The study described in this book was motivated by important debates about the meaning and the effects of major transformations that expertise has undergone in Western societies. Over the last few decades, expertise has come to occupy a rather paradoxical position. On the one hand, we are surrounded by more expertise because ever more domains of our lives have come under the authority of “experts” and because expertise has been increasingly claimed by “non-experts,” by people lacking official accreditations. On the other hand, the right and authority of experts to make decisions that impact the lives of many and the grounds upon which such decisions are made have been called into question, as the rise in anti-elitist and populist feelings over the last decade indicates. We live, therefore, at a time, when expertise is both ubiquitous and very much challenged. The Internet has played an important role in these developments, as it has provided new avenues for the production, dissemination, and evaluation of knowledge. Since this medium has been widely used by a broad range of stakeholders in the field of mental health, throughout this book I have scrutinized the different ways in which it has shaped the performance of expertise about bipolar disorder.

The use of the Internet in mental healthcare has given rise to numerous fears and expectations. Some have hailed this medium’s potential to help improve the accessibility and affordability of mental health therapies and have also welcomed the possibility it affords people diagnosed to come together and exchange insights with others with the same diagnosis. Others have worried about the varying quality of online mental health-related information and about the ways in which such insights may affect relations between medical professionals and people diagnosed, which have been rather strained historically. These fears and hopes are to a certain extent well founded, yet they neglect the multiple, dynamic character the Internet can have as well as the different forms of engagement users can take up online, depending on their personal interests, skills, and goals, as well as on the social and cultural perspectives which shape their understanding of specific mental health conditions. In this book, I have tried to reconcile the agency of users with a perspective which sees online technologies as value-laden and capable of actively influencing people’s behaviors. This way I have been able to compare the ways in which different stakeholders seek to make their expertise manifest by using different online technologies and I have scrutinized how the design and online affordances available on different online platforms shape such performances. I have also investigated the transformations that such online engagements may lead to in the relations between people diagnosed and medical professionals, and have probed the individualizing or collectivity-generating effects the Internet can have in regard to the online performance of expertise. Furthermore, I have also focused on how different cultural norms and values inform these transformations.

The main finding of this book is that the online performance of expertise is not a straightforward process by which offline practices, tools, and approaches are transferred online, but involves additional skills and complex negotiations, which sometimes lead to unexpected configurations. Despite expectations that the availability of health-related information would lead to patient empowerment and would allow people diagnosed to re-position themselves in relation to medical professionals, the findings presented here have revealed a more complex picture, where individuals diagnosed with bipolar disorder can become highly influential through their skillful use of the Internet and by developing and cultivating important alliances with “traditional” experts. Contrary to expectations that the focus on “radical” differences in personalized and precision medicine would lead to increased individualization, online contributors diagnosed with bipolar disorder performed solidarity with others with whom they shared numerous similarities, which prompted them to engage in epistemic practices as well as affective labor. These findings prompted the development of a new conceptualization of expertise, understood as a practical and collective achievement, realized through coordination and affective labor among stakeholders who occupy multiple and shifting positions within a complex ecosystem.

Each of the empirical chapters has highlighted different ways in which expertise on bipolar disorder was performed online by various stakeholders. In this chapter, these findings are brought together and reflected upon to consider what they can tell us about the meaning and relevance of expertise in the context where promises generated by big data analytics suggest that substantial knowledge no longer describes (solely) a set of human capabilities and may no longer require human intervention in order to be applied in a variety of contexts. I highlight what the findings presented in the empirical chapters mean in relation to the key questions this book has addressed.

