Part III - Digital Media and Adolescent Mental Disorders

9 Depression and Anxiety in the Context of Digital Media

Megan A. Moreno and Anna F. Jolliff

Over the past two decades, scientists have strived to understand the relationship between digital media use and two common mental illnesses in adolescents: depression and anxiety. The lifetime prevalence of depression or anxiety among youth increased from 5.4% in 2003, to 8% in 2007, to 8.4% in 2012 (Bitsko et al., 2018). In this chapter we begin by defining depression and anxiety, and addressing the state of the science around the relationship between these two mental illnesses and social media use. We then consider both potential problematic digital media behaviors for depression and anxiety, as well as potential benefits of social media for youth with these conditions. Throughout this chapter we consider other factors that may influence the proposed relationships among digital media use, depression, and anxiety. We conclude the chapter with considerations of clinical implications and future research directions.

Theories of Depression and Anxiety

This chapter will frequently discuss symptoms of major depressive disorder (MDD) and symptoms of generalized anxiety disorder (GAD). We will refer to these as “depression” and “anxiety” for short, but keep in mind that, first, there are many types of depression and anxiety; and second, the research described here is not limited to participants with clinically significant MDD or GAD, but often simply with depressive or anxious symptoms.

The fifth edition of the Diagnostic and Statistical Manual (DSM-5) defines a major depressive episode as a period of at least two weeks during which an individual experiences either a depressed mood or a markedly diminished interest in normal activities (American Psychiatric Association, 2013). In adolescents, the depressed mood many manifest as irritability. Also key to the diagnosis of depression is decreased performance or increased distress in a major area of life, such as school, work, or relationships. In contrast, GAD is characterized as a period of at least six months during which a child experiences excessive and uncontrollable worry, worry that is inappropriate or out of proportion to the anticipated event. In children, this worry is often about competence or performance. Additional symptoms of GAD include restlessness, difficulty concentrating, or sleep disturbance.

There are many theories to explain the development, maintenance, and treatment of depression and anxiety. It will be helpful to have a working theory of depression and anxiety to understand its relationship to digital media use; as such, we will describe two example theories here. However, keep in mind that there are many qualified theories to describe the etiology and maintenance of mental illness – many more than can be discussed in this handbook.

One such theory is cognitive theory. According to cognitive theory, “cognition is at the core of human suffering” (Sommers-Flanagan & Sommers-Flanagan, 2018, p. 273). Factors such as early life events, genetic predisposition, and caregiver modeling lead individuals to develop rigid and negative beliefs about the self, other people, and the world at large. When faced with a life stressor, an individual’s core beliefs are triggered and present as automatic thoughts. Over time, the repeated activation of automatic thoughts results in information processing, emotions, and behaviors that are consistent with depression or anxiety. Core beliefs consistent with depression or anxiety might include “I’m unlovable,” “I’m powerless,” or “I’m defective.” There are many critical events during adolescence – and, relevant to this chapter, events on social media – that might activate thoughts like these. According to cognitive theory, depression and anxiety can be reduced through the conscious revision of automatic thoughts and the core beliefs underlying them. This “validity testing” is often performed in partnership with a therapist or another trusted person.

A second theory through which we will view depression and anxiety, in relation to digital media use, is multicultural theory. This is not so much a theory for the etiology of illness as it is a lens, or an orientation, that all theorists must integrate in order to effectively explain and diagnose illness as well as guide treatment (Bitsko et al., 2018). In short, multicultural theory suggests that mental illness develops in response to the oppressive nature of the dominant culture. According to multicultural theorist Derald Wing Sue, “people of color from the moment of birth are subjected to multiple racial micro-aggressions, from the media, peers, neighbors, friends, teachers and even in the educational process” (Sue et al., 2010, p. 212). It is easy to imagine how symptoms of depression and anxiety might silently develop in response to these social forces. Although multicultural counseling does not emphasize diagnosis (in part because psychopathology has been defined using Westernized notions of normativity) treatment is possible. Healing from depression and anxiety, from a multicultural lens, must integrate culturally responsive processes and practices, often in community with culturally competent others.

State of the Science: Social Media, Depression, and Anxiety

Over the past decade, the empirical literature and lay news media have addressed at length associations between social media, depression, and anxiety. The sheer volume of studies in this area has led to a recent upswing in published systematic reviews on this topic. Two such systematic reviews found a small positive association between social media use and these two mental illnesses; however, these reviews noted that the quality and practical significance of these studies are often low, and they are typically not designed to capture the nuance of the effect (Keles et al., 2019; Piteo & Ward, 2020). Another 2016 systematic review analyzed 70 studies looking at the relationship between social media use and depression or anxiety, and found that while passive use of social media was not associated with depression, specific behaviors (e.g., self-comparison) were (Seabrook et al., 2016). In sum, recent systematic reviews indicate that research to date is not designed to piece apart the nuanced relationship between social media use and mental health.

Challenges in Studying Depression and Social Media

The relationship between social media, depression, and anxiety is a challenging area of research for several reasons. As illnesses that wax and wane over time, assessments of depression or anxiety at a single time point may not fully capture the illness experience. A critical approach is to use measurements of mental illness that are shown to be valid measurements of the illness in question, such as the Center for Epidemiologic Studies.

The Depression Scale for Children (CES-DC), the Patient Health Questionnaire-9 (PHQ 9), the Generalized Anxiety Disorder scale – 7 (GAD 7), and the Screen for Child Anxiety Related Emotional Disorders (SCARED) are all empirically supported measures of depression or anxiety in adolescents and young adults (Cannon et al., 2007; Keles et al., 2019; Kroenke et al., 2001; Piteo & Ward, 2020; Richardson et al., 2010; Weissman et al., 1980). Despite the availability of valid and reliable screening tools for depression and anxiety, some studies have not employed such tools and have selected instead ad hoc measurement tools, making the results of these studies difficult to interpret (Twenge & Campbell, 2019).

A second challenge when studying depression and social media is determining a measurement approach for social media use. Most commonly, studies focus on quantity of social media use in terms of hours or minutes. Screen time is typically measured using self-reported estimates, which are often inaccurate (Ellis, 2019; Moreno, Jelenchick, et al., 2012). Other studies have employed passive observation to understand screen time, which involves asking a participant to download an application on their phone to track their social media use (Messner et al., 2019). However, these studies tend to capture time on a specific device, and adolescent technology use is known to incorporate multiple devices. It is recommended that future studies focus on other aspects of adolescents’ technology experiences, such as quality or importance placed on use. These measurement approaches are less common, and present new ways to examine media’s relationship to adolescent health.

Third, many studies examining social media use, depression, and anxiety do not focus on normative social media use. There is a wealth of studies measuring problematic social media use or social media “addiction,” specifically, as opposed to various qualities of normative use (Duradoni et al., 2020; Hussain et al., 2020). While these constructs are relevant to adolescent mental health, positive associations between problematic social media use and mental illness may not apply to normative social media use (Przepiorka & Blachnio, 2020). Thus, much of the research on social media use and depression or anxiety among adolescents actually comments on nonnormative use, and the implications for the general population of adolescents cannot be inferred.

Key Hypotheses on the Relationship between Social Media, Depression, and Anxiety

As we consider several key hypotheses in the literature on the relationship between social media screen time, depression, and anxiety, we ask you to keep in mind the measurement and study design issues that may influence these study findings.

The first hypothesis posits that there is a positive linear relationship between social media, depression, and anxiety. That is, as social media use increases, so does risk for anxiety and depression. From a cognitive theoretical perspective, it may be that exposure to certain stimuli on social media (e.g., a photo-shopped image, a photo from a party to which one was not invited, a heartbreaking news story) might activate or reinforce existing negative beliefs about oneself (“I’m worthless”) or the world (“everything is out of control”). Further, time spent on social media might displace time spent on other behaviors, behaviors that may have resulted in mental health-promoting thoughts (e.g., “I have a knack for piano” or “I’m a good teammate”). The “crowding out” hypothesis explains positive associations between depression and screen time by saying that screen time is related to depression when it occurs at the expense of other beneficial activities (McDool et al., 2020; Twenge, Joiner, Martin, & Rogers, 2018). As discussed, multiple systematic reviews support a weak positive relationship between social media use and adolescent depression or anxiety. However, because the studies reviewed are often cross-sectional, it is difficult to ascertain whether social media use causes depression or anxiety, or whether the presence of anxiety or depression makes one more prone to use social media (and less prone, for example, to activities such as exercise, in-person socialization, vocational pursuits, or recreation). Certain studies have detected greater risk for anxiety and depression after a certain threshold of social media use is met, which has been cited as three hours per day (Riehm et al., 2019), four hours per day (Barman et al., 2018), and nearly five hours per day (O’Keeffe et al., 2011).

A second hypothesis states that a U-shaped curve best captures the relationship between internalizing symptoms and social media use, with negative mental health associated with very low or very high use. There is some empirical support for this hypothesis (Belanger et al., 2011; Liu et al., 2016; Moreno, Jelenchick, et al., 2012). From a multicultural perspective on anxiety or depression, adolescents with nonnormative (very high or very low) use may be in other ways alienated from the dominant culture. High social media use might indicate a lack of participation in other areas of life, while very low social media involvement might signify estrangement from what now constitutes a developmentally appropriate activity: social media. Further, youth at the very high and very low ends of social media may be socioeconomically disadvantaged; they may live in low-resource settings, without consistent access to the Internet, or in contexts where social media use is the only activity available. Financial hardship, or disempowerment, is associated with depression and anxiety (Selfhout et al., 2009). Further, the U-shaped curve may also result from the stimuli encountered on social media. Frequent social media use puts minority youth at risk of daily, and sometimes hourly, evidence of minority oppression in the form of news media. Similarly, those who choose to stay off of social media may be trying to avoid these stimuli. Importantly, detecting the U-shaped curve requires the use of analytic approaches beyond traditional linear regression. Therefore, it is possible that studies presumed to support a linear positive relationship actually support the U-shaped curve hypothesis.

A third hypothesis is that there is no relationship between social media and depression, or social media and anxiety. More specifically, this hypothesis suggests that there is no population-level clinically significant relationship between these illnesses and social media use. Rather, certain subgroups may be at elevated risk for depression and anxiety due to social media use (Radovic et al., 2017) while for others there is no relationship, and for still others social media use actually promotes mental health. At a population level, this variability cannot be detected. From a multicultural perspective, this makes sense; one would never expect to find a “population level” effect of social media on depression or anxiety, in a world where oppression (and consequential mental illness) is not equally distributed. A white, cisgender child from a middle-class household is likely to see aspects of their own life reflected online; in contrast, those with any number of minority identities may feel “othered” by going online. The “no relationship” hypothesis is supported by several studies that have identified no population-level statistically significant association between social media use and depression or emotional problems (Anjum et al., 2019; Fardouly et al., 2018; Ferguson, 2021; Jelenchick et al., 2013). Given the additional difficulty of publishing statistically insignificant findings, it may be that more studies have detected the “null” relationship than have been published.

The fourth and final hypothesis is “it’s complicated,” which mirrors a common relationship status adolescents themselves like to use. The majority of studies focus on screen time as a measure of social media use, and it may be that other aspects of social media use relate more to depression and anxiety than does screen time. This hypothesis finds theoretical support from a cognitive perspective of anxiety and depression. Different online behaviors generate different thoughts, thoughts that may either reinforce or challenge beliefs about the self. Cognitive theory further states that avoidance is a key behavior maintaining illnesses like anxiety and depression. If an adolescent scrolls through social media primarily as a means of avoiding – rather than confronting – dreaded stimuli, social media use would likely contribute to the maintenance of anxiety. In contrast, a youth who uses their Finsta (Fake Instagram) to air the “less acceptable” sides of themselves may learn over time, through this online “validity testing,” that what they thought were unacceptable features are warmly received by peers. Last, children who already have depression or anxiety might assign social media different worth; they may compulsively check social media for evidence in support of their own worth, while a child who affords social media no such power would not feel this attachment toward use.

Few studies have been designed to test Hypothesis 4; that is, few assess the specific features or intentions underlying adolescents’ social media use. As such, it remains to be seen how the specific uses of social media differentially relate to depression and anxiety. A recently developed tool to measure the quality of use, the Adolescents’ Digital Technology Interactions and Importance scale, is a promising means of evaluating the importance that adolescents assign to different affordances of technology, including technology to bridge online/offline experiences and preferences, technology to go outside one’s identity or offline environment. and technology for social connection (Moreno et al., 2020). Tools like these are needed to understand the nuanced relationship between social media use on depression and anxiety in adolescents.

Where Are We Now?

A 2020 paper synthesized data from systematic reviews and meta-analyses between 2014 and 2019. This included cohort, longitudinal, and ecological momentary assessment studies (Odgers & Jensen, 2020). They authors concluded that most research has been correlational, focused on adults, and has led to a mix of conflicting results. They also observed that most studies report “small associations … that do not offer a way of distinguishing cause from effect and, as estimated, are unlikely to be of clinical or practical significance” (p. 336). It has become increasingly evident that the current literature may not support Hypothesis 1, but that Hypotheses 2–4 above remain available for more nuanced and high-quality studies to address.

Potentially Problematic Digital Media Behaviors for Depression and Anxiety

As discussed, the relationship between screen time and depression and anxiety in adolescents is not straightforward. Thus, rather than focusing on time spent on social media, an alternative approach is to focus on specific digital behaviors and their relationships to depression and anxiety. Problematic or addictive social media use is discussed elsewhere in this handbook. The present chapter will discuss problematic aspects of normative use that are associated with depression and anxiety among adolescents. As a reminder, depression is often characterized by symptoms such as low mood, fatigue, diminished pleasure in activities, and thoughts of death, while anxiety is characterized by symptoms like excessive worry, sleep disturbance, and restlessness. In this section, we will examine how symptoms of depression and anxiety are related to specific adverse experiences on social media, including exposure to cyberbullying, troubling news media, and certain types of highly visual social media. Next, we will consider other variables that strengthen the observed relationships between social media use, depression, and anxiety, including fear of missing out, sleep, and gender.

Risk 1: Adverse Online Experiences

Cyberbullying

The majority of teens have experienced an instance of cyber-victimization at some point. The most common categories of cyber-victimization include name-calling, spreading of rumors, and receiving explicit or unwanted images. However, cyberbullying is less common – and often more serious (Anderson, 2018). Cyberbullying has occurred when cyber-victimization is repeated, intentional, and unwanted (Ansary, 2020). Unsurprisingly, experiencing cyberbullying is linked to depression and anxiety (Alhajji et al., 2019; Barry et al., 2019; Tian et al., 2018; Willenberg et al., 2020). Indeed, a previous study found that online harassment was key to explaining the observed relationship between social media use and depressive symptoms (Kelly et al., 2018). From a cognitive theoretical perspective, experiences of cyber-victimization may cause or reinforce core beliefs associated with depression and anxiety (e.g., “I am unlovable,” “I am powerless”).

It is also important to adopt a multicultural perspective when understanding the relationship between cyberbullying and depression or anxiety. Adolescents of color frequently experience online racism, including online micro-aggressions, discrimination, and hate crimes (Moreno et al., 2016). Some research suggests, however, that racial minority adolescents are actually less likely to report cyberbullying (Alhajji et al., 2019; Edwards et al., 2016). It is unclear whether this finding is due to a real difference in the prevalence of cyberbullying; increased stigma in certain racial or ethnic groups around reporting cyberbullying; or because racism is so common to minority adolescents’ online experience that they do not recognize it as cyberbullying. Just as offline experiences of racism and bullying are linked to symptoms of depression and anxiety, so these symptoms can emerge from the same interactions online (Cannon et al., 2007).

Adolescent females and members of sexual minority groups are also more at risk of upsetting experiences online (Kroenke et al., 2001; Richardson et al., 2010). Research suggests females are more likely to report cyberbullying and are more negatively affected by it (Alhajji et al., 2019; Rice et al., 2015). Adolescent females are also more likely to be victims of digital intimate partner violence (Burns et al., 2016). In combination with poor sleep, experiences of cyberbullying have been shown to fully explain the relationship between high social media use and psychological distress among females (Kelly et al., 2018).

News Media

Research shows that 77% of adolescents obtain their news through social media (Robb, 2020). The news is frequently troubling and, as such, exposure to it, via social media, may elevate symptoms of anxiety and depression for certain adolescents. This may be especially true for members of stigmatized or disenfranchised groups, as well as people with existing depressive or anxiety symptoms (Caporino et al., 2020; Sahoo et al., 2020; Weinstein, 2018). In the year 2020, for example, adolescents in the United States could not open their most-used social media apps without confronting news of a global pandemic, racial injustice, wildfires across California and Oregon, and a highly contentious election. Black and Hispanic or Latino teens describe finding the news to be more important, and feeling more affected by the news, than their white counterparts (Mundt et al., 2018). Adolescents living in the United States who identify as black, transgender, or undocumented risk facing news of injustice against themselves or others with their same identities nearly every time they log into social media (Campbell & Valera, 2020; Ince et al., 2017; Leopold & Bell, 2017; Robb, 2020). While social media is a platform on which many people can and do effectively advocate for social justice and raise awareness about social injustice, both cognitive and multicultural theories help to explain why encounters with news on social media might perpetuate depression and anxiety. The news can reinforce negative beliefs about the world (it’s dangerous), oneself (I’m powerless), and one’s future (people like me don’t make it very far).

Risk 2: Highly Visual Social Media

Exposure to certain highly visual social media (HVSM) is a risk factor for depression and anxiety in some adolescents. Undoubtedly, visual social media can be positive. However, in this section, we use HVSM as shorthand for risky HVSM – for example, media that enables users to modify or “improve” their appearance before uploading (Weissman et al., 1980). Many of the most popular social media platforms for adolescents (Instagram, Snapchat, and most recently TikTok) are visual platforms that allow for appearance modification (Anderson & Jiang, 2018). While use of HVSM has also been associated with disordered eating, the relationship between social media and disordered eating is covered elsewhere in this handbook. The present section will explore the relationship between HVSM and depression and anxiety.

Some of the thoughts and feelings that characterize depression and anxiety may be triggered by exposure to HVSM. Feelings of worthlessness or fears of inadequacy may be sparked or exacerbated by frequent exposure to visually “perfected” images.

Youth may compare themselves to the people they “follow,” and find themselves lacking (Marengo et al., 2018). From a multicultural perspective, visual media are uniquely able to transmit messages from the dominant culture: how to look, how to behave, and the types of people and behaviors that are deserving of praise.

People with existing tendencies toward poor body image are at particular risk of depression or anxiety as a result of exposure to HVSM (Kelly et al., 2018; Marengo et al., 2018). The tendency to be bothered if tagged in an unflattering picture is associated with depression among college students (Robinson et al., 2019). HVSM also allows for taking, editing, and uploading pictures of oneself online, which has been linked to anxiety in college students (Mills et al., 2018; Wick & Keel, 2020). Appearance-related social comparisons, which are uniquely afforded by HVSM, have been associated with depression (Choukas-Bradley et al., 2020; Hawes et al., 2020). Engaging with “pro-ana” (pro-anorexia) media or “thinspiration,” which often contains depictions of thinness, “clean” foods, and calorie-deficient diets, has been linked with depression and anxiety (Fitzsimmons-Craft et al., 2020; Jennings et al., 2020). From a cognitive perspective, social media may reinforce the negative belief that one’s worth is tied to bodily appearance. If viewers perceive themselves as failing to meet these standards, depressive or anxious symptoms may increase. It may also be that adolescents who are anxious or depressed and dissatisfied with their bodies are more likely to engage with HVSM in pursuit of information (a like, comment, or share) challenges or confirms of their self-beliefs.

However, despite the theoretical justification and some empirical support, a recent scoping review on HVSM and depression found that the relationship between HVSM and depression is inconclusive (McCrory et al., 2020). It may be that the relationships between HVSM and depression are simply better explained by other variables. The absence of an effect may also be due in part to a lack of research studies designed to detect this effect: research on social media use does not always distinguish between HVSM and other social media, let alone differentiate between positive and negative forms of visual social media. Further, research typically relies on quantitative methods to evaluate the relationship between HVSM and depression and anxiety, which lacks richness and possibility for participants to elaborate on their experiences.

The relationship between social media use and depression or anxiety also may be dependent on a variety of factors that increase risk for internalizing symptoms. Several of these potential moderators are discussed below.

