.

W. Phillips and R.M. Milner, 2017. The ambivalent Internet: Mischief, oddity, and antagonism online. Cambridge: Polity Press.

D.J. Rickwood, F.P. Deane, and C.J. Wilson, 2007. “When and how do young people seek professional help for mental health problems,” Medical Journal of Australia, volume 187, number 7 Supplement, pp. S35–S39.
doi:, accessed 15 April 2019.

K.E. Rosengren and S. Windahl, 1972. “Mass media consumption as a functional alternative,” In: D. McQuail (compiler). Sociology of mass communications: Selected readings. Harmondsworth: Penguin, pp. 119–134.

J. Sarasohn-Kahn, 2008. The wisdom of patients: Health care meets online social media. Oakland, Calif.: California Health Care Foundation, and at, accessed 15 April 2019.

J.K. Sheffield, E. Fiorenza, and K. Sofronoff, 2004. “Adolescents’ willingness to seek psychological help: Promoting and preventing factors,” Journal of Youth and Adolescence, volume 33, number 6, pp. 495–507.
doi:, accessed 15 April 2019.

H. Skinner, S. Biscope, B. Poland, and E. Goldberg, 2003. “How adolescents use technology for health information: Implications for health professionals from focus group studies,” Journal of Medical Internet Research, volume 5, number 4, e32.
doi:, accessed 15 April 2019.

A.E. Smith, 2003. “Automatic extraction of semantic networks from text using Leximancer,” NAACL–Demonstrations ’03: Proceedings of the 2003 Conference of the North American Chapter of the Association for Computational Linguistics on Human Language Technology: Demonstrations, volume 4, pp. 23–24.
doi:, accessed 15 April 2019.

A.E. Smith and M.S. Humphreys, 2006. “Evaluation of unsupervised semantic mapping of natural language with Leximancer concept mapping,” Behavior Research Methods, volume 38, number 2, pp. 262–279.
doi:, accessed 15 April 2019.

P. Snickars and P. Vonderau (editors). The YouTube reader. Stockholm: National Library of Sweden.

M. Strangelove, 2010. Watching YouTube: Extraordinary videos by ordinary people. Toronto: University of Toronto Press.

M. Sveningsson, 2001. “Creating a sense of community: Experiences from a Swedish Web chat,” doctoral thesis, Linköpings universitet, at, accessed 15 April 2019.

K. Vance, W. Howe, and R.P. Dellavalle, 2009. “Social Internet sites as a source of public health information,” Dermatologic Clinics, volume 27, number 2, pp. 133–136.
doi:, accessed 15 April 2019.

W. Westenberg, 2016. “The influence of YouTubers on teenagers: A descriptive research about the role YouTubers play in the life of their teenage viewers,” Master’s thesis, University of Twente, at, accessed 15 April 2019.

W.H. Wiist, 2004. “Social networks and peer education,” American Journal of Public Health, volume 94, number 8, p. 1293.
doi:, accessed 15 April 2019.


Editorial history

Received 23 November 2018; revised 28 March 2019; accepted 5 April 2019.

Copyright © 2019, Simon Lindgren and Ragnar Lundström. All Rights Reserved.

Tube therapy: Dealing with mental health problems in social video comment threads
by Simon Lindgren and Ragnar Lundström.
First Monday, Volume 24, Number 5 - 6 May 2019