The Online Performance of Expertise About Bipolar Disorder

Different Stakeholder’s Use of the Internet

One of the unambiguous conclusions of this study is that different stakeholders used online technologies differently and that the choice thereof was not only informed by the resources at their disposal, but also by their goals and priorities. Thus, the analysis revealed that even though the National Institute of Mental Health (NIMH) and La Haute Autorité de Santé (HAS), the two governmental institutions studied in this book, have vast financial and technical resources at their disposal, they are reluctant Internet users, who have opted for non-interactive platforms and for conservative online technologies. I have argued that such an approach allowed them to push perspectives and research orientations that were no longer popular to the background, and it also enabled them to obscure from view the disagreements and struggles between different medical professionals who study or provide medical care for this condition. Furthermore, the choice for non-interactive platforms made it possible for them to share their perspectives about bipolar disorder authoritatively and without directly addressing the challenges brought to psychiatry by people diagnosed, their families, and even medical professionals. These governmental agencies used, however, different online tools, which were better aligned with their main goals and audiences. For instance, through the affordances available on the pdf files HAS provided, such as the audio option and the ruler, I have argued that this governmental agency used its platform for very specific educational purposes, meant to facilitate information retention and decision-making in clinical practice. The insights acquired from the empirical analyses discussed in the other chapters of this book suggest, however, that these governmental agencies need to consider the heterogeneity of the population they address in their online performance of expertise about bipolar disorder. While no access could be secured to find out how the information posted online was developed and by whom, studying how aspects of the identity of those designing and implementing this information shape their work and affect the envisaged users would be a fruitful direction for future research.

Unlike the governmental agencies discussed, people diagnosed with bipolar disorder used blogs and fora, where they could share their doubts and insights, either by initiating forum threads or by commenting on blog posts or threads written by others. It was shown that this allows for a bottom-up accumulation of perspectives and for a dialogical performance of expertise, as various treatment experiences, the advantages and disadvantages of various therapeutic approaches, and the results of self-experiments can be extensively discussed among numerous contributors. Whereas the people diagnosed studied in Chap. 4 used the Internet to share their treatment experiences and to acquire specific information, the online contributors studied in Chap. 6 used fora to come together and support others with whom they shared important commonalities. I have shown that how they engaged with the affordances available on these online platforms played an important role in the development of digital biocommunities, as it constituted a relevant similarity and reinforced sharing practices among the members of such groups. This way members of such communities came to use these online platforms as public diaries, through which they and their readers could keep track of developments in their mental, physical, and emotional states.

The different reasons people diagnosed with bipolar disorder chose to use the Internet also led to different dynamics. In general, online contributors interested in specific information shared their treatment experiences across one or several exchanges within a brief period of time and returned sometimes after a long interval to share new insights. In contrast, people interested in developing closer ties with others put a lot of time and effort into sharing their experiences online and reacting to those of others, either on the main page of the fora and/or through private messages with specific contributors. Only a very limited number of medical professionals have shared their insights on the platforms studied here, which has prevented the development of clear perspectives on how they use the Internet to acquire or share information about bipolar disorder. Since the prevalence of this condition suggests that a considerable number of medical professionals must also be diagnosed with bipolar disorder, it would be relevant to find out more about their perspectives in this regard and on the ways in which they call upon their different types of knowledge to perform expertise about bipolar disorder online. Even though the concept of performance implies that meaning comes into being through the efforts of the actors and of the audience, through the interactions between users and the various elements of online interfaces (Drucker, 2010), it was not possible to conduct interviews or to observe users within the same physical space. Addressing this aspect in future studies would lead to important contributions.