Fear of Missing Out

Fear of missing out, or FOMO, is defined as the “pervasive apprehension that others might be having rewarding experiences from which one is absent” (Przybylski et al., 2013, p. 1841). FOMO in adolescents has been independently associated with both depression and anxiety and, less consistently, with social media use (Barry et al., 2017; Franchina et al., 2018; Przybylski et al., 2013). Given that social media is a place where the (often enviable) experiences of others are constantly on display, it is not difficult to explain the link between social media use and FOMO. More complex is to explain why adolescents who are higher in FOMO are more at risk for depression or anxiety as a consequence of social media use (Fabris et al., 2020). It may be that, for adolescents with tendencies toward FOMO, exposure to friends’ and influencers’ “highlight reels” creates feelings of worthlessness, worry, and dissatisfaction with one’s own daily life. From a cognitive perspective, scrolling through social media might trigger automatic thoughts, such as “no one invites me to anything” or “my life sucks in comparison with hers.” The action of scrolling through social media may also be motivated by FOMO, as depressed or anxious adolescents search for evidence to assuage or confirm the belief that they are missing out.

At present, research is mixed on whether FOMO affects the relationship between social media use and depression or anxiety in adolescents. While there is ample evidence that FOMO is associated with problematic social media use and problematic smartphone use (Franchina et al., 2018; Przepiorka & Blachnio, 2020), there is insufficient evidence that FOMO explains or strengthens the relationship between typical use and depression or anxiety at a population level. In some cases, this absence of an effect may be due to insufficient measures of social media; as discussed, insufficient measures focus solely on time spent, rather than activities performed or experiences had while online.

Sleep

Sleep is critical to consider in any study of social media and mental illness. Indeed, sleep is perhaps the most consistently supported variable to explain the relationship between social media use and depression or anxiety (Alonzo et al., 2019; Kelly et al., 2018; Lemola et al., 2015; Oshima et al., 2012). However, studies suggesting that sleep explains the relationship between social media use and depression and anxiety have been largely cross-sectional, meaning directionality is subject to interpretation. Before exploring these hypotheses, it is important to note that poor sleep (e.g., sleeping too much or too little, trouble falling asleep) is actually a symptom of both depression and anxiety. Thus, sleep trouble is central to the experience of depression and anxiety for many people.

One hypothesis suggests that social media use causes sleeplessness, which in turn causes or exacerbates symptoms of depression or anxiety. Social media may cause sleeplessness by displacing sleeping hours and delaying bedtime (Quante et al., 2019). The blue light exposure associated with social media use may disrupt melatonin and cause wakefulness (Blass et al., 2006; Levenson, 2016; Wahnschaffe et al., 2013). It may be that social media is uniquely stimulating as compared to nonsocial online activities, given that it contains a wealth of self-relevant social information and capacity for social interaction.

A second hypotheses interprets the association in the reverse direction. That is, it may be that adolescents who are already depressed or anxious are more prone to sleep disruption. In turn, disrupted sleep leads to social media use, perhaps as adolescents seek distraction or support online. However, this hypothesis is contentious. Some research has shown that the relationship between poor sleep and social media use cannot be explained by existing depression or anxiety (Twenge & Campbell, 2019; Woods & Scott, 2016). Thus, all teens – not just those who are anxious or depressed – may benefit from finding soothing activities that are less stimulating than social media, and from following the American Academy of Pediatrics’ recommendations to keep devices out of bedrooms at nighttime (Moreno et al., 2016).

Gender

Some studies have suggested that gender may influence the effect of social media use on depression and anxiety. Several individual studies have found that adolescent females are more likely than males to experience depression associated with social media use (Barthorpe et al., 2020; Kelly et al., 2018; Twenge, Joiner, Rogers, & Martin, 2018; Twenge & Martin, 2020; Waterloo et al., 2018). From a multicultural theoretical perspective, the potentially unique negative effects of social media for adolescent females finds support. The design of social media (including affordances for appearance feedback and negative self-comparison) may be uniquely oppressive to adolescent females, particularly females of color, who face great pressure from the dominant culture to conform to a certain beauty ideal (Coyne et al., 2019; Messner et al., 2019). Under multicultural theory, symptoms of anxiety and depression (e.g., guilt, worry, restlessness, diminished pleasure) should not be understood as reflecting psychopathology, but instead reflecting reasonable reactions to the dominant culture.

However, summaries of research found in recent systematic reviews conclude no consistent effect of gender on the relationship between social media use and internalizing symptoms, and typically conclude that more research is needed on this topic (Keles et al., 2019; Piteo & Ward, 2020). A meta-analysis evaluated 67 independent samples of a combined 19,652 participants, and found that the effect of gender on the relationship between time spent on social media and psychological well-being was insignificant (Huang, 2017). A recent review of reviews reported something slightly different: after controlling for confounding variables, the least-depressed adolescent females in the sample had “slightly increased risk for depressive symptoms with daily social media use”(Odgers & Jensen, 2020, p. 341).

While the moderating effect of gender is inconclusive, research supports that males and females do use social media differently (Boyle et al., 2016). Research suggests that females spend more time online and are more likely to say they are nearly constant online users compared to adolescent males (50% vs. 39%) (Anderson & Jiang, 2018; Duggan, 2013). Females are more likely than males to use social media for self-expression, including expression of joy and pride, as well as expression of negative emotions, such as worry, stress, and depression (Egan & Moreno, 2011; Moreno, Christakis, et al., 2012; Waterloo et al., 2018). Given that females compared to males are likely to use online platforms for emotional expression, it is also possible that females with depression turn to social media more readily for support. Thus, it may be that social media use does not predict depressive symptoms, but greater depressive symptoms predict more frequent social media use, especially among females (Heffer et al., 2019).

Potentially Beneficial Digital Media Behaviors for Depression and Anxiety

Certain uses of social media may promote mental health among adolescents. As previously mentioned, cognitive theory would suggest that online experiences that confirm positive beliefs about the self, and those that challenge or invalidate negative beliefs, are likely to reduce depression or anxiety. From a multicultural perspective, uses of social media that create identity-affirming alternatives to offline spaces may mitigate depression and anxiety. However, the hypothesis that social media use directly reduces anxiety or depression is difficult to test. Similar to studies that try to assess whether social media use directly increases depression or anxiety, there are methodologic barriers to these assessments. That being said, empirical research does support a positive association between social media use and adolescents’ mental health. This is especially true for adolescents with depression and anxiety, adolescents with unique and marginalized identities, as well as typical adolescents who seek to maintain or promote mental wellness online.

Benefits of Social Media for the Typical Adolescent

Typical adolescents report using social media in ways that may ward off depressive and anxious symptoms, both by seeking information related to these symptoms and by finding support and connection online (Rideout et al., 2018). Research has shown that adolescents often feel happy, amused, or closer to friends while using social media (Weinstein, 2018; Wenninger et al., 2019). While studies over the years have repeatedly demonstrated social networks and support contribute to overall and mental health, too much online social networking may put one at risk for negative experiences, cognitions, and emotions (Ahn, 2012; Longobardi et al., 2020; Negriff, 2019; Rajani et al., 2011). Those who experience isolation, stress, and unmet needs in their offline worlds may find corrective experiences or buffering effects by going online (Nick et al., 2018; Prochnow et al., 2020). One qualitative interview study found that young people naturally and organically developed close-knit communities of close friends, often in the form of private Instagram accounts, on which privacy was a priority and emotional disclosure was safe and commonplace (Gibson & Trnka, 2020). These findings support the positive, adaptive, and strategic use of social media for typical adolescents.

Benefits of Social Media for Adolescents with Depression and Anxiety

Adolescents with depression and anxiety use social media differently than their mentally well peers (Radovic et al., 2017). Thus, the commonly cited associations between social media use, depression, and anxiety may be explained in part by the unique offerings of social media for depressed and anxious youth. Youth have described feeling motivated to share their depression online because it is perceived as easier than sharing in-person, and because they are hoping to connect with others who understand and have had similar experiences (Carey et al., 2018; Rideout et al., 2018). A systematic narrative review of 28 studies on online help-seeking among adolescents found that adolescents commonly cited anonymity, ease of access, and sense of community as driving motivators to find mental health support online (Pretorius et al., 2019). Thus, youth who are already experiencing depression and anxiety may find particular mental health benefits by going online.

Benefits of Social Media for Marginalized Adolescents

Social media may be particularly beneficial to marginalized adolescents, for whom it may not be safe, feasible, or appealing to find support in the offline world. This may include homeless youth, as well as racial, sexual, and gender minority youth. From a multicultural perspective, the possibility of finding support for mental illness while remaining anonymous may help adolescents to overcome shame and stigma, imposed by the dominant culture, around help-seeking. Further, in the wide world of online support, adolescents may be more likely to find support that is tailored to their cultural values and worldview.

A scoping review of 19 studies on individuals experiencing homelessness and their social media use found that for homeless youth, seeking help online minimized barriers and prejudices often encountered in-person (Calvo & Carbonell, 2019). Perhaps for the same reason, sexual and gender minority youth are significantly more likely than straight and cisgender youth to go online for information about depression and anxiety (Marengo et al., 2018; Rideout et al., 2018). Transgender youth have affirmed that social media is a place to garner emotional, informational, and “appraisal” support, or the validation in seeing their same experience reflected in others (Selkie et al., 2020). In sum, both qualitative and quantitative research studies support that homeless youth, as well as sexual and gender minority youth, use social media to find affirming communities and avoid discrimination (Craig et al., 2015; Escobar-Viera et al., 2020; Jenzen, 2017).

Another area of study has focused on experiences of racial minority youth. Perhaps due to a lack of culturally competent healthcare providers offline, black youth are more likely to go online to share their health stories (Rideout et al., 2018). One study interviewed 25 racially and economically diverse undergraduate students to understand the empowering and disempowering aspects of social media (Brough et al., 2020). Interviewees noted that social media allowed them to find and connect with similar others (e.g., by using the #blackLGBTQ hashtag), as well as to represent their voice both by sharing their own stories and observing as others share theirs. However, the same youth noted that social media can have the opposite effect, encouraging conformity to the dominant culture and exposing them to lifestyles that were not relatable (Brough et al., 2020). Thus, while social media may have unique affordances for marginalized youth, its potential to “other” its end users could also worsen mental health symptoms.

With the exception of the studies mentioned above, there is less support for the differential use of social media by racial or ethnic minorities. None of the recent systematic reviews on the relationship between social media use and internalizing symptoms mention race, although two call for more diverse samples (Odgers & Jensen, 2020; Orben, 2020). This suggests that there is little conclusive evidence on the differential use of social media by race, as well as any differences in associated mental health outcomes, whether positive or negative. Given that certain racial and ethnic groups may have fewer opportunities for culturally competent in-person mental health care and support, it is important to understand how they have built alternative spaces online.

Future Research Directions

After describing the literature to date, including studies that examine the relationships between depression and social media, problematic behaviors and experiences on social media, variables that may affect the relationship between social media and mental health, as well as the ways in which social media may alleviate symptoms of depression and anxiety, it is time to consider future research directions. The content above has noted gaps in the current understanding of these topics and exciting opportunities for future research in this area.

From the evidence surrounding depression and social media, we conclude with four critical considerations to move the research forward. These include improved assessments, advanced and nuanced analysis approaches, interpreting results with regard to their practical significance, and improving transparency in linking findings to conclusions. Further, we recommend that future studies incorporate measurements and hypotheses to address potential positive and negative associations between social media use, depression, and anxiety.

First, for improved assessments, many studies of depression do not use validated measurements for depression, leading to findings with limited clinical implications. Further, assessing technology use has most often focused on self-reported quantity of use, leading to biased and inaccurate assessments. Knowing that the vast majority of youth carry smartphones in their pockets, and often use devices passively (for example, walking while listening to music) and other devices simultaneously (for example, performing schoolwork on one’s laptop while using a smartphone as a calculator), accurately reporting the time spent on technology is next to impossible and not always meaningful. Improving technology assessments may involve further considerations of quality of use, such as through understanding emotional investment in use, importance placed on use, and the extent to which device use displaces other activities. Further, because offline activities are limited (either due to availability of resources or, the reality of offline discrimination, or recently, by the global COVD-19 pandemic), technology use may not be a marker of risk so much as a necessary path for education, entertainment, support, and connection.

Second, for advanced and nuanced analytic approaches, many previous studies have used population-level analyses such as linear or logistic regression across single populations. Future studies should consider more nuanced analysis approaches, such as quadratic analysis or latent class analysis to identify differences within groups. This approach would allow for detection and appreciation of individual differences that shape interactions with technology (Orben, 2020).

Understanding of practical significance, represented by statistical effect sizes, is also important, as many studies of media identify small effect sizes that are unlikely to drive clinical illness states. Putting these results into context is critical to help readers understand what behaviors are necessary to modify, and what behaviors lose practical significance in the context of an adolescent’s whole health.

Finally, we recommend that researchers evaluating social media, depression, and anxiety consider hypotheses that incorporate the potential for both positive and negative health effects, especially within at-risk subgroups. One study using this approach found that 46% of adolescent participants indicated that social media had a positive effect on their mood, while 41% reported neither a positive nor negative effect, and only 6% reported a negative effect (Wright et al., 2020). Measuring diverse uses and motivations for use, alongside validated measures of depression and anxiety, would allow for fuller consideration of social media’s effects on a study population, subgroup or individual. Specifically, the social media use among racial minority youth is underexplored. Thus, research should aim to understand the effects of social media on mental health within subgroups and individuals, especially individuals who are frequent targets of discrimination.

Clinical and Intervention Resources

Resources to promote healthy social media use may benefit both clinicians working with adolescents, and interventionists seeking new approaches to test.

There are several key tools and concepts that can be considered toward these goals:

1. 1. The American Academy of Pediatrics policy statement, “Media Use among School-aged Children and Adolescents,” recommended that parents establish media use rules to promote safe and healthy media use (Moreno et al., 2016). The policy statement proposed that families create a Family Media Use Plan to select and engage with media use rules. This plan is available online and includes a Media Use Plan in which families can select family rules and expectations around media use. It also includes a Media Time Calculator that allows teens to plan and consider how they spend their time during a given day, including time for media use.

2. 2. Healthy Internet Use Model. The Healthy Internet Use Model focuses on three key concepts: balance, boundaries, and communication (Moreno, 2013).

   • Balance: The balance between online and offline time is a critical concept to discuss with youth. Spending time offline, including hanging out with friends, exercising, or spending time outside, is critical to adolescent development. Further, achieving balance provides protection against concerns such as problematic technology use.

   • Boundaries: Boundaries refers to setting limits around what youth are willing to display about themselves online or on social media, as well as setting limits in where adolescents spend their time online. Discussing guidelines on what types of personal information are not appropriate to post on social media sites with teens can help prevent them from several online safety risks. These risks include being targets of bullying, unwanted solicitation, or embarrassment.

   • Communication: Just as with many tenets of adolescent health, parents should discuss social media and technology with their adolescents early and often. Establishing home rules for social media and technology use as soon as the child begins using these tools is an important way to promote healthy technology use from the beginning.

Further, adolescents should be advised that social media can promote mental health but can also make it worse. Social media can negatively affect health when it displaces other health-promoting activities, like sleep and physical activity. However, social media use that falls within normative ranges should not be the focus of modification. Rather, adolescent patients should be encouraged to pursue those aspects of social media use that research suggests promote mental health, while reducing or eliminating social media use associated with depression and anxiety.

10 The Role of Digital Media in Adolescents’ Body Image and Disordered Eating

Savannah R. Roberts , Anne J. Maheux , Brianna A. Ladd , and Sophia Choukas-Bradley

Social media is a normal part of life for adolescents in the United States. According to nationally representative data, the majority of adolescents (83%) use social media, and of those who do, 70% of teen girls and 56% of teen boys check it every day (Rideout & Robb, 2018). Research on social media has been rapidly increasing, as scholars attempt to understand how social media could both help and harm adolescents’ well-being. Prior research suggests that social media has an effect on users’ body image, with individuals simultaneously sharing images of themselves at their most attractive while experiencing preoccupation over how their appearance will be perceived by others. The effects of social media on body image may be heightened during adolescence, a developmental stage in which individuals often prioritize their physical attractiveness over other domains of self-worth. In this chapter, we first describe the developmental features of adolescence, and how they intersect with social media, with implications for body image and disordered eating. Next, we provide an introduction to relevant theoretical frameworks for considering social media’s effect on body image. Then, we examine how specific features of social media affect adolescents’ body image and disordered eating. Finally, we explore specific social media platforms and content devoted to body image concerns and disordered eating.

The Adolescent Developmental Period

Adolescence is a developmental period marked by substantive changes in interpersonal relationships, identity, and autonomy (Dahl et al., 2018). Biological, interpersonal, and sociocultural factors intersect to increase adolescents’ concerns about body image and physical appearance. These concerns may take the form of body dissatisfaction, when individuals dislike some element of their appearance, or disordered eating, when individuals engage in eating pathology in an attempt to modify their weight or shape. One key developmental feature of adolescence is the heightened focus on peer relationships (Brechwald & Prinstein, 2011). Increased sensitivity to social reward makes adolescents highly attuned to their peers (Kilford et al., 2016). Importantly, social status among peers is closely tied to appearance, as adolescents perceived to be the most attractive are often also the most popular (Kennedy, 1990; Lease et al., 2002). Concomitantly, adolescents experience increased self-focus and self-consciousness, including the imaginary audience – a sense that one’s peers are watching one’s every move (Elkind, 1967). When peer evaluation centers on appearance, the imaginary audience may increase adolescents’ body image disturbances.

Gender differences in sociocultural and biological factors produce differences in adolescents’ body dissatisfaction and disordered eating. Girls in particular are socialized to prioritize physical appearance (Daniels et al., 2020; Fredrickson & Roberts, 1997), and adolescent girls experience higher levels of body dissatisfaction and disordered eating than do boys (Neumark-Sztainer et al., 2006). Although ideal beauty standards differ by cultural context and by race/ethnicity, the average ideal body type for women in the USA is unattainably thin yet curvy, while the average ideal body type for men is muscular (Deighton-Smith & Bell, 2018; Edwards et al., 2016). Biologically, patterns of weight gain and fat distribution during adolescence bring girls on average further from the thin beauty ideal, while increased muscularity brings boys on average closer to the muscular beauty ideal. Many girls experience body dissatisfaction due to the perceived discrepancy between one’s body and the ideal feminine body, and may engage in disordered eating in an effort to reduce this discrepancy (Halliwell & Harvey, 2006). Adolescent boys may engage in muscle-building behaviors or excessive exercise in pursuit of the masculine ideal (Calzo et al., 2016). These developmental features considered together, adolescence is a period marked by increased risk for body image disturbances and disordered eating. Social media may increase the likelihood of these phenomena by allowing for social support and connection, while leaving adolescents vulnerable to exposure from negative social influences (Dahl et al., 2018). At a time when peer approval and status are of the utmost importance, social media allows for more frequent peer interactions, leading to increased appearance-related feedback (de Vries et al., 2016). Indeed, among adolescents, more frequent social media use is associated with higher investment in one’s appearance (de Vries et al., 2014). With increased frequency of appearance-related feedback and higher investment in this feedback, social media use may lead to increased body image concerns.

These concerns occur on a spectrum, ranging from low levels of body dissatisfaction to extreme preoccupation with weight and shape. Body dissatisfaction can be conceptualized as negative evaluations of one’s body, typically resulting from a discrepancy between one’s ideal and perceived appearance (Grogan, 2016). Body dissatisfaction has been identified as the most powerful predictor and risk factor for the development of disordered eating (Stice et al., 2011). Once disordered eating reaches the level at which it significantly impairs an individual’s physical health or daily functioning, that person may meet criteria for an eating disorder, such as anorexia nervosa (AN), bulimia nervosa (BN), or binge eating disorder (BED) (American Psychological Association, 2013). While few adolescents may receive a diagnosis of an eating disorder, the prevalence of body dissatisfaction and disordered eating is relatively common (Swanson et al., 2011). Across three large population-based studies, approximately 81% of adolescent girls and 63% of adolescent boys report body dissatisfaction (Kelly et al., 2018; Lawler & Nixon, 2011; Neumark-Sztainer et al., 2006). Further, population-based studies indicate that disordered eating is highly prevalent among adolescents, estimating that approximately 54–57% of adolescent girls and 30–33% of adolescent boys engage in at least one disordered eating behavior (Croll et al., 2002; Neumark-Sztainer et al., 2011). This chapter will focus primarily on how social media contributes to body dissatisfaction and disordered eating across the general adolescent population.