People Diagnosed, Medical Professionals, and the Internet

The findings discussed in this book also indicated that the Internet is not a neutral medium through which expertise about bipolar disorder can be performed, but that it shapes in some notable ways the position people diagnosed can occupy in relation to medical professionals. Through the use of blogs and fora, people diagnosed with bipolar disorder were able to engage in processes of knowledge production about this condition, thereby acquiring more influence and agency. As discussed in Chap. 4, through their online affordances which allow for the longitudinal accumulation in the same spaces of numerous insights, blogs and fora facilitated the development of what I have called “digitally informed hypotheses” about the effects and side effects of medications. This may have immediate empowering effects for individual users, by confirming the importance of their insights and thereby encouraging them to assume a more confident position in their interactions with medical professionals. In other cases, online platforms seemed to represent therapeutic approaches in themselves, as getting in touch with other people diagnosed and talking to them about their issues was often framed as contributing to one’s well-being and mental stability. Also from this point of view, the Internet appeared to empower some online contributors, as it assisted them in developing more agency over bipolar disorder. If recognized by relevant others, such as family members, medical professionals, and current or potential employers, this ability of people diagnosed to better manage their condition could, in turn, improve their quality of life by leading to more equal relations and more collaborative exchanges.

As people with different understandings of bipolar disorder, who had different relations with their medical professionals and different needs and possibilities in their daily life, came together online, even in brief exchanges, the Internet may also have contributed toward a more open dialogue about this condition. In time, this may lead to the development of new standards to determine what accounts for reliable information at a more general level. Importantly, while at the moment the “digitally informed hypotheses” described in Chap. 4 require the assistance of medical professionals to become clinical evidence, in time this may no longer be necessary. As the integration of different types of data fueled by personalized and precision medicine may develop further, the insights provided by people diagnosed through their online interactions may come to be recognized as clinical evidence, even in the absence of significant interventions from medical professionals. Should this occur, it will be difficult to deny the contribution of blogs and fora in enabling people diagnosed with bipolar disorder to re-position themselves more authoritatively in relation to medical professionals, as these platforms facilitated the production of new knowledge and will have thus shaped the prescription of treatment for this condition.

Nevertheless, Chap. 5 showed that while the Internet does not always favor the powerful, only a small number of individuals are able to re-position themselves and to acquire a high standing. The two bloggers studied in this chapter managed through the skillful use of this medium to become highly influential. Through their popularity, visibility, and credibility, Tracy and Fast not only shape how their readers understand and approach bipolar disorder, but also influence the production of knowledge about this condition, as they engage in productive collaborations with medical professionals. Yet, these bloggers did not use their standing to further democratize participation in the production of knowledge. Rather than promoting research projects developed by citizen scientists or various crowdsourced initiatives, Tracy and Fast informed their readers about projects undertaken by medical professionals with whom they were familiar, or with whom they personally collaborated. Through the efforts they made to familiarize readers with scientific methodology and through their advice on how to best approach medical professionals, these bloggers may, nonetheless, have already helped their readers develop more collaborative relations with the latter and may have facilitated the participation of a greater number of people diagnosed with bipolar disorder in research.

The data used in the study described in this book were collected from blog posts and forum threads with many comments; thus, the insights provided on the selected online platforms may be more representative of people’s experiences with certain therapeutic approaches than others. Focusing on the content made available on online platforms with limited (public) interactions and comparing differences in online interactivity in relation to various forms of treatment would therefore be a promising avenue for future research. Furthermore, since the results presented here are based on the experiences of readers who were motivated enough to contribute online, they do not represent all people diagnosed with bipolar disorder. More studies are therefore needed to understand the perspectives of people who use interactive online platforms for information purposes but refrain from contributing and those of people who refuse to use the Internet for health-related purposes or who do not have access to it. Studying the impact online contributions have on readers and how lasting their effects are among other online participants would also provide valuable contributions to the literature. To better understand how people diagnosed with bipolar disorder re-position themselves in relation to medical professionals through their use of the Internet, offline ethnographic studies on how medical professionals make sense of such online engagements, how online expertise is brought to medical settings, and the consequences this has are needed.

Cultural Markers and Expertise on Bipolar Disorder Online

The findings presented in this book suggest that local norms and values play an important role in how expertise on bipolar disorder is performed online. Content-wise, the differences between the contributions on American and French online platforms were largely shaped by the ways in which mental healthcare was organized in these countries and by the preferred scientific approach to bipolar disorder. Thus, differences in the insurance system lead to different uses and engagements with the technologies of online platforms. For (temporarily) uninsured American contributors, online platforms constituted valuable alternatives or stand-ins for medical professionals, as these contributors relied upon the advice of other people diagnosed to identify affordable and effective medications and to determine some alternative practices they could take up, to heighten their chances to remain stable. Furthermore, discussions about generic drugs and their different effects were very popular on American blogs and fora, but did not occupy a prominent position on French platforms. In contrast, many French contributors worried about overmedication in France, as surveys positioned France among the countries with the highest consumption rates in Europe. Another noteworthy difference concerns the social impact ascribed to bipolar disorder. Whereas most American contributors framed this condition as a disability which prevented them from gaining or maintaining meaningful employment, many French contributors were employed, at times in highly demanding positions.

The relations between medical professionals and people diagnosed seemed fairly balanced on French platforms, with some contributors confessing to more strained relations and denouncing the practice of forced hospitalization, while many others expressed trust in their medical professionals and described their interactions with them as collaborations, even though between unequal partners. In contrast, on American platforms there were more pronounced tendencies for people diagnosed to complain about the quality of medical care they received, to highlight the lack of trust medical professionals displayed toward their experiences with medications. While some contributors explained such strained relations by invoking the close financial links between psychiatrists and pharmaceutical companies, there were also many who ascribed them to the different levels of expertise of different medical professionals. Furthermore, in the US, the antipsychiatry movement seemed to have remained influential enough to be taken up on many of the blogs and fora studied. In France, psychoanalysis was still considered a possible therapeutic approach, making its presence felt through the use of its terminology in discussions about the causes of bipolar disorder and about the elements that may affect people’s response to treatment.

There were also important differences regarding the preferred types of online technologies people diagnosed with bipolar disorder from these countries used. Whereas in the US, blogs acquired more readers and more comments, in France, fora were by far the most popular, with an impressive number of readers and online contributions. There was also a notable difference regarding the extent to which online and offline practices were integrated in the US and France. In contrast to French contributors, most American online contributors rarely used blogs and fora to organize offline events or to inform readers about them. More studies are needed to acquire a better understanding of the factors that account for such differences in the use of online platforms and their affordances among American and French online contributors, and of the standing the latter may come to occupy online.

The Internet and its Individualizing or Collectivity-Generating Effects in Relation to Expertise

This book has shown that the Internet has both individualizing and collectivity-generating effects in relation to expertise, as people diagnosed with bipolar disorder shared their insights on this condition online. The analysis in Chap. 4 showed that the Internet can contribute to the development of new individual-group configurations. The online contributors studied shared their treatment experiences as individuals rather than representatives of certain groups, and their interactions with others were most of the time too fleeting for actual communities to develop. Nevertheless “light” forms of collectives did come into being. Such collectives did not develop, however, through the agency of the online contributors alone, but were facilitated by the online affordances of blogs and fora, which allowed for the accumulation of their insights in the same spaces. Importantly, even though online contributors could decide never to return to a specific online platform again, unless they deleted their contribution, they continued to be part of that collective long after their last visit, through the insights they had shared. Furthermore, certain online affordances available on these platforms may further serve as reminders of their participation, as notifications about new comments or about specific reactions to their own contributions may allow people diagnosed to keep track of reactions on that platform, even when they no longer participate actively.

Closely knit collectives could also develop online, as was shown in Chap. 6. I have termed such (sub)groups “digital biocommunities” to highlight the fact that they are not only brought together by increasingly more specific commonalities of experience, but also by shared approaches to the digital technologies used. Thus, online contributors were drawn together by similar interests, as they reacted to specific forum threads, by common attitudes about their condition and the social-political circumstances they were living in, which they could refer to in brief or more elaborate forms through the use of specific online affordances available on these platforms. In contrast, Chap. 5 revealed that the use of the Internet contributes to the development of a new type of individual stakeholder, what I have called “online expert mediators.” This new stakeholder category emerges at the intersection between the diagnosis of bipolar disorder, the acquisition of particular types of knowledge, and an individual’s skillful use of the Internet. It has also been made possible by the limited trust some of the people diagnosed and their families have toward medical professionals due to various scandals over the last two decades regarding the close ties between psychiatrists and pharmaceutical companies. The rise of online expert mediators may mark a turn from community activism to exceptional entrepreneurial selves. This book has therefore described a new form of individualization as well as a new form of collectivization that the Internet contributed to, and has shown that both forms exist simultaneously online.