Relevant Theoretical Frameworks

There are a number of psychological theories relevant to understanding associations among social media use, body image disturbances, and disordered eating. While many of these theories were developed before the advent of social media, they nonetheless explore concepts that are implicated in social media use. The following section details leading theoretical frameworks for the development of body image disturbances and disordered eating, all of which have robust empirical support. Further, we explore a newly developed psychological theory, the transformation framework, which describes the ways in which social media has transformed adolescents’ lives and further increased the importance of physical appearance.

Objectification Theory

Objectification theory was proposed as a framework for explaining the psychological consequences women experience from growing up in a society that sexually objectifies the female body (Fredrickson & Roberts, 1997). It argues that women and girls in Western society learn to adopt an observer’s perspective of their own bodies – a process called self-objectification – after being exposed to frequent sexual objectification, which reinforces the societal message that a woman’s interpersonal value is based primarily on her physical appearance (Fredrickson & Roberts, 1997). Self-objectification is linked to body shame, depression, anxiety, and the development of disordered eating (Butkowski et al., 2019; Calogero et al., 2011; Erchull et al., 2013). Research now suggests that boys and men also experience self-objectification, as they are also exposed to sociocultural appearance pressures and may experience sexual objectification (Vandenbosch & Eggermont, 2013). The act of curating one’s social media profile can be thought of as a behavioral manifestation of self-objectification, as the user is specifically creating content about one’s identity that is meant to be consumed by others (Choukas-Bradley et al., 2019, 2020, 2021). In this way, social media users are encouraged to adopt an observer’s perspective of themselves and post social media content that will elicit positive feedback from their social media audience. Later in this chapter, we discuss specific behaviors and experiences on social media that are associated with self-objectification.

Social Comparison Theory

Adolescents who derive self-esteem from their physical attractiveness are likely to engage in social comparison, evaluating their attractiveness by comparing it to other social media users. Festinger’s (1954) social comparison theory argues that individuals engage in social comparison in order to estimate their own social status relative to others. While this is a natural process, it can be problematic in the case of physical attractiveness. Festinger’s seminal paper on social comparison theory (1954) argues that individuals have a tendency to make upward appearance comparisons when evaluating physical attractiveness (i.e., individuals tend to compare themselves to people they perceive as more attractive than themselves), resulting in worse body image. Furthermore, when engaging in social comparison, people try to compare themselves to similar others. Taken together, peers on social media may be perceived as realistic comparison targets, but by presenting highly edited images, these “similar” comparison targets may in fact serve as upward comparison reference groups depicting unattainable attractiveness. Regardless of whether adolescents compare themselves to individuals perceived to be more or less attractive, engaging in social comparison is associated with body dissatisfaction, especially among adolescent girls (Jones, 2001). Indeed, social appearance comparisons appear to be a primary mechanism through which social media exerts influence on body image disturbances and disordered eating during adolescence. Later in this chapter, we describe specific features of social media that encourage social comparison.

Tripartite Influence Model

A third theory relevant to understanding social media’s influence on adolescents’ body image and disordered eating is the tripartite influence model (Thompson et al., 1999), which was developed to explain the mechanisms through which body dissatisfaction originates. This model proposes that through peers, parents, and the media, adolescents are frequently exposed to unattainable standards of beauty. After encountering such exposure, adolescents internalize an unattainable appearance ideal and, like in social comparison theory, engage in appearance comparisons, processes known to lead to greater body dissatisfaction (Keery et al., 2004; Thompson et al., 1999). Internalization of an appearance ideal (often the “thin ideal” for adolescent girls and the “muscular ideal” for adolescent boys) refers to the extent to which an individual ascribes to culturally defined standards of beauty. Social media perpetuates these unattainable ideals and encourages social appearance comparisons through comments, images, and interactions that communicate societal expectations for adolescents’ bodies, ultimately fostering body dissatisfaction because these appearance ideals are unattainable for the majority of individuals (Thompson & Stice, 2001). Given its ubiquity, social media has become a primary source of appearance pressure in adolescents’ lives.

The Transformation Framework

The aforementioned theories were all developed before the advent of social media. However, scholars have recently identified features of social media directly implicated in the development of body image disturbances. The transformation framework argues that widespread adoption of social media among today’s adolescents has fundamentally changed the ways in which they are interacting with one another (Choukas-Bradley et al., 2021; Nesi et al., 2018a, 2018b). Here, we discuss three of the seven specific features of the transformation framework that are most relevant to understanding social media’s effects on body image: visualness, publicness, and quantifiability.

First, social media is characterized by visualness and publicness, meaning that users rely on photographs and videos to communicate to broad, public audiences (Nesi et al., 2018a). This reliance on visual forms of communication can make adolescents hyperaware of their own physical appearance. Currently, highly visual social media (HVSM) – such as Instagram, Snapchat, and TikTok – is the most popular type of social media among adolescents (Anderson & Jiang, 2018). When adolescents use HVSM, they increase their focus on others’ attractiveness, are exposed to unattainable beauty standards, and may engage in appearance-driven self-presentation techniques to elicit positive peer feedback in the form of “likes” or comments. These “likes” and comments represent quantifiability, or the Numerical indicators of popularity and attractiveness indicated by one’s peers and social media audience. We have provided a specific section later in the chapter describing how quantifiability of appearance-based feedback influences adolescents’ body image and disordered eating. Collectively, these features of social media may encourage self-objectification and social comparison.

Social Media Behaviors

The unique features of social media offer opportunities to engage in new, social media-specific behaviors, some of which have been linked to body image disturbances and disordered eating. The visual, public, and quantifiable aspects of social media contribute to a heightened focus on appearance and peer feedback. Below we discuss how specific behaviors on social media, including taking, posting, and editing “selfies,” and giving and receiving “likes” and comments on one’s content, may be implicated in adolescents’ body image and disordered eating.

Selfies

Social media offers adolescents the opportunity to take, edit, and post photos of themselves – “selfies” (Lim, 2016). Selfie behaviors, including taking and posting selfies, are relatively common among adolescents (Dhir et al., 2016; McLean et al., 2019), with nationally representative US data reporting that 45% of adolescents often or sometimes post selfies (Anderson & Jiang, 2018). The association between selfie behaviors and body image and disordered eating outcomes is not yet fully understood. Some evidence from adolescent girls and young adult women in China and Australia suggests an association between selfie posting and body dissatisfaction, overvaluation of shape and weight, greater internalization of the thin ideal (McLean et al., 2015), greater engagement in appearance comparisons (Mingoia et al., 2019), self-objectification (Meier & Gray, 2014; Zheng et al., 2019), and restrained eating (Niu et al., 2020). Among samples with both adolescent boys and girls, posting a selfie is associated with self-objectification (Meier & Gray, 2014), body shame (Salomon & Brown, 2019), and restrained eating (Wilksch et al., 2020). Interestingly, other research has found that disordered eating behaviors are associated with greater avoidance of posting selfies among adolescent boys and girls (Lonergan et al., 2020), and that Singaporean adolescent girls with greater body esteem are more likely to post selfies than those with lower body esteem (Chang et al., 2019). Some research with adolescent boys and girls in the USA (Nesi et al., 2021) and China (Wang et al., 2019) has found no association between selfie posting and body esteem. Experimental research with adolescent girls and adult women in lab settings shows that those assigned to take and post a selfie to social media report heightened anxiety, less confidence, and feeling less physically attractive afterwards (Mills et al., 2018). Notably, these outcomes were found whether participants were uploading an unedited selfie or had the opportunity to edit and choose a preferred selfie, highlighting that simply focusing on one’s appearance and posting it to a semi-public audience may help explain this association. The somewhat conflicting results suggest a need for more research in this area, particularly with mixed-gender samples.

Photo Editing

Social media allows for adolescents to manage their online self-presentation by editing and applying filters to photos before posting. Editing one’s photos and selfies, including applying filters, cropping, and modifying one’s appearance directly, is not uncommon among adolescents and is more common among girls than boys (see McLean et al., 2015). Qualitative work suggests that adolescent girls engage in “meticulous backstage planning,” spending hours planning and editing their photos to meet societal beauty norms, a practice that many consider “necessary” to be “pretty enough” online (Chua & Chang, 2016, p. 193). Editing one’s own photos may exacerbate the deleterious effects of social media by encouraging self-objectification, social comparison, and internalization of the thin ideal. Research with adolescents has shown that editing one’s photos is associated with self-objectification, which in turn is linked to appearance anxiety, body shame, negative appearance evaluation (Terán et al., 2020), and body image concerns (Wang et al., 2019). Photo-editing encourages social appearance comparisons (Mingoia et al., 2019) and disordered eating behaviors (Lonergan et al., 2020), even when controlling for time on social media and internalization of the thin ideal (McLean et al., 2015). Additionally, some qualitative work suggests that, especially for girls, the curation of one’s photos and selfies happens before the editing phase, including scrupulous photo planning and taking of multiple photos to ensure a desired outcome (Chua & Chang, 2016; Mascheroni et al., 2015), processes that some adolescents girls describe as “work” (Yau & Reich, 2019, p. 203).

Exposure to Others’ Photos

Emerging evidence suggests photo-based social media activity, rather than total time spent on social media, contributes to adolescents’ body image disturbances (Choukas-Bradley et al., 2020; Cohen et al., 2017; Marengo et al., 2018; Meier & Gray, 2014). HVSM in particular allows adolescents unprecedented opportunities to view the idealized and edited photos of their peers. Viewing others’ photos on social media is thought to engender risk for disordered eating and body dissatisfaction through internalization of cultural appearance ideals and social appearance comparisons (see Rodgers et al., 2020). Indeed, recent research with adolescents has shown that engaging in social appearance comparisons with others’ photos on social media is associated with body dissatisfaction (Chang et al., 2019) and disordered eating (Zimmer-Gembeck et al., 2020), and that monitoring peers’ attractiveness on social media is associated with internalization of cultural appearance ideals (Vandenbosch & Eggermont, 2015). Further, adolescent girls high in trait social comparison (those who engage in greater social comparison than their peers) may be especially vulnerable to the deleterious effects of viewing others’ photos on body image (Kleemans et al., 2018).

Peer Approval: “Likes” and Comments

Adolescents are also highly attuned to quantifiable metrics of peer approval in the form of “likes,” comments, friends, and followers. Neuroimaging studies have demonstrated greater activation in the brain’s reward circuitry (e.g., the nucleus accumbens) when adolescents view photos that receive high numbers of “likes,” especially when these were their own photos (Sherman et al., 2016; Sherman, Greenfield, et al., 2018; Sherman, Hernandez, et al., 2018), suggesting that quantifiable approval of one’s online self-presentation may be especially rewarding. Among adolescent girls in Australia, number of friends on social media has been shown to positively correlate with body image concerns (Tiggemann & Slater, 2013) and dieting (Tiggemann & Slater, 2014).

Peer approval can also be conveyed through comments on adolescents’ posts. As expected, longitudinal evidence suggests that social media use generally is associated with more appearance-related peer feedback (i.e., comments) on adolescents’ social media posts, though the same study found that the reception of peer appearance-related feedback is unrelated to body dissatisfaction (de Vries et al., 2016). Interestingly, positive appearance-related comments (compliments) have been implicated in adolescent girls’ self-objectification, possibly more so than negative comments or “teasing” (Slater & Tiggemann, 2015). However, negative appearance-related comments may be linked to adolescent girls’ lower self-esteem and depression and to boys’ tendency to act out (Berne et al., 2014). Some work with young adults suggests that the link between social media use and social comparison may be exacerbated by adolescents’ viewing “likes” and comments on others’ posts (Fardouly et al., 2017; Fox & Vendemia, 2016), to which they ostensibly compare their own peer feedback. Some longitudinal work also suggests that more liking and commenting on others’ social media content is associated with decreased appearance self-esteem across development (Steinsbekk et al., 2021).

No prior work to our knowledge has examined the experience of receiving or giving likes on adolescents’ disordered eating outcomes specifically, though research with adult women has demonstrated that Facebook use is implicated in the maintenance of disordered eating by providing reinforcement of shape and weight concerns (Mabe et al., 2014). Theoretically, if adolescents receive “likes” and comments on photos that have been edited, or promote an idealized version of their appearance, adolescents may infer that they receive positive feedback for altering their appearance, reinforcing their body dissatisfaction. For adolescents who engage in disordered eating, these “likes” and comments may provide reinforcement for disordered eating behaviors, though this should be studied directly in future research. Indeed, social reinforcement plays a role in adolescents’ disordered eating behaviors, and research has demonstrated that adolescent girls in particular encourage dieting and disordered eating among one another, and that girls who engage in disordered eating are more likely to be perceived as popular by their peers, despite having lower body esteem (Lieberman et al., 2001).

Subjective Social Media Experiences

Although social media-specific behaviors clearly play a role in adolescents’ body image and disordered eating, researchers are increasingly turning toward subjective, psychological experiences on social media to explain individual differences in these outcomes. Indeed, investment, or the degree of importance adolescents place on social media experiences, has been more strongly linked to negative outcomes than merely engaging in the behavior. Below we describe the roles of investment in one’s appearance online, investment in peer feedback on one’s posts, and heightened appearance-related social media consciousness (ASMC).

Investment in Appearance

Likely due to the sociocultural emphasis on appearance, aspects of adolescent development, features of social media, and, for girls, gender socialization, adolescents are highly invested in how they present themselves online. Although girls report generally placing more importance on appearing attractive online, boys report investment in their online appearance as well (e.g., de Vries et al., 2014; Mingoia et al., 2019; Yau & Reich, 2019). Investment in one’s selfies, including putting in more effort to take and edit selfies, is associated with greater body dissatisfaction and dietary restraint, even after controlling for overall social media use and internalization of the thin ideal, among adolescent girls in Australia (McLean et al., 2015), and with greater appearance comparisons among adolescent girls and boys in Australia (Mingoia et al., 2019). Work with young adult women is more extensive and finds a similar pattern (e.g., Cohen et al., 2018; Lonergan et al., 2019). Importantly, photo editing and investment in photos are highly correlated among adolescents and young adults (Cohen et al., 2018; Mingoia et al., 2019; McLean et al., 2015), suggesting that photo editing may be a behavioral manifestation of appearance investment.

Investment in Peer Feedback

Adolescents are also often highly invested in receiving peer feedback on their social media posts in the form of “likes,” followers, friends, and comments. Qualitative work suggests that adolescents, especially girls, post selfies for the primary purpose of appearing attractive or favorable to peers and ultimately receiving positive peer feedback (Burnette et al., 2017; Chua & Chang, 2016; Yau & Reich, 2019). Research also shows that adolescents and young adults engage in various behaviors to earn more “likes” on their content, including editing their photos, uploading photos at certain times of day, deleting photos when they do not get enough likes and reposting at a later time, purchasing followers and likes, asking their friends to like their photos, and liking others’ photos in exchange for more likes (Dumas et al., 2017; Yau & Reich, 2019). Among adolescents, this behavior is associated with negative mental and behavioral health outcomes (Nesi & Prinstein, 2019), problematic social media use (e.g., using social media to cope with negative emotions; Martinez-Pecino & Garcia-Gavilán, 2019), and lower global self-esteem (Meeus et al., 2019). Some preliminary work suggests that concern about peer feedback on one’s selfies specifically is associated with worse body esteem (Nesi et al., 2021). Among young adult women, greater investment in selfie feedback from peers was associated with body surveillance, body dissatisfaction, and drive for thinness, but not bulimic tendencies (Butkowski et al., 2019). Notably, young adults who engage in negative feedback seeking (i.e., eliciting negative feedback to confirm negative perceptions of oneself) and who receive more negative comments on Facebook are more likely to report disordered eating concerns and behaviors a month later (Hummel & Smith, 2015).

Appearance-Related Social Media Consciousness

The visual nature of social media that leads to a focus on physical appearance, such as HVSM, may manifest as a broader preoccupation with one’s social media self-presentation, even in offline spaces. Appearance-related social media consciousness (ASMC) has been proposed as a novel subjective experience among adolescents and adults, defined as a preoccupation with one’s attractiveness to a real or potential social media audience (Choukas-Bradley et al., 2019, 2020). ASMC is common among both adolescents and young adults, especially among young women (Choukas-Bradley et al., 2019, 2020). In some ways, this experience reflects the extension of self-objectification to a social media audience, whereby adolescents and young adults imagine how their social media photos look to outside observers, overvalue their physical appearance on social media relative to other social media experiences, and even remain vigilant during in-person social interactions with the knowledge that at any moment a photo could be taken and posted to a larger social media audience (Choukas-Bradley et al., 2019, 2020). ASMC is correlated with self-objectification, body surveillance, body shame, body comparison, depressive symptoms, and disordered eating among adolescents (Choukas-Bradley et al., 2020). Additionally, even when controlling for body surveillance (a behavioral manifestation of self-objectification) and overall time on social media, ASMC is associated with greater disordered eating behaviors for adolescent girls (Choukas-Bradley et al., 2020), suggesting the unique effect of social media-specific appearance cognitions.

Social Media Devoted to Body Image Concerns

The prior sections demonstrate how social media plays a role in adolescents’ body image and disordered eating. Adolescents may also turn to social media for guidance or inspiration on attaining their desired body type. In addition to universal features such as edited photos, “likes,” and comments, social media includes content designed specifically for the purpose of encouraging users to attain a specific body shape or appearance, and influences users’ perceptions of body image.

Weight Loss and Fitness Social Media Content

“Thinspiration” and “fitspiration” refer to social media images meant to inspire viewers to be thin or fit, respectively. While social media users may believe this content teaches viewers healthy lifestyle and dieting techniques, it can be problematic if it encourages inaccurate, or even dangerous, health content (Carrotte et al., 2015). Thinspiration and fitspiration images frequently depict weight loss techniques or fitness regimens, though there is no guarantee that these messages come from certified health professionals. More likely, the images have been posted by celebrities, models, influencers, or peers. Moreover, teenage girls with preexisting body image concerns are especially likely to seek out this type of content, hoping to gain inspiration for changing their own weight or appearance (Carrotte et al., 2015). The presentation of these images on social media, where adolescents frequently see the personal life experiences of their peers, may make them appear more relatable and thus attainable, despite many negative outcomes related to viewing these images. Alarmingly, companies that manufacture “wellness” products such as FlatTummyShakes and FitTea hire popular social media influencers and celebrities to advertise their products, though these supplements contain appetite suppressants and laxatives, and thereby facilitate disordered eating (Auguste et al., 2019; Wong, 2018). Studies examining young adults’ exposure to and posting of such content consistently show associations with body dissatisfaction and disordered eating behaviors (e.g., Griffiths & Stefanovski, 2019; Holland & Tiggemann, 2017), often mediated by appearance comparisons (e.g., Tiggemann & Zaccardo, 2015). To our knowledge, only one study on the topic has included adolescents, finding that participants with a diagnosed eating disorder were more than twice as likely to view fitness-related social media content, and consumption of such content was highest among adolescent girls relative to boys and young adult women (Carrotte et al., 2015).

Social Media Content Encouraging Eating Disorders

Taken to the extreme, some social media content is dedicated to promoting and encouraging eating disorders. This content, often referred to “pro-ED” (pro-eating disorder), “pro-ana” (pro-anorexia nervosa), or “pro-mia” (pro-bulimia nervosa), facilitates community discussion by individuals with these disorders to maintain their disordered eating behaviors and cognitions. The majority of followers of pro-ED profiles are adolescent girls (Bert et al., 2016). Content includes images of emaciated figures to inspire extreme thinness, challenges and competitions for caloric restriction, techniques for avoiding treatment, and anti-recovery messages (Arseniev-Koehler et al., 2016; Bert et al., 2016; Ging & Garvey, 2018). Eating disorders may be acquired or exacerbated through social learning processes. Indeed, research on group treatments for adolescent eating disorders demonstrates that patients may bond over their weight loss goals, share tricks for preventing effective care, vomit together, or compete with one another for the most severe case presentation (McGilley, 2006; Vandereycken, 2011). Whereas in clinical contexts, trained clinicians are able to monitor and address these phenomena, such an opportunity is unavailable on social media. Adolescents who are most at risk for disordered eating and who are more easily impressionable may be especially at risk for valuing the potential social support these pro-ED platforms provide (Arseniev-Koehler et al., 2016). At a developmental stage when peer evaluation and feedback is of paramount importance, these platforms pose a dangerous threat for encouraging and exacerbating adolescent eating disorders.