By using the concept of interactional expertise to study the activities of these online expert mediators, expertise about bipolar disorder was approached as the property of individuals. In so doing, however, the role of medical professionals and other people from whom these bloggers learned before starting their own blog and the inspiration and new insights they acquired from the contributors on their platforms were not taken into account. While it may well be that Fast is right and that readers “appreciate quality above all else” (e-mail interview, 2016), more studies on the behavior of different online audiences and their online preferences are needed, to understand what such popularity is based on. Given that socioeconomic inequalities have been shown to influence the quality of (professional) care one receives when healthcare is provided through public–private partnerships (Engel & Van Lente, 2014), it is necessary to understand whether online expert mediators represent a cheap(er) way to access medical information for people with limited resources or an additional means through which those sufficiently well-off may seek to manage their health. As the concerns discussed in Chap. 2 in regard to digital and AI-based technologies highlighted, the Internet could also contribute to harming online contributors on “free” platforms, if the platform owners sell such data to employers, insurance companies, banks, and other institutions that importantly shape one’s quality of life.

Knowledge Production in the Digital Age: Contributions

In this section, the significance of the findings described above is discussed by considering their relevance in regard to processes of knowledge production in the digital age and to the role the Internet plays in them. In so doing, I argue that we need to move beyond rather simplistic approaches which see the Internet either as a quick technological fix or as a postmodern version of Pandora’s box.

The findings presented in this book are the result of a qualitative study, which focused on the narrative accounts provided by various stakeholders on different types of online platforms. Yet, as was highlighted in Chap. 2, this research unfolded against a background where developments in digital technologies and data analytics have led many to believe that the days of the relevance of human knowledge as we know it are counted. As we have seen, enthusiastic about the capacities and potential of high-performing intelligent machines, such commentators—often developers or owners of digital companies—believe the acquisition of huge amounts of data from numerous individuals (will) enable computers and algorithms to provide better, more relevant solutions to all sorts of problems (Mayer-Schönberger & Cukier, 2013). Such processes are underlined by a shift from causation to correlation in the development of knowledge, as such technologies can process at great speed previously unimaginable volumes of data, and identify relevant patterns. The insights thus acquired are appealing to many, especially to those who find themselves at sufficient distance from these technologies, not only because of the remarkable calculations upon which they are based, but also because of their apparent objectivity, as human bias is (mistakenly) thought to be largely removed from them. Such digital technologies are thought to become better alternatives to human expertise because of the many flaws that are imputed to human professionals, such as their insufficient or fragmented access to information, their limited capacities to process all relevant data and to make accurate predictions about complex phenomena, and their difficulty to overcome personal or collective ideologies and interests (Topol, 2019). In this context, knowledge produced by digital technologies and algorithms is made to shine brighter, due to its perceived neutrality and objectivity.

This book told, however, a different story, as it emphasized the importance of several typically human competencies, such as sensitivity to context, norms, and values, for the performance of expertise. By studying how governmental agencies and people diagnosed with bipolar disorder performed expertise online, I argued that expertise could better be approached as a practical achievement, realized through coordination and affective labor among stakeholders with different types of knowledge, who occupy multiple shifting positions across a complex ecosystem. One of the merits of this conceptualization is that it focuses on affective labor, showing that it plays an important role in epistemic practices. This is particularly important in the current context, where big data enthusiasts have started to invoke affective practices to argue in favor of using artificial intelligence (AI) in medical practice. Thus, Topol (2019) claims that the use of AI as diagnostic tools would free up time for medical professionals to be more “present,” “humane,” and “empathetic” during their encounters with patients. Furthermore, he argues that “it is essential that we upgrade diagnosis from an art to a digital-driven science” (Topol, 2019: Loc. 893). As this last statement shows, such visions do not consider affective labor and practices as part of (medical) expertise, of epistemic practices, but rather as something different, that can be separated from the former, and that can be called upon and managed.

The findings presented in this book have shown, however, that affective labor is closely tied to the development and performance of expertise and that this relation is not limited to people diagnosed (Chap. 6) but also mobilized in the development and translation of knowledge among different communities of practice (Chap. 5). Thus, affective practices, creativity, and adaptability are typically human capabilities that have been shown to play a significant role in the production of knowledge, as well as in its accurate interpretation and successful implementation. This is in line with arguments developed by Collins (2018) in his study on artificial intelligence, where he highlights the importance of context sensitivity and language acquisition in regard to expertise. In his opinion, such abilities can only be acquired through acculturation in a given society and are importantly tied, from a collective point of view, to embodiment, to the ways in which one makes sense of the world with and through one’s body. Furthermore, Collins emphasizes that in their interactions with others, people constantly engage in substantial, highly complex forms of “repair work,” as they make sense of, adapt, fill in, and modify the broken and incomplete information they receive, in order to develop appropriate responses. The findings presented in Chap. 4 of this book support this perspective, because they highlighted the adaptative character of expertise, as general views on the effectiveness of medications or on the manifestations of bipolar disorder were enriched by being assessed and applied to the specific circumstances, needs, and preferences of various individuals. That epistemic practices are also affective in various ways is important not only from a scientific point of view, but also politically, especially given the skewed gendered distribution and income inequality characterizing professions where care rather than knowledge is highlighted (Hochschild, 1983/2003).

Another problematic aspect is that affective labor is framed by big data enthusiasts as something that occurs without the mediation of technology, when doctors can take their eyes away from computers and scans, and look instead at the patient. The findings presented here have shown, however, that digital technologies are an integral part of certain affective practices, shaping how people perform affective labor, and who the performers and recipients of affective labor can be. Studying affective labor online is important also because of the growing amount of value that is nowadays generated from the cognition, communication, affect, and the immaterial actions of online “prosumers” (Berardi, 2009; Hardt & Negri, 2004) and because scholars are divided about the role of such labor in digital media economics. Thus, many have criticized users’ engagement with digital technologies as a form of free labor (Lupton, 2014; Mitchell & Waldby, 2010; Terranova, 2000; Waldby & Cooper, 2008), particularly since people are typically required to give up ownership over their data and any claims over potential profits that can be made from them. The findings presented in this book provide, however, a more nuanced perspective, as they show that online contributors perform affective labor in order to assist others, but they also benefit in various ways from these efforts, either because such practices contribute to their well-being or because they receive similar assistance from others, when they need it.

This book also showed that norms and values play an important role in the performance of expertise, not as sources of bias that need to be overcome, but as factors that motivate people to contribute in specific ways to knowledge production. Thus, Chap. 4 highlighted the role personal values and preferences play in how people assess treatment effectiveness and the therapeutic improvements or new treatments that they seek to contribute to, by sharing their personal experiences and insights. Whereas currently dominant neoliberal imperatives encourage individuals to assume responsibility for their health, the findings in Chap. 6 showed that lay expertise is importantly tied to the value of solidarity. These insights are aligned with the work of several philosophers of expertise (Goldman, 2018; Quast, 2018), who have highlighted the moral dimension of expertise. These authors have argued that people endowed with expertise are expected to behave responsibly, reflexively, and virtuously, and to assist others to the best of their abilities. While solidarity is largely neglected from such considerations, the findings presented in Chap. 6 suggest that it is a value that is worth paying more attention to in relation to expertise. By revealing the importance of affective labor for epistemic practices, this book helps expand the category of behaviors people endowed with expertise engage in and need to be aware of.