Body Positivity Social Media Content

In response to the increased popularity of appearance-focused photo and video sharing on social media (Anderson & Jiang, 2018), there has been an emergence of body positive content that focuses on challenging the unrealistic beauty standards depicted on social media by reconceptualizing body acceptance. More specifically, the social movement known as “the body positivity movement” has developed on social media with the intention of increasing body acceptance through broad definitions of beauty and the depiction of a greater range of body types and appearances, along with limited photo editing and manipulation (Cohen, Irwin, et al., 2019; Lazuka et al., 2020; Tylka & Wood-Barcalow, 2015; Webb et al., 2017). Indeed, content analyses indicate that posts related to body positivity present varying constructs of beauty (Lazuka et al., 2020), and have gained popularity on mainstream online communities (see Rodgers et al., 2020). Research has begun to examine the potential benefits of exposure to this content, with recent experimental studies finding associations between young women’s exposure to body positive images and boosts in body satisfaction and body appreciation, when compared to viewing thin-ideal images (Cohen, Fardouly, et al., 2019; Williamson & Karazsia, 2018). Despite this promising evidence, there is debate regarding how body positive content may continue to place value on physical appearance and may increase shame for individuals who have lower body acceptance (see Cohen et al., 2020). Consistent with this critique, studies have found that despite women’s frequently encouraging responses to body positive images, such exposure is associated with higher levels of self-objectification and salience of physical appearance (e.g., describing the self through the lens of physical appearance rather than other attributes; Betz & Ramsey, 2017; Cohen, Fardouly, et al., 2019). The numerous negative outcomes associated with self-objectification (Fredrickson & Roberts, 1997) pose the possibility that body positive posts may have long-term negative impacts that need to be further investigated.

From a theoretical standpoint, there are also potential benefits of body positive social media. For example, the tripartite influence model (Thompson et al., 1999) offers another framework for evaluating the relationship between body positive social media content and body image concerns. Exposure to a more diverse range of bodies may lead to a decrease in the internalization of media’s unrealistic appearance ideals (i.e., the thin and muscular ideals), improving viewers’ body image. Additionally, it is possible that social media users’ engagement in social comparisons with more inclusive and realistic social media targets may positively affect body image outcomes. Since much is currently unknown regarding the impacts of body positive content, future research should investigate the short- and long-term benefits and consequences associated with exposure to body positive social media content, especially among adolescent girls, who may be distinctly vulnerable to these associated effects.

Future Directions, Implications, and Conclusions

Social media has transformed the lives of adolescents. Although research is mixed regarding the overall effect of social media on adolescents’ well-being, extant research suggests that the highly visual nature of social media may lead to body image concerns and disordered eating. It may be useful to assess the ways in which an adolescent is using social media, and whether it is causing disruption to their well-being or body image. Given the ubiquity of social media use among adolescents, it is imperative that scholars and mental health care providers consider the effect of social media on adolescents’ body image and disordered eating.

While many novel social media behaviors have been linked to body dissatisfaction and disordered eating, social media is constantly evolving. New behaviors and opportunities beyond posting and viewing others’ posts are rapidly becoming central for adolescent social media use. For example, many adolescents now have two Instagram accounts – one on which they post polished posts fit for a more public audience and another – a “finsta” or fake Instagram – where they post more private, personal topics and photos (McGregor & Li, 2019). Additionally, Snapchat, a social media site used by approximately 70% of adolescents (Anderson & Jiang, 2018), allows for ephemeral sending where, unlike on more permanent platforms, photos are seen by interaction partners but then immediately deleted (Bayer et al., 2016). With the increasing popularity of TikTok, video-based sites also require increased research attention. It is unclear how these sites may affect adolescents’ body image and disordered eating behaviors, and more research is needed to investigate the role of these novel behaviors.

An additional key priority for future research includes applying intersectionality theory (Crenshaw, 1989) when investigating the relationship between photo activity on social media and body image concerns. To this point, the majority of literature in this area focuses heavily on presumably heterosexual White cisgender girls, despite social media use being ubiquitous among all adolescents, regardless of gender, racial/ethnic identity, and sexual orientation. Future research should examine the unique intersection of marginalized identities across race/ethnicity, class, gender, and sexual identity to determine social media’s particular effects on specific populations, such as Black girls and young women.

Conclusion

This chapter describes theoretical and empirical work on adolescents’ social media use, body image, and disordered eating. Although a few examples highlight the potential benefits of social media for adolescents’ body image, the majority of work in this area underscores the role of social media in perpetuating and encouraging body dissatisfaction and disordered eating behaviors, particularly among adolescent girls, by overemphasizing physical attractiveness and body ideals on HVSM. Many of the social media behaviors and experiences described in this chapter are normative and thus insidious in potentially causing harm. Others, such as pro-ED sites, are more flagrant. Adolescents, their parents, and clinicians should be made aware of the potential detriments and dangers of these platforms. Future research should continue to investigate these processes and develop intervention, prevention, and dissemination strategies to foster adolescents’ healthy body image and eating behaviors across development.

11 Digital Media in Adolescent Health Risk and Externalizing Behaviors

Michaeline Jensen , Mariani Weinstein , Morgan T. Brown , and Jessica Navarro

Adolescent externalizing and health risk behaviors are some of the leading causes of morbidity and mortality among young people (Blum & Qureshi, 2011; Kann et al., 2018) and are of significant public health concern. Adolescence is a key period for understanding these types of behaviors, as they tend to emerge and peak in this stage (Claxton & van Dulmen, 2013; Krieger et al., 2018). Importantly, adolescence is not only a key risk corridor for risky and problem behaviors, but also for entry into new social and digital spaces; most social networking sites (and their regulators) set age 13 as the age at which youth can have their own accounts (Jargon, 2019). Co-construction theory (Subrahmanyam et al., 2006) asserts that adolescents create (and co-create) their online worlds and experiences to match developmental needs, and thus we should not be surprised that adolescents’ developmentally appropriate affinities for risk taking, boundary testing, and affiliation would all manifest in some form in digital spaces, and that digital activities and offline behaviors would be mutually influential.

How youth digital media use and externalizing/risk-taking behaviors intersect is somewhat more complicated. In many domains, adolescent rates of health risk behaviors (substance use, sexual risk taking, violence perpetration) are at their lowest levels in decades (Lewycka et al., 2018; Twenge & Park, 2017), which some have asserted may be related to the proliferation of digital media and displacement of time (previously spent engaging in risk behaviors) in favor of time online and new forms of leisure, entertainment, and relationship formation (Kraut et al., 1998). Others have posited that youth engagement in online communities allows for covert or hidden coordination or reinforcement of deviancy and rule breaking, and thus technology may be linked with increased problem behavior (Ehrenreich & Underwood, 2016). In fact, the associations are not always straightforward, and thus this chapter seeks to summarize and integrate the research findings that have been published to date on these mutual influences and the mechanisms that underlie them.

State of the Evidence on the Role of Digital Media Use in Externalizing Behaviors

Here, we consider the intersections of digital technologies and several domains of externalizing and health risk behaviors (including delinquency, aggression, sexual risk taking, and substance use). For each externalizing or risk-taking behavior, we will review the research around two key questions: 1) Does the quantity of engagement with digital media impact adolescents’ externalizing and health risk behaviors? 2) What is the role of adolescents’ qualitative experiences online in these behaviors?

Problem Behavior and Delinquency

Problem behavior is generally conceptualized to include rule breaking, delinquency, antisocial behavior, and other acts that go against societal norms. In the digital age, problem behavior can (and does) occur online, and thus here we attend both to online manifestations of problem behavior alongside the ways in which adolescent engagement with digital media is associated with offline delinquency. As with all the externalizing and health risk behavior outcomes included here, we first consider whether there are consistent associations between the quantity of adolescent digital media engagement (e.g., screen time) and their problem behaviors before turning our attention to the quality/nature of online experiences.

Quantity of Digital Media Use and Problem Behavior

Some recent studies have suggested that more frequent social media use is tied to more concurrent conduct problems and delinquency among both younger (Ohannessian & Vannucci, 2020) and older (Galica et al., 2017) adolescents. However, these cross-sectional associations have not entirely held up in longitudinal research, as seen in a recent study where time online was linked to later internalizing symptoms and to comorbid internalizing and externalizing symptoms, but not externalizing symptoms in the absence of internalizing (where externalizing was measured as a combination of inattention, impulsivity, and antisocial behavior; Riehm et al., 2020). Similarly, our own research suggests that social media use and phone ownership in early adolescence are not associated with later conduct problems (once baseline conduct problems are accounted for) and that days on which young adolescents use more technology for a variety of purposes do not tend to be days when they report a greater likelihood of conduct problems (Jensen et al., 2019). However, some longitudinal associations have been found: Research with Korean adolescents suggests that technology use for entertainment is related with later online and offline delinquency, and internet use for communication is related to later offline delinquency (though internet use for information seeking seems to protect against offline delinquency; Lim et al., 2019). Other studies have investigated the opposite direction of effects (that earlier conduct problems might increase later social media engagement), which has been supported from adolescence (delinquency) into young adulthood (social media use; Galica et al., 2017) but not from childhood (behavior problems) into adolescence (screen time; Männikkö et al. 2020). Taken together, the displacement hypothesis is not strongly supported by the literature (i.e., there is little evidence that those youth who are online most are getting into less trouble) and there is considerable inconsistency in findings around whether digital media engagement might be linked with higher problem behaviors over time. More experimental, longitudinal, and ecologically valid research is needed in this domain.

Overlap between Online and Offline Delinquency

Online delinquent and problem behavior can take many forms. A commonly used typology classifies cybercrime and cyberdeviance into four types: cybertresspass (e.g., malware), cyberpornography, cyberviolence (e.g., cyberbullying, trolling, flaming), and cyberdeception and theft (e.g., digital piracy; Graham & Smith, 2019; Wall, 2001). For instance, some youth trespass into off-limits online spaces in ways that could have severe criminal penalties (e.g., cracking into bank accounts) whereas others trespass in ways that are less likely to be prosecuted but nonetheless problematic (e.g., hacking into a peer’s social media account). The prevalence of these (usually covert) behaviors among teenagers is understudied and hard to ascertain, but surveys from the security industry suggest that up to 40% of youth have hacked into a social media account, email, or bank account (primarily “for fun” and “out of curiosity;” Richet, 2013).

In reality, the line between online and offline spaces in delinquency is a blurry one. Indeed, emerging evidence suggests that long-standing types of offline delinquency now also manifest online, and the two contexts are not entirely separable. For example, qualitative interviews with ex-gang members and violence-prevention workers have revealed the existence of so-called digitalist gangs (Whittaker et al., 2020) who use social media as a tool for attention for themselves and their gang. These gangs are more likely to be newer and less established (compared to less digitally connected “traditionalist” gangs), and to engage in activities like boasting, taunting, and posting videos of violent confrontations online. These types of online posts can serve to spark very real offline violence, as seen in the so-called Twitter feuds covered by the popular press (Patton et al., 2013). In a recent study of Black youth involved in gangs in Chicago, 11% of posts included a picture of a gun, although not all these pictures were necessarily shared with aggressive intent (Patton et al., 2019). Further, research suggests that gang members are more likely than nongang members to engage online in piracy, harassment, threats, and the facilitation of drug sales, assault, theft, and robbery (Pyrooz et al., 2015), suggesting considerable overlap between online and offline crime.

Youth who engage in delinquent behavior in both online and offline formats may be at particular risk. A recent study found that those adolescents (ages 12–17) who committed both online and offline delinquency were the most likely to experience increased risk factors and fewer protective factors, whereas the online delinquency only group had fewer risk and more protective factors and the offline delinquency only group fell in between the two (Rokven et al., 2018). In a rare longitudinal study, Korean youth who engaged in cyber-delinquency were more likely to report more engagement in later offline delinquency (Nam, 2020), which may suggest that, at least for some, online delinquency may serve as a gateway to later offline (and potentially higher consequence) crime.

Online Depictions of Offline Delinquency

In addition to delinquent acts performed online, social media can be used to portray delinquent acts performed offline. A study of undergraduate students revealed that exposure to online depictions of delinquency (including abusing an intimate partner, illegally carrying a weapon, physical fighting, selling drugs, driving while under the influence, setting fire to property, stealing, and vandalism) was frequent, with 81% of students being exposed to at least one offending behavior online (McCuddy & Vogel, 2015). Furthermore, those students who viewed more delinquency in their online social networks were more likely to engage in delinquent behaviors themselves (though this was a much stronger association in smaller social networks). Unfortunately, the cross-sectional nature of this study does not allow us to ascertain the direction of effects (i.e., whether youth who engage in delinquent behaviors are more likely to affiliate with other youth who do so and post about it online, or whether exposure to online depictions of delinquency may shift youth norms and behaviors).

In an innovative program of research, the Blackberry project (Underwood et al., 2012) has followed a sample of students (and their text messages) over the course of high school. Qualitative coding of real, naturalistic text message data has revealed that most of these teens engaged in at least some antisocial text messaging, and that this text messaging about antisocial activities was associated with increases in multiple reporters’ accounts of rule-breaking behavior (Ehrenreich et al., 2014). Furthermore, findings suggest that the reason for associations between peer network delinquent texting topics and youth externalizing problems might be better characterized as selection (externalizing adolescents choosing deviant peer groups) rather than socialization (deviant peer groups driving externalizing behavior; Ehrenreich et al., 2019).

Aggression, Bullying, and Violence

Here, we consider how digital media use may relate to both physical and social/relational forms of aggression (the latter of which is particularly relevant online; Archer & Coyne, 2005). Indeed, aggression online can take a number of forms, including online bullying, harassment, and discrimination. Prevalence estimates vary widely and range from 1.0% to 61.1% of youth experiencing cyber-victimization and 3.0% to 39.0% of youth engaging in cyber-perpetration of aggression, suggesting that social media is a prominent context for cyberbullying (Brochado et al., 2017; Kowalski et al., 2019; Thomas et al., 2015).

Research suggests that many of the social roles that serve to instigate and sustain traditional/offline bullying also can be seen online. Sterner and Felmlee (2019) identified distinct roles of Perpetrator, Reinforcer, Victim, Defender, Bystander, and Informer around cyberbullying on Twitter. Reinforcers and defenders tended to enact these roles by commenting or by liking posts of the perpetrator or victim respectively, whereas informers tended to alert a site administrator to the cyberbullying incident. Interestingly, there were an average of 12 people directly involved (in one of the above roles) in each case of aggression on Twitter, suggesting that some features of social media (e.g., its permanence; Nesi et al., 2018a, 2018b) may increase the reach of cyberbullying experiences beyond those typically seen in face-to-face bullying.

Quantity of Digital Media Use and Online and Offline Aggression

Some have asked whether level of engagement with digital media (e.g., time spent online) presents a risk factor for cyber and traditional aggression. In a recent meta-analysis, links between general social media use and offline violence-related behaviors could not be formally synthesized because only three studies were available; however, the available studies each show that youth who are using social media more frequently tend to report more concurrent violence-related behaviors (Vannucci et al., 2020). Some cross-sectional research has also suggested that adolescents who spent more time online were more likely to be cyberbullying perpetrators (Hinduja & Patchin, 2008), with those who spend particularly high and problematic levels of time online being at the most risk (Kircaburun et al., 2020) and those with particularly low levels of time being (understandably) at very low risk of cyber-perpetration (Zych et al., 2019). It may be that in the average range of technology use, time online and time on social media are not closely related to cyberbullying perpetration.

Overlap between Online and Offline Aggression

Youth who perpetrate bullying online appear to mostly be the same youth who perpetrate bullying offline (Fanti et al., 2012; Hinduja & Patchin, 2008; Olweus, 2012; Sourander et al., 2010) as confirmed by a meta-analysis that concluded that traditional bullying perpetration is among the strongest predictors of online bullying perpetration (Kowalski et al., 2014). It is common for cyberbullying perpetrators and victims to know one another in person – for example in 57% of the cyberbullying cases at a high school the victim reported that the perpetrator was a schoolmate (P. K. Smith et al., 2008). In a profile analysis, youth who engaged in cyberbullying tended to engage in all other types of bullying as well (relational, verbal, and physical offline bullying) and were at elevated risk for other externalizing behaviors (e.g., using substances and carrying weapons; Wang et al., 2012). A longitudinal analysis of the transactional associations between face-to-face bullying perpetration and cyberbullying perpetration found that higher levels of earlier offline bullying perpetration predicted increases in cyberbullying perpetration (controlling for previous cyberbullying perpetration), but cyberbullying perpetration did not predict increases in offline bullying perpetration (Espelage et al., 2012); this suggests that cyberbullying does not appear to be a first foray that grows into later offline bullying perpetration, but rather that offline bullying perpetration may come to extend to online environments.

Exposure to Online Violent Content and Offline Aggression

The impact of exposure to violent content in video games has been much talked of and controversial. Scholars have proposed that violent video games normalize aggression and can elicit and reward aggressive cognitions (e.g., hostile attributions), quick violent reactions, and aggressive fantasies (Gentile et al., 2014), though others have noted that selection effects are also likely at play (Breuer et al., 2015; Heiden et al., 2019). Early in the field’s history, a meta-analysis of early video game research concluded that evidence strongly supports exposure to violence in video games as a causal risk factor for increased aggressive behavior (Anderson et al., 2010), but this finding has not entirely held up over time, with more recent registered reports (e.g., Przybylski & Weinstein, 2019) and meta-analyses of high-quality longitudinal studies finding zero to tiny associations between violent video gaming and later violent behavior (Drummond et al., 2020). One domain that has not yet been extensively researched is that of the potential intersections between social aspects of online gaming and in-game aggression, which has gained growing attention with the advent of online multiplayer gaming (with live video, audio, and or/chat streams; Freeman, 2018). More information is needed on whether the synchronous and semi-anonymous online multiplayer gaming context may socialize and/or reinforce youth verbal (e.g., hate speech, insults) or even serious physical aggression (e.g., the phenomena of SWATting; Lamb, 2020) in ways not yet captured in the literature to date.

Sexual Risk Taking

In adolescence, high risk sexual behaviors include behaviors that increase risk of unintended pregnancy, HIV infection, and other STIs, including early age at first intercourse, multiple sexual partners, concurrent sexual partners, having one-night stands, using drugs or alcohol prior to having sexual intercourse, having sex in exchange for money, and lack of pregnancy prevention methods (Kann et al., 2018). Sex and sexual risk taking have always been salient in adolescence, and in the digital age they are increasingly also taking shape in online spaces.

Social media and platforms that allow private messages are prevalent among youth to develop and maintain their romantic relationships, with only a small minority of adolescents accessing formal dating apps (which are meant to be illegal for minors; Vandenbosch et al., 2016). About 8% of all teens have met a romantic partner online (Lenhart et al., 2015) and 30% of sexually experienced adolescents have met a sexual partner online, with those who met partners online more likely to engage in unprotected sex and with multiple concurrent sexual partners (Ybarra & Mitchell, 2016). In this domain, social media may also contribute to health, safety, and privacy risks. Youth are exposed to and engage with sexual content in media, including pornography and sexting, that may impact their offline sexual behavior. In addition, youth may engage in online sexual behaviors such as cybersex or coordinating encounters with potential partners (including strangers). People have been very concerned about the risk that children will be targeted by sexual predators online, but empirical research suggests that this is in actuality very rare (Ybarra & Mitchell, 2016).

Quantity of Digital Media Use and Sexual Risk Taking

In a recent meta-analysis, the average association (across 14 cross-sectional studies) between social media use and sexual risk taking was r = 0.21 (95% CI 0.15, 0.28), representing a small to medium significant association, with stronger associations for younger adolescents and very small associations for later adolescents (Vannucci et al., 2020). Three of these studies included in the meta-analysis captured online sexual acts, including frequency of sexy online presentation (Vandenbosch et al., 2016), frequency of risky sexual online self-presentation (Koutamanis et al., 2015), and frequency of sending sexts (Gregg et al., 2018) whereas the remaining 11 studies captured more traditional indicators of adolescent risky sexual behavior. It does, then, appear that social media use and sexual risk taking tend to co-occur, though the cross-sectional nature of all studies makes it impossible to parse the direction of effects.

Exposure to Online Sexual Content and Offline Sexual Risk Taking

Exposure to sexual content online (e.g., internet pornography) has been linked to offline sexual risk taking, though, as with much research reviewed in this chapter, a lack of longitudinal or experimental designs limits ability for causal inference. For instance, a meta-analysis of six cross-sectional studies revealed that exposure to sexually explicit websites was linked to higher odds of intercourse without a condom in two studies and was perhaps related to having ever had sexual intercourse and having had multiple partners, though significant statistical heterogeneity made meta-analysis difficult, and most studies were weakened by their limited accounting for important potential confounding variables (L. W. Smith et al., 2016). In a relevant experiment on social norms, young adults who were assigned to and viewed sexual content posted by “peers” in a lab-generated Facebook feed tended to estimate that more of their peers engaged in sex without a condom, and in turn expressed higher willingness to engage in this risky behavior themselves (relative to young adults assigned to view nonsexual content on the Facebook feed; S. D. Young & Jordan, 2013). This highlights the important role of descriptive norms in intentions around risky behaviors and is consistent with longitudinal research that shows that adolescents’ self-report of exposure to online sexual content is related to normative beliefs and, in turn, increased likelihood of intentions to engage in and actual sexual behavior (Bleakley et al., 2011).

Sexting, Cybersex and Offline Sexual Risk

Sexting refers to the exchange of sexually explicit text or images, usually via private messaging, in a way that need not be synchronous or reciprocal (Daneback et al., 2005). Cybersex is a related concept that can occur via computer (rather than just by text or private message) and encompasses synchronous sexual talk and/or behaviors with a partner over video, voice, or text chat and that often includes an element of sexual gratification through masturbation (Daneback et al., 2005; Judge & Saleh, 2013). Although sexting and cybersex share some features with other types of exposure online to sexual content (e.g., pornography), they are also distinct, as they are usually characterized as more interactive as opposed to one-sided consumption.

Sexting is prevalent in adolescence, with between a quarter to a half of teens reporting engaging in sexting to some extent (Baiden et al., 2020; Frankel et al., 2018; Maheux et al., 2020). Sexting can take many forms, with qualitative research with emerging adults revealing that sexting occurs in various relational contexts including casual sexual, dating and intimate relationships, and nonsexual peer contexts (Burkett, 2015). A study conducted in Belgium found high rates of textual and visual online sexual behavior (with consistently higher rates among boys than girls); about half of teens (55% of boys, 40.6% of girls) had attempted to sexually arouse their romantic partner via online communication, 20% of teens reported sending sexy pictures to a dating partner, and 7.6% of adolescents reported undressing in front of a webcam for a romantic partner (Beyens & Eggermont, 2014). A profile analysis of adolescent women revealed that they tended to follow one of four patterns with relation to online sexual behavior: abstinent, participating in multiple behaviors including risky behaviors, mostly seeking sexual content, and mostly receiving sexual contacts (Maas et al., 2018). Motivations for sexting include sexual arousal, humor, flirtation, and seeking reassurance about appearance. Sexting and cybersex are in some ways normative (and present little risk for negative outcomes like STI and unintended pregnancy) but can also carry their own risks, including receiving unwanted and unsolicited sexts, privacy violations, and feeling pressured to engage in sexting (Burkett, 2015).

Cross-sectional research seems to suggest that those youth who are more sexually active and (to a somewhat lesser extent) who engage in certain types of sexual risk behaviors are also more likely to be engaged in sexting (Frankel et al., 2018; Romo et al., 2017), with photo-based sexting being more strongly tied to offline sexual activity than text-based sexting (Houck et al., 2014). A meta-analysis of 8 studies that examined sexting risk for sexual and risky sexual behaviors concluded that those youth who sexted were significantly more likely to be sexually active, to have had multiple past year partners, and to have used alcohol or drugs before sex (L. W. Smith et al., 2016). A separate meta-analysis of 15 studies (14 cross-sectional) with a wider age span (including adolescents and young adults) found that youth who engage in sexting are moderately more likely to have lifetime and recent sexual experience, and slightly more likely to engage in unprotected sex and have more sexual partners (Kosenko et al., 2017). Rare longitudinal studies on this topic suggest that sexting may serve to increase risk for later offline sexual activity and risk taking. For instance, one study concluded that sexting is associated with later sexual activity but not with later risky sexual activity (sex without a condom, substance use before sex, and multiple sexual partners; Temple & Choi, 2014). Similarly, degree of engagement with chat rooms, dating websites, and erotic contact websites has been associated with later sexual activity in both sexually experienced and nonsexually experienced Belgian adolescents (Vandenbosch et al., 2016). Finally, a study of objectively coded text message content suggests that evidence of sexting at age 16 was associated with reporting an early sexual debut, having sexual intercourse, having multiple sex partners, and engaging in drug use in combination with sexual activity two years later (Brinkley et al., 2017). This is consistent with a profile analysis that suggested that youth who engaged in the riskiest behavior over time engaged in both online sexual risk behaviors (e.g., sexting or arranging a sexual encounter with someone met only online) and offline sexual risk behaviors (e.g., hooking up and unprotected sex; Baumgartner et al., 2012).

As with the other outcomes reviewed here, more longitudinal and experimental research is needed to ascertain what drives these associations: Are sexually active youth more likely to also express that sexuality in sexting? Does sexting serve as a gateway to later in-person sexual behaviors and risk taking? Are sexting, sexual activity, and sexual risk taking driven by other risk factors (e.g., disinhibition; Dir & Cyders, 2015)? Only well-designed empirical studies will tell.

Substance Misuse

Substance misuse is a major public health concern among adolescents, with implications for long-term mental and physical health (Grant & Dawson, 1998; Substance Abuse and Mental Health Services Administration, 2019). Here, we consider research at the intersection of technology and all classes of substance use (including alcohol, prescription and over-the-counter medicine, tobacco, marijuana, and other illicit drugs), though the existing literature (and thus too our review) focuses most closely on the most prevalent adolescent substance use type: alcohol use and misuse.

As with the other externalizing and health risk outcomes considered here, we will review studies on both the quantity of engagement with digital media (and its potential implications for adolescent substance misuse) and research on how adolescents engage around alcohol online. Unlike previously considered outcomes of problem behavior/delinquency, aggression, and sexual risk taking, substance use does not have an online analogue. Although teens can (and do) engage in online expression of sexual behavior and risk (e.g., sexting), delinquency (e.g., hacking and cracking), and aggression (e.g., cyberbullying), there is as of yet no way that adolescents can consume alcohol or other substances online. They do, however, post in both text and pictures (Moreno et al., 2015) about offline alcohol and drug consumption, view such posts from their friends, and use digital media to glorify, rehash, coordinate, and even lament drinking episodes online (D’Angelo et al., 2014; Hebden et al., 2015; Jensen et al., 2018). We will thus here consider whether engaging with digital media in these different ways is associated with riskier adolescent substance use outcomes. Although alcohol-related marketing does occur online, research suggests that most adolescent exposure to alcohol-related content online is noncommercial (posted by individuals in the social network; Cavazos-Rehg et al., 2015) and thus alcohol marketing is not considered here.

Quantity of Digital Media Use and Substance Use

On the whole, research does seem to suggest that those youth who are most engaged with digital media are at least somewhat more likely to misuse alcohol and other substances. This is captured in a recent meta-analysis that identified 14 cross-sectional studies of amount social media use and adolescent substance misuse, with an average pooled effect size of r = 0.19, in the small to moderate range (Vannucci et al., 2020). Individual study findings suggested that adolescents who are more engaged with social media are also more likely to report regular alcohol use and binge drinking, tobacco use, and marijuana use compared to those who are less digitally connected (Gommans et al., 2014; Kaufman et al., 2014; Ohannessian et al., 2017; Sampasa-Kanyinga & Chaput, 2016; Spilková et al., 2017). These associations also seem to persist in adolescents even once potential confounds of impulsivity, sensation seeking, peer relationships, and symptoms of depression are controlled for (Brunborg et al., 2017). One recent longitudinal study suggested that frequency of social media posting and “checking in” on social media was associated with greater likelihood of subsequent initiation of tobacco and cannabis use, though other types of digital media use (e.g., “chatting and shopping” and “reading news/articles and browsing photos) were less consistently linked to risk of subsequent tobacco and cannabis initiation (Kelleghan et al., 2020). Of note, some research has suggested that much of these observed associations may be due to exposure to alcohol-related content on social media, and that once this mediator is partialed out there is no unique association between digital media engagement and alcohol use (Erevik et al., 2017). We thus turn our attention next to the types of alcohol-related content posted and viewed on social media.

Alcohol- and Drug-Related Posting and Substance Use Behaviors

Adolescents post about substance use on social media in a myriad of ways and for various purposes. These can include text-based posts describing alcohol attitudes, intentions, and behaviors (that make up over half of youth alcohol-related posts) as well as image-based alcohol depictions (Moreno et al., 2015). For the most part, when images featuring alcohol or other substances are shared on social media, they tend to be posted by someone in the picture rather than others (Morgan et al., 2010). and alcohol depictions tend to be incidental images (e.g., a person holding a drink while a photo is taken) rather than the primary focus of the image (e.g., a picture of drinking games or a person visibly drunk; Hendriks et al., 2017). Among this sample of Dutch young people aged 12–30, alcohol posting among adolescents under age 18 (legal drinking age) was rare, but young adults endorsed mostly posting images that include alcohol for “entertainment” and choosing not to post alcohol-related images because they thought it was “stupid,” because they drank little, to reduce risk of a future employer seeing it, and because it was not consistent with their identities (Hendriks et al., 2017). A distinction between legality or illegality of behavior is also relevant for marijuana depictions on social media, which an even larger majority of youth see as inappropriate to post (Lauckner et al., 2019). Nonetheless, when adolescents post about substance use on social media, posts are usually positive in nature, pro-alcohol posts outnumber anti-alcohol posts by a factor of more than 10, and negative consequences of use (e.g., hangovers or embarrassment) are rarely depicted (Cavazos-Rehg et al., 2015; Moreno et al., 2010, 2013).

It is quite clear from the literature that adolescents who post more alcohol-related content on social media tend to drink more (Roberson et al., 2018; Stoddard et al., 2012; Westgate & Holliday, 2016). In a meta-analysis of 19 studies on alcohol-related social media use (that included posting, viewing, and liking others’ alcohol-related posts), alcohol-related social media use was moderately and significantly related to alcohol consumption and alcohol-related problems, with stronger associations emerging in cross-sectional and self-report (of alcohol-related social media use) studies compared to longitudinal and observational research (Curtis et al., 2018). Indeed, posting about alcohol is associated with self-reported drinking frequency, heavy drinking, drinking quantity, and likelihood of alcohol use disorder (Glassman, 2012; Marczinski et al., 2016; Moreno & Whitehill, 2014).

Although far less studied, there is also some evidence that similar linkages may be at play for other substances as well. For tobacco, adolescents who posted positive tobacco-related content on Twitter were more likely to report past month cigarette and any tobacco use relative to those who did not post about tobacco on Twitter (Unger et al., 2018), and although posting about tobacco use is much less common than alcohol use among Dutch emerging adults, cigarette-related social media posts are nonetheless associated with real-life cigarette use (Van Hoof et al., 2014). For marijuana, research in young adults suggests that they do indeed post cannabis-related images on Instagram (Cavazos-Rehg et al., 2016) and that posting marijuana-related content to social media is associated with more pro-marijuana attitudes and actual marijuana use among racial-ethnic minority college students from low-income areas; however, no such associations emerged for alcohol depictions, alcohol attitudes, and alcohol use, which may suggest that these associations are most relevant when a behavior is illegal or less normative (Lauckner et al., 2019). Recent research suggests that marijuana-related posting is not uncommon even in adolescence, however, which underscores the necessity of more research in this domain. For instance, in Washington (a state where cannabis is legal for recreational use among adults over the age of 21), nearly a third of adolescents reported sharing marijuana-related content on social media, with about 11–13% sharing images or videos of people smoking marijuana and 24% sharing marijuana-related memes (Willoughby et al., 2020).

Nearly all of the above research has examined the role of alcohol- and drug-related posting to public (e.g., Twitter) or semi-public (e.g., Facebook, Instagram) platforms, but much less research has attended to the role of private communications (e.g., private direct messaging and text messages). However, the research that has examined private messaging suggests it plays a key role. One study found that about a quarter of late adolescents (in the summer after 12th grade) reported discussing substance use on public social media, whereas nearly half report doing so via private digital channels (George et al., 2019). In our own work (Jensen et al., 2018) college students in the USA and Korea have reported that they prefer private text messages to public-facing social networking sites to facilitate alcohol involvement, and private text messaging was more related than public social media to frequency of alcohol use and heavy episodic drinking. We have also shown that counts of alcohol-related words in sent and received private text messages are associated with higher odds of same-day drinking (Jensen & Hussong, 2019). Longitudinal research suggests that these associations may be bidirectional, with those youth who had previously been using substances being more likely to evidence later public and private substance-related discussions, and public and private conversations predicting later increases in marijuana use (but not alcohol or tobacco use; George et al., 2019). Taken together, these findings highlight the importance of future research that attends to how private digital communication channels may be uniquely indicative of substance use risk.

Exposure to Others’ Alcohol- and Drug-Related Posts and Substance Use Behavior

In addition to adolescents’ own posting behaviors being associated with substance use and misuse, so too is there a sizable body of evidence to suggest that adolescents’ peers’ posts also have the potential to impact their behavior. The majority of studies seem to support the hypothesis that exposure to others’ substance use online is related to pro-substance attitudes and actual substance use behavior (Cabrera-Nguyen et al., 2016; Curtis et al., 2018; Pegg et al., 2018). Results from recent longitudinal designs are particularly informative. Even after controlling for developmental risk factors for initiation of alcohol use, exposure to peers’ alcohol-related social media content predicted an adolescent’s likelihood of drinking initiation one year later (Nesi et al., 2017). Similarly, adolescent exposure to alcohol-related social media content predicted alcohol consumption six months after exposure after accounting for both the adolescent’s and their peers’ drinking habits (Boyle et al., 2016). Some studies suggest that different types of exposures may be more influential and long-lasting: Adolescents who had more exposure to pictures (but not text) about friends partying or drinking in their social networks were more likely to increase or maintain their smoking levels over time (Huang, Unger, et al., 2014). This is consistent with findings that image-based alcohol-related content posted by college freshmen may be more related to substance use intentions down the road than purely text posts on social media (D’Angelo et al., 2014). Among young adults in Norway, disclosure of and exposure to alcohol-related content online was tied to later alcohol use, though the strength and consistency of these associations were reduced once relevant covariates were accounted for (Erevik et al., 2017).

An innovative experiment confirms this pattern: Litt and Stock (2011) created two Facebook profiles, one that portrayed alcohol use as normal and a control that displayed no alcohol; after viewing one of the two profiles participants were assessed on willingness to use alcohol and alcohol attitudes. Participants who viewed the alcohol normative profile had higher levels of willingness to use alcohol, more favorable images of alcohol users, more positive attitudes toward alcohol, and lower perceived vulnerability to the consequences of alcohol use, suggesting that exposure affects attitudes concerning alcohol. Results from Roberson and colleagues (2018) build on this idea – higher numbers of people who display drinking in an individual’s online network predict more pro-alcohol attitudes. Taken together, it does appear that exposure to substance use in adolescents’ online peer networks is associated with increased risk for substance use and misuse, and we thus turn next to potential explanatory mechanisms for this association.

Mechanisms

As seen above, largely separate literatures suggest that adolescent externalizing (aggression and delinquency) and health risk (substance use and sexual risk taking) behaviors intersect with digital media use in myriad ways, with more support for the importance of activities youth engage in online rather than just the amount of time they spend on screens in co-occurring with and potentially impacting their risky behaviors. Here, we consider several potential mechanisms for these observed associations (shared vulnerability, peer selection and socialization/influence, identity expression, and whether there are unique predictions to be gained) that largely apply across the spectrum of externalizing and health risk outcomes.

Shared Vulnerabilities

A long body of research suggests that externalizing and health risk behaviors (e.g., sexual risk taking, substance use, aggression, and problem behavior) frequently co-occur, and are likely driven by the same vulnerabilities (S. E. Young et al., 2009). So too we are beginning to find that youth who are engaged in online risky or externalizing behaviors are likely to be involved in other behaviors on the externalizing spectrum. For instance, we have seen that perpetrators of online bullying are more likely to engage in substance use and offline conduct behaviors (Sourander et al., 2010; Ybarra & Mitchell, 2004). We also see that sexting is related to nonsexual risk-taking behavior, with adolescents who engage in sexting having higher odds of tobacco and alcohol use (Kosenko et al., 2017).

One compelling explanation for this co-occurrence is that the same risk factors likely predispose youth to multiple types of (online and offline) externalizing spectrum and health risk behaviors. For instance, online antisocial behaviors are associated with many of the same risk factors for in-person antisocial behaviors (i.e., narcissism, exhibitionism, and exploitativeness; Carpenter, 2012). Online aggression and cyberbullying seem to be facilitated by long-known individual (e.g., low agreeableness, moral disengagement, hyperactivity), family (e.g., low parental monitoring), peer (e.g., deviant peer group), and community factors (e.g., low school safety; Espelage et al., 2012; Kowalski et al., 2014; Marín-López et al., 2020). Likewise, similar risks are associated with youth engagement in online and offline sexual behavior: sensation seeking, low levels of education, less parental monitoring, and less family cohesion (Baumgartner et al., 2012; Ševčíková et al., 2014). In particular, risk factors for externalizing problems that are developmentally salient in adolescence (like behavioral disinhibition and its sister concepts of impulsivity, sensation seeking, and low self-control; Steinberg, 2010) stand out as contributors to both offline and online behaviors. This pattern of shared risk across outcomes highlights the importance of accounting for relevant covariates in studies that seek to parse the nature of associations between digital media and externalizing and health risk behaviors and for ensuring that observed associations are meaningful and interpretable, and not just a result of a “third variable” problem.

In fact, some theorize that the online environment may be particularly well-suited for disinhibition. The online disinhibition effect theory posits that a confluence of factors that facilitate disinhibition are inherent in the online space (dissociative anonymity, invisibility, asynchronicity, solipsistic introjections, dissociative imagination, and minimization of authority; Suler, 2004). Although social media is increasingly dropping some of these features (e.g., synchronous dyadic or group conversations via video or voice chat are increasingly common), it still may be the case that the Internet provides some psychological distance from the impact of one’s actions and lowers the threshold to rash action to a lower point than what would be present in face-to-face interactions.

Peer Selection

One of the most potent predictors of youth risk taking and externalizing behavior is the peer context, whether that be digital or in traditional, face-to-face spaces (Chan et al., 2019; Leung et al., 2014). Adolescence lies at the nexus of susceptibility to peer influence, concern for social reward, and engagement with digital peer contexts. Some features of digital media and online social networks make them particularly powerful conduits for peer influence: This is articulated in Nesi, Choukas-Bradley, and Prinstein’s transformation framework (Nesi et al., 2018a, 2018b), which asserts that traditional peer relations constructs are transformed via the features of social media.

We know from decades of research that adolescents tend to be similar to their peers (homophily), with support for similarly minded peers choosing one another as friends (selection) as well as social influence by adolescents on their peers’ attitudes and behavior (socialization). The classic question of whether peer similarity is driven by selection or socialization (e.g., Kandel, 1978) is equally relevant in the digital age. That is, are the many associations seen here between peers’ online behaviors and adolescents’ own online and offline behaviors a result of selection (i.e., choosing people with shared interests and behaviors) or socialization (i.e., peer influence)? Although peer socialization processes are the most frequent intervention target for preventing externalizing and health risk behaviors (Henneberger et al., 2020), selection is often also at play, and it can be difficult to disentangle the two and their influences (Gallupe et al., 2019; Samek et al., 2016). Selection and socialization processes are often mutually influential, such that youth select into antisocial networks and then they reinforce each other over time (Brechwald & Prinstein, 2011). Modern statistical methods like social network analysis and stochastic actor-partner modeling have allowed for scholars to parse the two more finely than ever before, and in fact, selection has been shown to be a stronger explanation for peer similarity in substance use behaviors than socialization effects (Rebellon, 2012).