The findings of this book which highlighted the importance of communities for knowledge production are also important, as we currently “live in an environment in which datafied individualization of health responsibility appears like an inevitability” (McFall, 2019: 61). From this point of view, the Internet could be seen to finally live up to some of the visions it generated in its early days, when many expected it to lead to the development of virtual communities, as people with the same diagnosis could thus easily come together and talk about matters of interest, regardless of where they lived (Eysenbach, 2005; Hardey, 1999). By showing that people diagnosed with bipolar disorder perform expertise online, this book contributed to a growing body of literature (McFall, 2019; Prainsack & Buyx, 2017; Sharon, 2017; Van Hoyweghen & Rebert, 2012) that seeks to highlight the importance of this value in (mental) healthcare practices.

The conceptualization of expertise that I put forward highlights the complexity and diversity which characterize the production of knowledge in the digital age and the dynamic ways in which authority and influence can be (re)distributed among relevant stakeholders. This is important because in many situations a priori agreements about the identity of the “experts” or of the stakeholders involved in the development of expertise no longer hold, the relevance and impact of their input differ depending on the problem they are called to solve, and their role may change as new tools or practices are introduced (Waardenburg et al., 2018). From this point of view, the findings presented here constitute a contribution to the growing body of expertise studies, which emphasize the fact that nowadays individuals or groups socialized within different epistemic cultures have to intensely negotiate the relevance of their insights and the reach of their influence (Holst & Molander, 2018). By focusing on the knowledge with which different stakeholders are endowed, the new approach I put forward also highlights the importance not only of combining different types of theoretical knowledge, but of bringing together scientific insights with relevant local factors that only specific stakeholders may have knowledge of. This new conceptualization of expertise is thus in line with arguments recently provided by Barrotta and Montuschi (2018: 395), who state that “local knowledge coming from ‘other sources’ is often necessary to fill the gap between experts’ knowledge and correct judgement calls.” Another merit of this approach to expertise is that it draws attention to the importance of the ecosystem within which relevant stakeholders are based, as even sound recommendations may be thwarted or lead to less desirable effects depending on the available infrastructures, policies, legal provisions, and dominant political climate.

Through the findings presented, this book is therefore a contribution to the work of STS scholars who have cautioned against the hype surrounding digital technologies and their epistemic potential (Neff, 2013) and have argued, instead, that certain typically human capabilities continue to be very much needed in processes of knowledge production. These findings are thus important reminders of the value of nuanced perspectives, developed through rich, detailed studies, for a better understanding of the different functions and roles that people and digital technologies fulfill in processes of knowledge production. Whereas considering the online knowledge thus generated as mainly the merit of the technologies used would be a serious mistake, neglecting the active ways in which the latter do shape the content produced would be a danger that STS scholars have repeatedly warned against (Swierstra, 2016; Wyatt et al., 2016).

This book showed that different stakeholders performed expertise about bipolar disorder in different ways, depending on their skills and resources, on the ways in which they (could) chose to react to broader social transformations, and on the values that underpinned their engagements. The ways in which science and knowledge production are currently structured and organized increasingly require experts to make bold claims and to issue clear, unwavering recommendations about the problems they are called to solve. The varied and numerous accounts, doubts, and uncertainties expressed by the people diagnosed with bipolar disorder studied here constitute important reminders, however, that modesty, diversity of opinions, and constructive criticism are essential elements in the development of sound knowledge. By cultivating and enhancing the Internet’s potential toward inclusive participation and (self)reflection, inquiries about the meaning and relevance of expertise will continue to generate enthusiasm, excitement, and heated debates. Many more exhilarating questions are thankfully opening up.