In some ways, digital media is well-suited to help us better understand homophily, as online communication and social networks leave behind digital traces of the selection and socialization processes that we suspect are at work. Ehrenreich and colleagues (2019) used adolescent text messages over the course of high school, which were coded for antisocial content, to delve deeper into this very question. They found that those youth who were engaging in more externalizing behaviors (a combination of aggression and rule breaking) at each grade were more likely to be exchanging antisocial text messages (about substance use and rule breaking) with a larger proportion of their peers in the subsequent grade (evidence of a selection effect), but the proportion of antisocial dyads did not predict next-grade externalizing (lack of support for a socialization effect). Interestingly, they did find some evidence of a socialization effect when they homed in specifically on the first year of high school, such that the proportion of peers exchanging antisocial texts in the 9th grade was associated with one’s own rule-breaking behaviors a year later. A study using social network analysis showed that both selection and socialization processes were relevant to adolescent substance use: Teens tended to select friends with similar social media use and substance use behaviors, but exposure to photos of substance use online also seemed to socialize adolescents’ later smoking behavior (Huang, Soto, et al., 2014).

Peer Socialization

Although studies of digital media and traditional peer interactions suggest that selection is likely more important than it is often given credit for, socialization is still relevant to understanding peer processes in externalizing behavior. Adolescent susceptibility to peer influence is evolutionarily driven (Ellis et al., 2012) and evident even in their neurobiology (e.g., Chein et al., 2011); adolescence is a period in which youth are keenly motivated for social affiliation (including romantic), and thus highly motivated to seek social approval. We review several forms of peer influence/socialization here.

Deviancy Training

Socialization takes many forms, and deviancy training is one mechanism of peer socialization (Dishion et al., 1996). The process often plays out with a youth discussing an antisocial topic, which is reinforced by the peer’s response (e.g., by laughter, encouragement, or more antisocial discussion; Piehler & Dishion, 2007). One of the central difficulties of studying deviancy training in youth is the difficulty of capturing their interactions as they play out, and thus a promising direction for future research is the time-linked analysis of deviancy training in naturalistic peer-to-peer interactions via digital media. Digital communication offers an unprecedented window of opportunity to observe and understand how youth communicate and reinforce one another in their real interactions. Evidence gleaned from the content of youth text messages suggests that those youth whose antisocial text messages are reinforced by peers’ positive responses are more likely to see increases in their problem behavior over time. A study of adolescents’ text message exchanges noted that antisocial comments in text are often met with laughter (e.g., “lol” and “haha”) from their conversational partners, which is similar to the deviancy training observed in past face-to-face observational research (Ehrenreich et al., 2014). Furthermore, these antisocial conversations were associated with increases in rule-breaking behavior a year later.

Some social networking sites include features that can serve to amplify the ability of peers to positively reinforce youth behavior. The Facebook Influence Model (Moreno et al., 2013) posits that peer influence is amplified within the online social networking environment, which in turn shapes downstream cognitions and behaviors around risk. Whereas the seminal studies on deviancy training in face-to-face interactions pinpointed communication features like laughing or encouragement as powerful (albeit minimal) reinforcers of deviant talk, Facebook and Instagram allow youth to send the same message with the click of a “like” or a “❤”. In fact, research suggests that the “like” is a powerful reinforcer (Sherman et al., 2016).

Social Norms

Selection and socialization processes on social media can alter perceptions of peer norms over time (David et al., 2006). Descriptive norms capture perceptions of how many of or how often peers engage in the relevant behavior (e.g., substance use, delinquency) and injunctive norms capture perceptions of how much peers approve of the behavior; both are strongly linked to adolescent behavior (Rimal & Real, 2005). Super Peer Theory (Strasburger et al., 2013) asserts that media can serve as a “super peer” in that it can expose teens to information that makes risk-taking behaviors seem normative, and that this normative influence will in turn cause youth to take risks themselves.

Research is generally supportive of the thesis that exposure to risky content online operates by reshaping youth perceptions of normativity. Qualitative studies with adolescents (Moreno et al., 2009) and college students (Moreno et al., 2012) tend to suggest that peers’ references to alcohol use on social media are indicative of their actual alcohol use behaviors offline, with younger youth perhaps being most susceptible to the impact of online depictions on normative beliefs. Our research suggests that the amount of “alcohol talk” in received (but not sent) text messages from college students’ entire text messaging network over the course of two weeks is associated with greater perceptions of peer descriptive and injunctive substance use norms, in addition to sent and received alcohol talk being tied to frequency of heavy episodic drinking (Jensen & Hussong, 2019). A longitudinal study of adolescents showed the exposure to sexual content in media increased youth perceptions of normative pressure (which captured both injunctive and descriptive norms), which in turn increased sexual activity intentions and behavior (Bleakley et al., 2011). This is highly consistent with experimental evidence that exposure to sexually suggestive photos impacts adolescents’ perception that more of their peers engage in sexual risk taking (S. D. Young & Jordan, 2013) and that college students who viewed a social networking site with alcohol-related content estimated that the average college student drinks more frequently than participants who did not view the alcohol-related content (Fournier et al., 2013).

Status

Adolescents have been known to engage in certain types of problem behaviors (e.g., carrying a weapon, substance use, physical aggression) in service of gaining the status that these behaviors confer (Dijkstra et al., 2010; Osgood et al., 2013; Rulison et al., 2013). Nesi and colleagues (2018b) assert that some features of social media (e.g., its publicness and widespread availability) may amplify youths’ quest for status through online spaces through selective self-presentation. Although there have been relatively few studies to date that explicitly test the role of status striving as a driver of youth externalizing and risk-taking behavior, some new research suggests that some adolescents are (and are known by peers for) engaging in “digital status seeking” behaviors (behaviors intended to increase “likes” and approval) online, and that these digital status seeking behaviors are longitudinally tied to later increases in substance use and sexual risk behavior (Nesi & Prinstein, 2019). Indeed, the Internet’s culture of “micro-celebrity” may facilitate the extent to which high-status “peers” can impact norms and exert influence (Marwick & boyd, 2011).

We are beginning to see the role of status in peer influence across the externalizing and risk-taking spectrum. For instance, partying is considered by many teens as a high-status activity, and attendance (and subsequent publishing online) of images and text about parties may boost status by association (Marwick & boyd, 2011; Nesi et al., 2018b). Students in a rural high school in the United States tended to drastically overestimate how many of their popular peers were sexting (and those who believed that popular peers had sexted were more likely to have sexted themselves than those who did not hold that perceived norm; Maheux et al., 2020). As reviewed earlier, digitalist gangs are also capitalizing on the attention and status that social media can afford (Whittaker et al., 2020). There is even some evidence that being a perpetrator of cyberbullying is predictive of increased peer status over time (Wegge et al., 2016).

Interestingly, youths’ search for status and desire to be perceived positively could also exert a “chilling effect” wherein adolescents may self-censor their real-life behaviors to avoid unfavorable exposure on social media (Marder et al., 2016). A mixed-methods study of the chilling effect revealed that teens do engage in impression management around depictions of substance use (e.g., hiding their drink/cigarette when they know a photo will be taken and likely end up online, presumably to avoid potential consequences if it is seen by a parent) but that they rarely alter their actual substance use behaviors (e.g., choosing not to drink or smoke at the party in the first place; Marder et al., 2016). Further research on impression management, status seeking, and behavior change will certainly better elucidate the nature of these associations in the years to come.

Unique Online Influences?

As reviewed here, online peer influence does seem to be a predictor of youth externalizing and health risk behaviors. An important question, though, is whether online peers exert unique influence, over and above that which would be expected (or is seen) from real-life, face-to-face peers (i.e., from school or neighborhood). Recent studies have tested this hypothesis, and overall, it seems that, although peers (in general) are still highly influential, there is significant overlap between online and offline networks, and online-only peer relationships seem to exert none to small effects. For instance, McCuddy (2021) sought to parse influence by adolescents’ peers who are known in person (and also sometimes online) from those peers who are uniquely known online (and not in person). They uncovered little evidence that online peers expose adolescents to new/unique support for delinquency (e.g., only 7% of those exposed to any general delinquency in a peer network saw this influence from online-only peers, whereas 64% were exposed to both online and offline peer delinquency). Rates were similar for violence (8% exposed only via online peers) and slightly higher for theft (17%) and substance use (21%). Exposure to online peer support for general delinquency and violence were not associated with adolescent problem behaviors in these domains, though online peers appeared slightly more influential for theft and substance use behaviors. In all cases, online peer influence was of lesser magnitude than traditional (face-to-face) peer influence. Another study has similarly failed to find support for unique influence by online-only friends on marijuana use (Negriff, 2019).

Identity

Adolescent online and offline experiences are increasingly interwoven and often indistinguishable into what Granic and colleagues (2020) call “hybrid realities” that are both important for the attainment of developmental tasks like identity development. The Media Practice Model asserts that adolescents choose to interact with media in ways that are most consistent with their identity (or what they aspire for their identity to be; Brown, 2000). We must consider, then, that adolescents’ online engagement in and depiction of risk-taking and externalizing behaviors (e.g., sexting, depictions of substance use, cyber-aggression) are best understood through the lens of identity development and intentional self-presentation.

This thesis is supported by evidence that adolescents engage in sexting and cybersex in ways that are consistent with sexual identity exploration and development (Eleuteri et al., 2017) and that depictions of alcohol use online are related to one’s identity as a “drinker” (Thompson & Romo, 2016; Westgate & Holliday, 2016). This is also consistent with research in college students that suggests that depictions of substance use in highly visible areas (i.e., a profile or cover photo, which may seem more tied to identity) are more strongly tied to alcohol use and binge drinking than depictions elsewhere on social media (e.g., in a status update or a photo post; Moreno et al., 2015).

Digital Media as a Tool in Reducing Externalizing and Health Risk Behavior

Although schools and community programs have traditionally been main avenues for health information and education, virtual spaces are also a growing venue for the delivery of educational information, interventions, and support related to externalizing and risk-taking behaviors. Particularly in 2020–2021, when most adolescents in the USA have been engaged in distance learning due to COVID-19 and many in-person intervention programs shuttered, the delivery of health information through social media is increasingly relevant. Social media platforms, text messaging, and web-based platforms offer three key affordances for the delivery of health information: accessibility, anonymity, and credibility. Adolescents often want answers to questions about risk-taking behavior in the moment (Selkie et al., 2011), and the temporal and spatial accessibility of information and support via social media offer youth this proximity and flexibility. Further, online spaces can offer the anonymity teens may need to seek out information related to the use of drugs or alcohol or sexual activity without worrying about their parents’ or peers’ reactions (Best et al., 2016). Social media also offers a degree of credibility to health information; adolescents can see who originally posted the information as well as those who have shared it, which may help them to determine the validity of the information (Dunn et al., 2018; Stevens et al., 2017). While existing research on the use of social media as a tool for health information is promising, further research is required, especially given the rapidly changing online mores of the adolescent population.

Health Information

Social media can be a powerful tool in disseminating public health information to adolescents, particularly given the omnipresence of social media in the lives of youth. Even before the advent of social media, the Internet was the primary source of health information for adolescents, especially those with few alternative accurate sources of information and for sensitive topics (Borzekowski et al., 2006; Gray et al., 2005). More recently, a number of qualitative studies with adolescents have confirmed that social media and text messaging are accessible and appealing sources of public health information (e.g., sexual health), though youth are also wary of potentially inaccurate or uncredible online sources (and have encountered barriers like inadvertently opening pornographic content; Selkie et al., 2011). In a study of African American and Latinx youth, Stevens et al. (2017) found that social media was an important source of sexual health information, and that participants felt social media was a more credible source than internet searches. Further, exposure to sexual health information on social media was significantly associated with reductions in sexual risk-taking behaviors (Stevens et al., 2017).

Delivery of Prevention Messaging

In addition to health information, social media can also be utilized to convey prevention messages to adolescents. Another qualitative study with US adolescents found that teens differentiate between social media platforms when engaging with drug prevention content and are highly conscious of how their peers might perceive their behavior (Dunn et al., 2018). Consequently, participants reported reading and liking prevention content, but were not likely to share it with their peers or create antidrug content themselves. Participants in this study recommended using short and humorous videos on platforms away from adult eyes, where teens might feel more comfortable, and the authors thus conclude that it is crucial to involve adolescents in creating effective prevention messaging on social media.

Numerous studies have found that internet-based interventions can reduce risk-taking behavior, albeit with small effects. Adolescent women who participated in a web-based drug prevention intervention were less likely to use drug and alcohol six months after the intervention than their peers in the control group. Further, participants in the intervention group also saw increases in understanding of normative beliefs and self-efficacy (Schwinn et al., 2010). A text-based intervention study of youth seen in the emergency department for drinking-related outcomes found that youth in the intervention group engaged in fewer binge-drinking episodes and drank fewer drinks per day than their peers in the control group at the three-months post-test (Suffoletto et al., 2014).

A 2014 systematic review of 11 intervention studies that examined social media and text messaging as a mechanism for sexual health education concluded that these mediums can increase knowledge of STI prevention and may reduce risky sexual behaviors (Jones et al., 2014). For example, a Facebook-based intervention saw small gains in condom use among adolescents in the intervention group at two months, though this difference diminished by the six-month follow-up (Bull et al., 2012).

Online Support

Although many studies have documented the benefits of online support groups (using a variety of modalities including social media, text messaging, and internet browser) for adolescents with health problems (e.g., cancer, asthma, type I diabetes), very few studies have analyzed the efficacy of online support groups as strategy to reduce adolescents’ externalizing and risk-taking behaviors (Selkie et al., 2011). We do know that adolescent participants report utilizing anonymous online chat rooms to discuss sensitive topics (e.g., drug and alcohol use), and that these anonymous interactions can yield feelings of emotional support (Gray et al., 2005).

Research with adults suggests that online support communities could also be a useful tool in mitigating risk-taking and externalizing behaviors in adolescents. Indeed, studies of adults suggest that web-based support through Adult Child of Alcoholic (ACoA) online support groups afford desired anonymity, accessibility, and support from any location or at any time of day (Haverfield & Theiss, 2014). Likewise, a 2020 study of adults in an online recovery group found that the social support offered through the online group interactions seemed to reduce social isolation and the risk of drug addiction alongside helping build “recovery capital” to aid in maintaining sobriety (Bliuc et al., 2020).

While further research with adolescent populations is needed to investigate the potential and efficacy of online support groups in mitigating risk-taking behaviors, we can likely assume that the affordances of online support (i.e., accessibility and anonymity) will also be prized by young people. The need for accessible and high-quality recovery and support services has never been as salient as it is today when most substance abuse recovery and mental health programs have been pushed online due to the COVID-19 pandemic.

Conclusions and Future Directions

Although research on digital media and adolescent externalizing and risk-taking behaviors is still in its infancy, we have already accumulated evidence of several fairly consistent patterns. Adolescents are dual citizens of both online and offline spaces, and as such their identities and risk profiles manifest in both spheres as well. We are increasingly seeing that the amount of time adolescents spend online seems to be less important than the ways in which they spend that time, which can provide a valuable window into adolescent behavior and risk. Our glimpses into that window thus far suggest that adolescent disclosures and self-presentation online largely overlap with their offline identities and behaviors; our next challenge will be to devise ways to harness this information to enhance the efficacy and reach of interventions targeting these risky behaviors. For example, digital indicators of risk may be useful in targeting of public health messaging, invitations to prevention programming, or even timing of interventions. We have also seen that peer influence is alive and well online, that it largely overlaps with and operates similarly to the offline peer influence processes we have long studied, and that online peers do not seem to be presenting much unique risk compared to the peer influences adolescents encounter in their schools and neighborhoods.

These insights and implications notwithstanding, we still have much to learn. The field requires longitudinal and experimental research that allows for causal inference; only armed with this strength of evidence will we truly be able to parse the direction of effects in observed associations between digital media engagement and externalizing risk. This causal inference will only be possible in well-designed studies that adequately account for shared risk factors (e.g., disinhibition) that may potentially confound associations. Similarly, we require studies that use representative samples from diverse populations that allow us to generalize findings beyond just specific subsets of youth. Understandably, much of the research to date has focused on late adolescents, emerging adults, and college students (populations that are more easily accessible and more amenable to research on sensitive topics like sex, drugs, and crime). The next wave of research, however, must make sure to assess the range of experiences across the full span of adolescence (10–24; Sawyer et al., 2018), with particular attention to how the experiences of early adolescents (who are more likely to be newer residents of the digital world) may differ from those of late adolescents and early adults (Vannucci et al., 2020). We must also ensure that our research speaks to the experiences of youth from diverse backgrounds and identities, with attention to unique ways in which different groups of youth may engage in both online and offline spaces. Finally, we require more research-informed recommendations for how prevention and intervention scientists can best harness adolescents’ deep attraction to and engagement with their online social networks in service of sustainable health behavior change.

As the digital world evolves, so too must our science. Researchers must be nimble to adapt their research questions and designs to the ever-changing digital landscape and adolescents’ shifting preferences, though it is worth noting that we likely stand to learn the most from studies that tap digital manifestations of well-supported, theoretically driven processes that are much more stable than the platforms on which we study them.

12 Problematic Digital Media Use and Addiction

Sarah E. Domoff , Aubrey L. Borgen , Bonny Rye , Gloria Rojas Barajas , and Katie Avery

Adolescents spend considerable amounts of time using digital media and social media. Although risks and benefits exist, clinicians, teachers, and parents have grown concerned about problematic use, or excessive use that interferes with adolescents’ health, well-being, and development. In this chapter, we explain the difference between problematic and typical media use; detail the measurement of problematic media use; review existing prevention and treatment approaches for problematic use; and provide recommendations for clinicians working with adolescents. As this research is still in its early stages, we conclude with directions for future research.

Problematic vs. Normative Digital Media Use

Historically, conceptualizations of pathological use of digital media have relied on other behavioral disorders, such as pathological gambling. Indeed, Dr. Kimberly Young pioneered early studies on internet addiction (e.g., Young, 1998a), forging the path for subsequent research on identifying how one’s use of digital/electronic communication and media may contribute to poor functioning and well-being. Adapting criteria from the DSM-IV-TR (American Psychiatric Association [APA], 2000)’s description of pathological gambling, Young created one of the first known measures of such problematic use: the Internet Addiction Scale (Young, 1998a). Since then, several measures using a similar paradigm have been developed targeting a range of electronic communication and digital media uses, ranging from pathological video game use (Gentile, 2009) to instant messaging addiction (Huang & Leung, 2009) to compulsive texting (Lister-Landman et al., 2017).

Across these measures, a constant is that pathological or problematic use is defined as excessively using digital media or internet/electronic communication to the point of dysfunction. In other words, similar to other “addictions” or “abuse,” frequency of use is not the defining or sole factor. It should be reiterated that how one uses digital or social media and the impact of such use on one’s functioning (e.g., in relationships, at work or school, with peers) delineates problematic versus normative use. Put in other terms, an adolescent may use social media very frequently and not have it negatively impact their life, whereas another adolescent may use social media to a lesser extent and it could have dire consequences for their well-being. Duration or amount of use may matter to a degree (i.e., of course, problematic social media use correlates with higher amounts of use); however, only considering duration of social media use misses the mark for capturing this idea. In this chapter, we discuss this conceptualization further, and explicate current research on assessing, preventing, and treating problematic social media use. We also highlight clinical practices carried out at the Problematic Media Assessment and Treatment Clinic (www.sarahdomoff.com) and other best practices for mental health clinicians seeking to more routinely assess and treat these concerns.

Internet Addiction, Social Media Addiction, and Other Problematic Digital Media Use

Prior to the release of the most recent edition of the DSM – the DSM-5 (APA, 2013), the majority of research on problematic use of digital media used internet addiction criteria (Young, 1998b; based on pathological gambling criteria from the DSM-IV-TR) to conceptualize dysregulated or “addictive” media use (Domoff, Borgen, et al., 2019). Currently, definitions of dysregulated (also termed “addictive” or “excessive”) digital media use draw from the DSM-5 criteria for internet gaming disorder (APA, 2013) or theories rooted in behavioral addiction (Domoff, Foley, & Ferkel, 2020). Since then, research has expanded the term “problematic” to encapsulate both one’s own dysregulated use and digital media use or internet/electronic communication that may harm individuals other than the user themself.

For example, Billieux et al. (2015) proposed a Pathway Model of Problematic Mobile Phone Use, which consists of pathways to three types of problematic mobile phone use: (1) addictive patterns of use (i.e., the primary focus of this chapter); (2) antisocial patterns of use (e.g., cyber-bullying or use in situations that would be deemed socially inappropriate); and (3) risky patterns of use (e.g., phone use while driving or in other situations where physical harm may ensue and unsafe sexting). Although the majority of the following sections will focus on dysregulated use, researchers and clinicians should be aware of these other components of social media interactions and excessive phone use. We elaborate further on antisocial and risky use of social media or digital devices in the clinical implications sections. Similarly, although online gaming is outside the scope of this chapter, it should be noted that many popular games are social in nature and involve multiple players (e.g., massively multiplayer online games). We refer readers to Gentile et al. (2017) for a review of internet gaming disorder and clinical implications for adolescents.

In addition to recent theoretical advances in defining problematic media use, there is a growing body of literature indicating that reward systems in the brain are activated when adolescents use digital media (e.g., gaming disorder; Wegmann & Brand, 2020) and social media (e.g., Nasser et al., 2020) – providing a compelling basis for concerns about their addiction potential. For example, Sherman et al. (2016) examined adolescents’ brain reactivity when viewing pseudo Instagram photos. They found that seeing photos with many “likes” was associated with reactivity of several regions of the brain, including those connected to reward processing (interestingly, these authors also found reward regions were activated when “liking” photos, as well, see Sherman et al., 2018). Although this area of research is still new, the initial evidence suggests that engaging with social media (and other types of digital media) are rewarding to adolescents.

Assessing, Preventing, and Treating Problematic Digital Media Use

Assessing Problematic Digital Media Use

There are several measures of various types of problematic digital media use with strong psychometric properties. Although most have been validated with adult samples, we review three that have been developed for adolescents and are specific to social media use. One measurement that has been used to assess problematic digital media use is the Bergen Social Media Addiction Scale (BSMAS), previously known as the Bergen Facebook Addiction Scale (Andreassen et al., 2012). This scale assesses how social media is used rather than the social media platform specifically (Lin et al., 2017) and social media use is assessed over the past year (Watson et al., 2020). The BSMAS is comprised of 18 items that assess 6 symptoms of addiction: salience, mood modification, withdrawal symptoms, tolerance, conflict, and relapse (Andreassen et al., 2012).

The BSMAS’ Cronbach’s alpha is 0.83 (Andreassen et al., 2012), suggesting strong internal consistency. Regarding convergent validity, this scale associated with the Addictive Tendencies Scale, the Facebook Attitudes Scale, and the Online Sociability Scale (Andreassen et al., 2012).

The Addictive Patterns of Use (APU) Scale is another reliable and valid measure that can be used to screen for smartphone addiction (Domoff, Foley, & Ferkel, 2020). The scale consists of nine items that ask adolescents to rate their frequency of symptoms of addictive phone use (Domoff, Foley, & Ferkel, 2020), based on criteria for internet gaming disorder from the DSM-5, adapted to smartphones. Items include “During the last year, how often have there been times when all you could think about was using your phone?” and “Have you experienced serious conflicts with family, friends, or partner because of your phone use?” (Domoff, Foley, & Ferkel, 2020). In addition to completing the nine items, adolescents are asked to list the features of their phone that they use the most, allowing researchers to identify the types of apps or smartphone functions that may be most problematic. Recently, additional research further supports the validity of APU, with this measure associating with media use (e.g., TV viewing frequency; Domoff, Sutherland, et al., 2020a) and other dysregulated behaviors (e.g., food addiction, dysregulated eating; Domoff, Sutherland, et al., 2020b).

Finally, the Social Media Disorder (SMD) Scale (van den Eijnden et al., 2016) similarly uses criteria of internet gaming disorder, but applied to social media use, to assess symptoms of dysregulated social media use. The developers recognize nine criteria to define disordered social media use within the adolescent population: preoccupation, tolerance, withdrawal, relapse, mood modification, external consequences, deception, displacement, and conflict (van den Eijnden et al., 2016). This scale is made up of 27 items, 3 items for each of the 9 criteria listed previously; a short version that consist of 9 items was also developed that selected the highest loading items on each of the 9 criteria (van den Eijnden et al., 2016). A cut-off score for disordered use was identified as endorsement of at least five of the nine criteria on the scale (van den Eijnden et al., 2016). Positive correlations between social media disorder symptoms on this scale and depressive mood, hyperactivity, and inattention have been demonstrated (van den Eijnden et al., 2016).

Rates of problematic social media use (or high scores on measures of “addictive” or disordered social media use) tend to fall around 7%, across 29 countries (Boer et al., 2020; consistent with gaming disorder rates, Gentile et al., 2017). That is, based on data from countries in Europe, the Middle East, and North America, approximately 7% of adolescent social media users experience impairment due to their use (Boer et al., 2020), such as trouble sleeping/poor quality sleep (e.g., Vernon et al., 2016) and poorer academic functioning (Domoff, Foley, & Ferkel, 2020). Given how recently problematic social media use measures were developed, there is limited research on whether this prevalence has changed over time. However, in terms of how the COVID-19 pandemic is impacting rates, early evidence suggests that burden caused by COVID-19 is associated with greater addictive social media use (Brailovskaia & Margraf, 2021) and some evidence that problematic social media use has increased in some samples from before to during the pandemic (among adolescents in Italy; Muzi et al., 2021). Future research should prioritize examining longitudinal trajectories of problematic social media use, particularly given drastic increases in media use during the COVID-19 pandemic.

Preventing Problematic Social Media Use

Due to the burgeoning interest in social media and smartphone use among adolescents, there has been a vast amount of research highlighting correlates of social media use overall. However, there has been limited research investigating correlates or contributors to problematic social media use. Many researchers have hypothesized that there is a relationship between problematic social media use and adverse mental health symptoms, with the most consistent research supporting links via disrupted sleep and shortened sleep duration (e.g., Vernon et al., 2016). There have also been various studies outlining demographic factors and social factors that are associated with dysregulated social media use. Across the studies described, it is critical to note that we focus on dysregulated use (often called problematic in subsequent research) and not amount of social media use. The research on duration or amount of social media use and various correlates is mixed and inconsistent (Odgers et al., 2020), and is too indiscriminate to adequately capture the scope of adolescents’ social media interactions. It is also important to note that, unless specified, most research is correlational and should not be inferred as causal.

Internalizing symptoms, such as depressive and anxiety symptoms, correlate with disordered social media use. Bányai et al. (2017) conducted a longitudinal study assessing how problematic social media use and depressive symptoms were related. It was found that both problematic social media use and depressive symptoms grew over a two-year span and that changes in problematic use correlated with changes in depressive symptoms (Bányai et al., 2017). Another study found direct associations between problematic social media use and depressive symptoms and indirect associations between problematic social media use and self-esteem (Kircaburun et al., 2019). It has also been found that those with a higher baseline of depressive symptoms showed a sharper incline in problematic use (Raudsepp & Kais, 2019).

Various demographic factors such as gender and age have shown differing associations with problematic social media use. Gender has been found to have an impact on how social media impacts adolescents. That is, for boys, anxiety was a predictor for higher social media use while for girls, problematic social media use associates with depression (Oberst et al., 2017). For adolescent girls, it is suggested that problematic social media use and depressive symptoms work in a cyclical fashion, whereas depressive symptoms exacerbate problematic social media use, which then further worsens depressive symptoms (Kuss & Griffiths, 2017). This suggests a possibility that adolescent girls with depressive symptoms may struggle to identify adequate coping techniques and instead use social media to ineffectively manage their symptoms (Gámez-Guadix, 2014). Another study found that younger adolescents and female adolescents had higher levels of problematic social media use (Kircaburun et al., 2019). Additionally, the type of social media behavior plays a role in how it impacts the social media user. There is a relationship between passive social media use (e.g., scrolling, low social interaction) and anxiety and depression symptoms, while active social media use (e.g., commenting, liking, communicating with peers) was related to lower symptoms of depression and anxiety in adolescents (Thorisdottir et al., 2019).

Several social factors have been shown to relate to problematic use in adolescents. Social norms and friends’ social media use frequency was directly associated with frequency of social media use, leading to an association with problematic use (Marino et al., 2020). Another study found that social connectedness and general belongingness were indirectly related to problematic social media use (Kircaburun et al., 2019). Fear of missing out and perceived academic competence predicted addiction to social media among high school students in one study (Tunc-Aksan & Akbay, 2019).

Regarding protective factors, self-esteem has been shown to be a moderator of problematic social media use and depression in adolescents (Wang et al., 2018). It has been proposed that adolescents with higher levels of self-esteem feel more confident in coping with adversity and are therefore less likely to have depression and subsequent problematic social media use (Wang et al., 2018). For girls who use Facebook actively and have perceived online social support have shown to benefit from social media use, and perceived online social support was found to have a negative association with adolescent girls’ depressed moods (Frison & Eggermont, 2016).

Treating Problematic Digital Media Use

Prevention Programs

Given the limited research on risk and protective factors of problematic social media use, it is not surprising that we could not identify any published, empirically supported problematic social media use prevention programs. However, the authors of this chapter have developed and have recently tested the Developing Healthy Social Media Practices (DHSMP) Prevention to address this gap. The DHSMP Prevention program was developed to promote healthy social media use and mitigate risks associated with social media use among youth in grades 6–8. DHSMP Prevention is a classroom-based prevention program, consisting of 6 classes, approximately 45 minutes per session. The program consists of providing adolescents with psycho-education on: (1) positive and negative effects of social media use (i.e., content of social media and user engagement); (2) the impact of various social media use practices on adolescent health and well-being (i.e., context of use); (3) how to critically evaluate content provided via social media (i.e., deciphering whether social media posts/shares are legitimate or “fake news”; (4) how to cope with cyber-bullying; (5) privacy and safety online; and (6) social gaming-specific risks and benefits (e.g., loot boxes and financial risks; app/game design principles to encourage longer game play; fostering positive interactions when gaming).

The DHSMP Prevention program has been piloted with approximately 160 6th graders in one public middle school in the Midwest. Acceptability and efficacy of this program indicate high acceptability based on student ratings, and increased skills in healthy social media use. Specifically, youth reported (on a scale from 1 to 5, with 5 being more confident/likely): feeling confident in their ability to recognize when social media use is harmful (M = 3.84, SD = 1.26); feeling more confident in identifying times and places when they shouldn’t be using social media (M = 4.18, SD = 0.96); being likely to reduce their use of social media around bedtime, mealtime, while talking with friends and family, during class, and while doing homework (M = 4.04, SD = 1.22); being likely to use the privacy tips they learned (M = 3.85, SD = 1.14); feeling more confident in recognizing what cyber-bullying is (M = 4.48, SD = 0.84); and a greater likelihood to use strategies to cope with being cyber-bullied (M = 3.58, SD = 1.26). Although not a randomized clinical trial (RCT), preliminary results suggest that a school-based psycho-education program on how to use social media in healthy ways may increase relevant skills in early adolescents. Currently this program is being tested in a nonrandomized trial to further establish its potential efficacy.

Treating Problematic Digital Media Use

Even though problematic digital media use is a significant issue among adolescents, there are no validated treatment options specific to social media use. Research in this area has focused on treating internet gaming disorder (IGD) or internet addiction (IA), with very few studies investigating the treatment of problematic social media use (Pluhar et al., 2019). However, the research about IGD and IA treatment provides a basis for future directions in helping adolescents improve their social media use.

Cognitive Behavioral Therapy: Many research studies investigating the treatment of IA have focused on methods influenced by cognitive behavioral therapy (CBT). One of these investigated treatments is CBT for IA (CBT-IA; Young, 2013). The first phase of CBT-IA focuses on the behavior of individuals with IA, particularly time management and engagement in offline activities. The second phase focuses on the cognitive aspects of IA, introducing participants to challenging and restructuring their maladaptive cognitions about internet use. Finally, the third phase of CBT-IA uses concepts of harm reduction therapy to address any other environmental or psychological problems that are associated with IA (Young, 2011). This treatment model has been tested in a sample of individuals meeting criteria for IA. Adult participants engaged in the 12-week treatment, and a significant majority (70%) were able to manage their symptoms 6 months after completing treatment (Young, 2013).

Using concepts of CBT-IA, a recent study investigated the effectiveness of a treatment model for social media addiction. This treatment model focused primarily on the cognitive aspects of social media addiction, using the methods of cognitive reconstruction, reminder cards, and diary techniques (Hou et al., 2019). College students with high scores on the BSMAS (Andreassen et al., 2017) engaged in a short-term intervention that took place over two weeks. Compared to a group that did not receive the intervention, those in the treatment group experienced decreases in symptoms related to social media addiction, increased self-esteem, and increased sleep quality (Hou et al., 2019). While this study included a small sample of college students, it provides a basis for future research of using CBT to treat problematic social media use.

Abstinence Treatments: As with other types of addictive or problematic behaviors, abstinence from social media has been proposed as a potential treatment option for problematic use. Research about abstinence from social media has mixed results: Some studies have found that withdrawing from Facebook for a week can benefit individual well-being (Tromholt, 2016), while other suggest that complete withdrawal from social media can result in negative effects on highly addicted individuals (Stieger & Lewetz, 2018). Using an ecological momentary intervention, researchers found that abstaining from social media for an entire week can result in frequent relapse and withdrawal symptoms such as craving, boredom, and increased social pressure to be on social media. Long-term abstinence of social media, especially among heavy users, may have just as many (or more) negative effects than positive effects.

However, integrating CBT components and short-term abstinence may result in a useful treatment for problematic social media use. Instead of instructing participants to take a week-long break from social media, researchers for one study instructed adults to take eight 2.5-hour breaks from social media over the course of two weeks (Zhou et al., 2020). As identified by these researchers, the main goal of abstinence is for the participant to begin engaging in substitution behaviors, which can just as easily be accomplished in short breaks from media. During the two-week intervention, participants also recorded their behaviors, feelings, and thoughts in daily records; the researchers included a control group that only completed these diaries, without participating in the abstinence process. Participants who engaged in both abstinence and daily records reported the largest increase in life satisfaction after the intervention. While this study still included a small sample (33 adults in the intervention group), this provides preliminary evidence for combining short-term abstinence and aspects of CBT in treating problematic social media use (Zhou et al., 2020).

Other Treatment Modalities: Additional research about treatment with adolescents indicates that group therapy and parent involvement may be particularly useful. Group therapy with other adolescents provides a form of offline social support that is beneficial to those experiencing IA (Kim, 2008). Meta-analyses of IA group therapy have provided support for this type of treatment, especially in groups of approximately 9–12 adolescents (Chun et al., 2017). In addition, parent training targeted at managing behavior associated with IA can be a helpful treatment component (Du et al, 2010). Both of these treatment modalities should be assessed in future research with adolescents experiencing problematic digital media use.

Clinical Implications

Because of the possible negative consequences of problematic digital media use, it is important that mental health care providers for adolescents are aware of risk factors and early indicators. The American Academy of Pediatrics has recommended that clinicians conduct routine screenings for problematic internet use and has also provided useful recommendations for how to go about initiating a screening routine (D’Angelo & Moreno, 2020). Three areas of competency are important for clinicians screening adolescents for problematic use: knowing risk factors, using a validated screening tool, and identifying when screening will occur. There are multiple factors that indicate an adolescent may be at risk for developing problematic digital media use, which include being male (Widyanto & Griffiths, 2006). Other studies have suggested that some mental health diagnoses can be risk factors for problematic use, most notably ADHD and depression (Pluhar et al., 2019). However, anxiety, sleep disorders, and autism spectrum disorder have also been found to be common diagnoses among adolescents with other types of problematic digital media use. When first meeting with a teen, other risk factors to keep in mind include: dependence on the Internet for relationships and managing mood, narcissistic traits, experiences of FOMO (fear of missing out), dissatisfaction with family relationships, or mental health issues among parents (D’Angelo & Moreno, 2020).

Once a clinician is aware of risk factors affecting their adolescent client, it is important to use a validated screening measure (see Domoff, Borgen, & Robinson, 2020 for additional screening questions for overall problematic digital media use). One of these screening measures is the Problematic Media Use Measure (PMUM; Domoff, Harrison, et al., 2019). The PMUM contains 27 items that were created based on criteria for IGD, and measure how media use is interfering with individual functioning. The PMUM is a parent-report measure that has been validated for use with children aged 4–11 years. Additionally, a short-form (PMUM-SF) has been validated with nine items. Both the original and PMUM-SF are helpful for screening young adolescents for problematic media use. Currently, a self-report version of the PMUM is being validated in the USA and internationally to facilitate screening of problematic media use in older adolescents. Additionally, the APU scale is useful for screening for problematic smartphone and social media use, specifically. Both the PMUM and APU are freely available for clinicians (request access via www.sarahdomoff.com).

Researchers at the University of Wisconsin have provided two screening instruments on their website: the Adolescents’ Digital Technology Interactions and Importance (ADTI) Scale and the Problematic and Risky Internet Use Screening Scale (PRIUSS). While the PRIUSS is meant to be used as a screener for adolescent problematic digital media use, it has primarily been validated among older adolescents and young adults, including samples of 18- to 25-year-olds (Jelenchick et al., 2015). The ADTI has been validated among a sample of 12- to 18-year-old adolescents (Moreno et al., 2020). Both of these screening instruments may be useful to clinicians in determining need for intervention services, and can be found at http://smahrtresearch.com/use-our-methods/. Additionally, a three-item PRIUSS has been validated (PRIUSS-3; Moreno et al., 2016).

After screening for problematic digital media use, we recommend administering narrow-band measures of media-specific problems, combined with a clinical interview. For example, at the Problematic Media Assessment and Treatment Clinic (www.sarahdomoff.com), we use the Video Game Addiction Scale (revised; Gentile, 2009) and the Social Media Disorder Scale (van den Eijnden et al., 2016) to further assess criteria for gaming disorder and problematic social media use, respectively. As mentioned, we also screen for other types of risky digital media use, including assessing the content that youth are exposed to, the individuals with whom youth interact online, the context of use (e.g., around bedtime, during other important activities), and parental management of adolescents’ digital media use. Although these implications are specific to screening and assessment in outpatient settings, mental health clinicians in the inpatient setting should review clinical recommendations outlined by Burke et al. (2020) for hospitalized youth and social media use in this setting.

Limitations and Future Research Directions

Measures and Consistency of Terminology

Assessing problematic digital media use has proven to be a difficult task because of the inconsistency in terminology and conceptualization of “problematic.” We argue that problematic should not be defined by amount of use – instead, clinicians should screen for dysregulated use (“addictive”), risky use (i.e., while driving, intimate/private interactions with unknown individuals), and antisocial use (cyber-bullying, trolling, etc.) routinely with each adolescent. An additional limitation is that screening tools, such as the APU (Domoff, Foley, & Ferkel, 2020) and PMUM (Domoff, Harrison, et al., 2019), do not yet have clinical cut-off scores, necessitating their validation in clinical samples to better identify youth at risk.

Research Design

As research in the area of problematic digital media use continues to grow, many limitations in this area of investigation have come evident. One of the primary limitations is accurate reporting of digital media use, particularly among adolescents. Research suggests that individuals of every age find it difficult to accurately report how much time they are spending using digital media each day (Ohme, 2020). While accurate reporting of screen time is important for research, it is even more important for researchers to measure how adolescents are using digital media and what daily activities the use is interfering with, as those are the primary concerns when determining problematic use.

To get around the limitations of adolescent self-report, some researchers are beginning to use technology to track technology use. Passive sensing technology in smartphones is gaining traction as a convenient way to measure adolescent behavioral patterns like app usage or interactions on social media, in addition to physical health indicators such as movement and sleep (see Trifan et al., 2019 for a review of passive sensing research). The first validated passive sensing app that measures adolescents’ mobile device use (e.g., type of app used, duration, timing of use) has recently been supported as feasible to use and acceptable to adolescents and their parents (Domoff et al., 2021). This app, eMoodie, has ecological momentary assessment (EMA) capacity and uses gamification principles to foster completion of surveys and EMA on adolescents’ mobile devices (see www.emoodie.com for more information). Using research designs that include objective, accurate measures of problematic digital media use will bring researchers closer to the goal of determining etiology and planning treatment.

Clinical Trials

Another area for improvement in this area of research is increased implementation of clinical trial studies. As the conceptualization and assessment of problematic digital media use expands, opportunities for clinical trial research will become more feasible. One of the few RCTs that has been conducted concerning treatment for problematic digital media use in adolescents was primarily aimed at internet addiction (Du et al., 2010). While the study provided evidence for using CBT to treat internet addiction in adolescents, they identified their limitation of only including participants without comorbid disorders. Anecdotally, problematic digital media use commonly occurs among adolescents who have been diagnosed with other mental health disorders. In order for clinical trials to be generalizable to clinic settings, samples should include adolescents who have comorbid disorders. Additionally, it is important that clinical trials include broader types of problematic digital media use, instead of only internet addiction. The lack of treatment options for these adolescents, in addition to the growing prevalence of problematic digital media use, indicate the need for increased clinical trial research.

Sample Demographics and Diversity

Research into problematic digital media use and internet/social media addiction is being propelled forward by the growing need for identification and resource development. This is most apparent within the growing population of youth who are native to the digital social networking world as well as among those learning to incorporate these new dimensions of their virtual selves into their social networking immigrant lifestyles (Prensky, 2001). Future investigation should seek to address the research limitations of clinical studies in order to maximize generalizability, while also parsing out what may be facilitating differential susceptibility for risks or rewards related to social media usage. In examining the limitations of samples, there is a need for validation of aforementioned screeners and assessment in non-WEIRD (Western, educated, industrialized, rich, and democratic) populations; further, problematic social media research has quite limited samples in terms of racial/ethnic diversity and across socioeconomic strata. Given that lower-income youth and racially/ethnically diverse youth have higher rates of digital media use (and may have different risks related to social media use; e.g., harassment, victimization), future research must address this major limitation of social media research.

13 The Effects of Digital Media and Media Multitasking on Attention Problems and Sleep

Susanne E. Baumgartner

With the rise of social and mobile media, not only has the amount of media use changed but also how and when adolescents use media. Almost half of US American adolescents claim that they are almost always online (Anderson & Jiang, 2018). Being constantly online also leads to new forms of media use, such as media multitasking. Media multitasking is commonly defined as using two types of media simultaneously, or using media while engaging in other non-media activities, such as using media while doing homework, during dinner, or during face-to-face conversations (Jeong & Hwang, 2012; van der Schuur et al., 2015). Media multitasking is highly prevalent, particularly among young people (Carrier et al., 2015).

The rise of digital media and media multitasking has led to concerns whether these forms of media use deteriorate adolescents’ attention. The main assumption is that if adolescents get used to using media wherever they are and whenever they want, they might have difficulties sustaining their attention, for example when doing their homework or when attending school (Ralph et al., 2015). Moreover, the constant use of digital media has been linked to sleep problems among adolescents (Hale et al., 2019). Since sleep is crucial for the healthy development of adolescents, including their attention and level of sleepiness in school, it is important to understand the ways in which digital media affects sleep. This chapter provides an overview of the current state of the field on the effects of digital media and media multitasking on attention and sleep.

Digital Media and Attention Problems: What Do We Know?

There is a long tradition in media effects research studying the effects of media on attention problems and ADHD-related behaviors. The focus was long on the effects of watching television or playing video games that have been the most popular forms of media use among adolescents in the past. For example, a meta-analysis from 2014 shows that there is indeed a small but significant association between the time children and adolescents spent watching TV and video games and ADHD-related behaviors (Nikkelen et al., 2014). This is further supported in a more recent review of the literature (Beyens et al., 2018). The effects of TV and video games on attention problems have been typically attributed to two main characteristics of these media types: their fast-paced and potentially violent content. It has been assumed that both of these characteristics might lead to higher arousal states to which adolescents potentially habituate (e.g., see Beyens et al., 2018). In the past decade, however, the media landscape and the types of media that are popular among adolescents have changed dramatically. This has resulted in a research shift away from the effects of traditional types of media (i.e., TV and video games) toward understanding the potential effects of social media and media multitasking on attention.

Media Multitasking and Attention

In 2009, Ophir, Nass, and Wagner published a seminal paper on differences in cognitive processing styles between heavy and light media multitaskers. Specifically, heavy media multitaskers were more easily distracted than light media multitaskers during a cognitive task they performed in the laboratory. It was the first study explicitly investigating the potential effects of media multitasking on cognitive processes. The authors interpreted their findings as an indication that people who multitask with media frequently have a completely different processing style than people who do this less frequently. Following this study, a plethora of studies have been conducted to understand the relationship between media multitasking and various aspects of attention (for reviews, see Uncapher & Wagner, 2018; van der Schuur et al., 2015). The literature can be differentiated into studies using self-report-based measures of attention in everyday life, and studies using cognitive tasks to measure the level of sustained attention in laboratory settings. It is, however, important to note that most of these studies focused on young adults (i.e., university students), and very few studies focused specifically on adolescents.

Studies using self-reports for attention problems in everyday life have consistently shown that adolescents who media multitask more frequently have more problems focusing their attention (for a review see van der Schuur et al., 2015). For example, media multitasking is positively related to increased attentional failures and mind wandering in young adults (i.e., undergraduate students; Ralph et al., 2013). Moreover, adolescents who media multitask more frequently have more attention problems and higher levels of impulsivity (Baumgartner et al., 2014, 2018). A recent meta-analysis supported these findings by showing that media multitasking and attention problems in everyday life are significantly positively related, with small to moderate effect sizes (Wiradhany & Koerts, 2019).

In contrast to the studies on everyday functioning, studies that tested differences in sustained attention with cognitive tasks in the laboratory show more mixed results. Whereas some find no differences between heavy and light media multitaskers on various tasks related to sustained attention or distractibility (e.g., Baumgartner et al., 2014; Ralph et al., 2015; Wiradhany et al., 2019), others find small effects (e.g., Cain & Mitroff, 2010; Madore et al., 2020; Moisala et al., 2016). Overall, the findings based on cognitive tasks are less consistent than those based on self-reports, and are more difficult to compare as different cognitive tasks are used across studies. Although the existing findings are rather mixed, a recent review of the literature concludes that for tasks measuring sustained attention, evidence points toward performance detriments for heavy media multitaskers in comparison to light media multitaskers (Uncapher & Wagner, 2018).

Despite rather mixed findings for performance differences in cognitive tasks, overall, the existing studies support the idea that adolescents who media multitask more frequently show more attention problems in their everyday lives. However, almost all of these studies are cross-sectional and therefore conclusions about the direction of the effect cannot be drawn. Notably, it is also possible that media multitasking does not lead to attention problems, but that adolescents who are more easily distracted in their everyday lives are more likely to engage in media multitasking. To date, only a few longitudinal studies exist that tried to establish the causal direction of these effects. One longitudinal study found that adolescents who used media more often during academic activities (such as while doing homework) reported increased difficulties in focusing their attention during academic activities over time (van der Schuur et al., 2015). Another study found effects of media multitasking on attention problems only among early adolescents (12–13 years old) but not among middle adolescents (Baumgartner et al., 2018). Thus, there is some but limited evidence for long-term effects of media multitasking on attention. In line with media effects theories, such as reinforcing spiral models (Slater, 2007), it has been proposed that the effects of media multitasking on attention problems might be reciprocal, with adolescents suffering from attention problems being more drawn to media multitasking, and media multitasking in the long run further exacerbating their attention problems (Baumgartner et al., 2018). However, more longitudinal research is needed to empirically test this proposition.

Social Media Use and Attention

Evidence for a relationship between social media use and attention is even more scarce. Only a few studies to date have specifically examined the relationship between the frequency of social media use and attention problems. These studies tentatively point toward a relationship between the use of social media and inattentiveness with adolescents using social media more frequently showing more signs of attention problems (Barry et al., 2017: Boer et al., 2020). The evidence for a relationship between attention problems and problematic or addictive social media use is more compelling. Several studies showed that adolescents who use social media in obsessive or problematic ways, also report more attention problems (e.g., Boer et al., 2020; Mérelle et al., 2017; Settanni et al., 2018; Yen et al., 2007). For example, one study found associations between problematic social media use and hyperactivity among a large sample of more than 20,000 Dutch adolescents (Mérelle et al., 2017), and another study found cross-sectional correlations between problematic social media use and attention deficits, impulsivity, and hyperactivity (Boer et al., 2020).

The question of causality across these studies is key. Does the use of social media deteriorate adolescents’ attention capacities or are those adolescents who have difficulties sustaining their attention more drawn to social media? Due to the scarcity of longitudinal studies in this realm this question cannot yet be conclusively answered. One longitudinal study investigating the reciprocal relationships between ADHD and social media use found no evidence for an effect of social media use frequency on ADHD over time but an effect of addictive social media use on ADHD (Boer et al., 2020). This indicates that not the frequency of use per se but more problematic usage patterns (such as uncontrollability of usage or displacement of social activities) might be detrimental to adolescents’ attention. Although this study found no evidence for attention problems being a predictor of developing problematic social media use patterns, another study found that ADHD symptoms in adolescents were the strongest predictor for developing internet addiction two years later (Ko et al., 2009).

Taken together, it seems likely that adolescents with attention problems are more drawn to social media in general, and that they are also more likely to show problematic usage patterns. The stimulating and arousing nature of digital media is particularly appealing to individuals showing symptoms of ADHD as they have a higher need for stimulation (Weiss et al., 2011). Digital media might provide the optimal level of stimulation to them. However, it is still unknown how far the (problematic) use of digital media further increases attention problems. The existing studies indicate that there is indeed a possibility that problematic usage patterns further deteriorate attention. However, due to the small amount of longitudinal studies, it is difficult to draw definite conclusions.

How Do Social Media and Media Multitasking Affect Attention?

To understand how social media and media multitasking affect attention problems among adolescents, it is important to identify theoretical explanations for such effects. Three potential explanations have been put forward to explain the potential effects of media multitasking on attention: 1) habituation to high arousal levels, 2) becoming increasingly sensitive to irrelevant information, and 3) deterioration of attentional control processes (see Baumgartner et al., 2018).

Similarly to the mechanism that was proposed for the effects of violent and fast-paced TV on attention, habituation to high arousal levels might also play a role in the effects of media multitasking and social media use on attention. Media multitasking is considered an arousing activity, and it has been shown that switching between media activities increases arousal levels (Yeykelis et al., 2014). Thus, it can be assumed that when adolescents engage frequently in media multitasking, they habituate to these rather high arousal levels. This in turn makes them favor stimulating and arousing activities in the future. That individuals can habituate to media stimuli has been previously shown for video games with gamers physiologically habituating to arousal levels after repeated video game play (Grizzard et al., 2015). In the context of media multitasking this could mean that adolescents habituate to the arousing nature of multitasking, and as a consequence find less stimulating single-task environments less appealing (e.g., sitting in class or listening to a lecture).

The second potential explanation is that media multitasking affects basic cognitive processes. Ever since Ophir et al. (2009) showed differences in cognitive processing among heavy and light media multitaskers, it has been suspected that engaging in media multitasking may cause these different processing patterns. Engaging in media multitasking requires individuals to attend to multiple streams of information. It has thus been argued that this type of information processing may train the brain to become more sensitive to irrelevant information (Ophir et al., 2009). If individuals get used to continuously attending to several streams of information, they might be more easily distracted by irrelevant external (and potentially internal) distractions (Adler & Benbunan-Fich, 2012).

The third mechanism that has been suggested is that by engaging in media multitasking, adolescents deteriorate their basic attentional control processes. This has been called the “deficit-producing hypothesis” (Ralph et al., 2013). The main assumption is that media multitasking might deteriorate adolescents’ ability to regulate their attention internally as they get used to external stimulations. A similar mechanism has previously been assumed for the effects of fast-paced TV content for which it was suggested that fast-paced content captures attention in a bottom-up fashion and does not train adolescents’ volitional attention processes (e.g., Lillard & Peterson, 2011). Thus, by engaging in media multitasking frequently, adolescents might not train their ability to guide their attention. This may lead to deficits in these attentional control processes over time (Rothbart & Posner, 2015).

Next to these three cognitive mechanisms, others have argued that digital media use may increase symptoms of ADHD among adolescents by replacing time spent with more developmentally beneficial activities (Weiss et al., 2011). Thus, even if digital media use has no direct effect on cognitive processes, it may still interfere with the healthy development of these skills because it replaces developmentally important activities, such as playing or having conversations with friends and family (Pea et al., 2012).

Importantly, although all of these mechanisms are theoretically plausible, empirical research assessing the mediating role of these mechanisms is still missing. Understanding the underlying mechanisms, however, is crucial as this will help to develop intervention studies that target the problematic aspects of digital media use rather than restricting digital media use in general.

Are There Any Positive Effects of Digital Media on Attention?

If digital media has the potential to affect attentional processes, the question is warranted whether digital media use may not also have positive effects on cognition and attention. Indeed, it has been argued that engagement in media multitasking may also train attentional processes (i.e., trained attention hypothesis: Kobayashi et al., 2020; van der Schuur et al., 2015). It has been assumed that people who engage frequently in media multitasking may improve their task switching skills and lower their switching costs by training these skills. Evidence for this trained attention hypothesis for media multitasking is scarce. However, one brain imaging study found some evidence for improved attentional brain activity among heavy media multitaskers (Kobayashi et al., 2020), and another study found better task switching performance among heavy media multitaskers (Alzahabi et al., 2013). Interestingly, it has been recently suggested that there are curvilinear relationships in that intermediate media multitaskers have better attentional control than low or heavy media multitaskers (Cardoso-Leite et al., 2016). More research is needed to establish whether such positive or curvilinear effects do indeed occur.

In contrast to the rather mixed findings on potential beneficial effects of media multitasking, research on the positive effects of playing action video games are more consistent. These studies show positive effects of playing action video games on several attentional skills, such as focused attention, selected attention, and sustained attention (for a recent meta-analysis, see Bediou et al., 2018, and for a review focusing specifically on attention, see C. S. Green & Bavelier, 2012). These effects were shown for cross-sectional studies but also for intervention studies that showed improvements in these cognitive skills after playing games for 20–40 hours. Most of these studies focused on young adults; however, a few also corroborated these effects for children and adolescents (Dye et al., 2009). Action video games pose a high demand on divided attention, information filtering, and motor control. It is therefore assumed that engaging in these games trains these attentional processes and can therefore benefit attentional control (e.g., Bediou et al., 2018).

In sum, there is some evidence that digital media has positive effects on attention skills. However, this highly depends on the content and type of media used. Particularly, first-person action video games seem to be beneficial. Moreover, effects are dependent on the amount of time spent with particular media. Extant literature suggests possible curvilinear relationships with moderate amounts of exposure being more beneficial than no exposure or too much exposure (Cardoso-Leite et al., 2016; Schmidt & Vandewater, 2008).

Future Research Directions for the Effects of Digital Media on Attention

Overall, research so far has found supporting evidence for a relationship between the amount of media multitasking and social media on the one hand and attention problems on the other hand. Adolescents who engage more frequently in media multitasking and who show more problematic social media use patterns, are also more likely to have attention problems in their everyday lives. The key endeavor for future research is to establish the causality of this relationship. It is yet unclear whether adolescents with attention problems are more drawn to engage in media multitasking, or whether media multitasking affects attention over time. Tentative evidence suggests a reciprocal relationship in that adolescents with attention problems are more drawn to specific types of media and media use patterns, and that spending too much time with these digital media further increases their attention problems (Baumgartner et al., 2018).

Next to the fundamental question of causality, it is crucial to understand the characteristics and affordances of digital media that lead to potential effects on attention. Which characteristics of social media and media multitasking impair attention, and how do these differ from other types of media? Understanding these characteristics is important for several reasons. First, this may help our understanding of the underlying mechanisms through which they are at work. Despite several theoretical assumptions about these mechanisms, empirical evidence is clearly lacking. Understanding these mechanisms might help adolescents to find more beneficial ways to use digital media without banning these completely from their lives. Moreover, a theoretical understanding of which characteristics are problematic would have crucial advantages in the current fast-changing media landscape. Currently, research lags behind new technological developments, and the same questions emerge with every new type of media. To create a more sustainable research agenda it would be helpful to understand the key characteristics of media that drive these effects, and compare and differentiate these among different media types (Orben, 2020).

Digital Media Use and Sleep: What Do We Know?

Sleep plays a critical role in the development of adolescents. Insufficient sleep has been linked to decreased cognitive functioning, increased risk of obesity, and diminished well-being, such as depressive symptoms and perceived stress (e.g., Shochat et al., 2014; Short et al., 2013). From late childhood to early adolescence sleep-related problems increase (Mitchell et al., 2020), with approximately 75% of students in their last year of high school getting insufficient sleep in comparison to only 16% of 6th graders (i.e., fewer than eight hours per night; National Sleep Foundation, 2006). Due to the importance of sleep for healthy psychological and physical development, it is concerning that so many adolescents today get insufficient sleep. Digital media are often seen as one of the main culprits for insufficient sleep and sleep problems, especially among adolescents (e.g., Bhat et al., 2018; Mireku et al., 2019). Particularly smartphones and social media are used extensively by adolescents, and frequently when already in bed or even during the night (e.g., Scott & Woods, 2019; van den Bulck, 2003, 2007).

There is consensus in the field that digital media use is linked to insufficient sleep in adolescents. Several reviews and meta-analyses support this notion (see, e.g., Carter et al., 2016; Hale et al., 2019; LeBourgeois et al., 2017). For example, a meta-analysis on the effects of mobile media devices on sleep, concluded – based on 20 studies with a total of more than 125,000 children and adolescents – that the use of media devices was consistently linked to insufficient sleep quantity, lower sleep quality, and increased daytime sleepiness (Carter et al., 2016). Similarly, in a more recent review of the literature, digital media use was related to adolescents going to bed later, needing more time to fall asleep, waking up during the night, showing signs of sleep problems, and daytime sleepiness (Hale et al., 2019). These effects have been shown for the general time that adolescents spent with media, but particularly for bedtime media use (Hale et al., 2019) and are consistent across various countries and cultural backgrounds (Hale et al., 2019).

Despite this strong evidence for cross-sectional relationships between digital media use and sleep, there are only a few longitudinal and experimental studies, and evidence from these studies is rather mixed. Some longitudinal studies found that digital media use was related to less sleep one or two years later (Johnson et al., 2004; Mazzer et al., 2018; Poulain et al., 2019). In contrast, others did not find longitudinal effects of media use on sleep (Tavernier & Willoughby, 2014), or only for specific subgroups (van der Schuur et al., 2018). For example, media multitasking was over time only related to increased sleep problems among girls but not among adolescent boys (van der Schuur et al., 2018).

To further establish the causality of the relationship, a few intervention studies exist that encouraged adolescents or young adults to reduce the use of specific media before bedtime to examine whether this improves sleep length and quality. These studies typically show improvements in sleep quality during intervention. For example, engaging in a smartphone app-based slow-breathing exercise improved subsequent sleep in comparison to using social media before going to bed (Laborde et al., 2019). Similarly, reducing adolescents’ screen time after 9pm on school nights was related to increased sleep duration and improved daytime vigilance (Perrault et al., 2019). A recent meta-analysis on 11 intervention studies concluded that interventions can be successful in reducing screen time and improving sleep time (on average by 11 minutes per day) among children and adolescents (Martin et al., 2020). These studies are promising as they show that reducing screen time can have beneficial effects on sleep. Longer intervention studies, however, are needed to further test the long-term effectiveness and willingness to comply among adolescent samples.

Why and How Do Digital Media Affect Sleep?

Three underlying mechanisms are typically put forward in the literature to explain the effects of digital media use on sleep (e.g., Bartel & Gradisar, 2017). First, the use of digital media before bedtime or when already in bed might displace sleep time. Second, the blue light emitted from digital devices might interfere with the secretion of the sleep hormone, melatonin. Third, the arousing content of digital media might make it difficult for adolescents to fall asleep after media use.

Sleep displacement may occur in two stages: it may lead adolescents to go to bed later and, once in bed, media use may delay the time when adolescents close their eyes and try to fall asleep (Exelmans & van den Bulck, 2017a). Evidence for sleep displacement is consistent for adolescent samples, and has been shown to occur for various types of digital media, such as smartphone, social media, video games, and TV (e.g., Hysing et al., 2015; Kubiszewski et al., 2013). Overall, the literature clearly points toward later bedtimes for adolescents who use digital devices in the evening. Delayed bedtimes and sleep times might be particularly problematic for adolescents who have strict school starting times and cannot easily sleep in. For adult samples, it has been shown that digital media use might lead to later bedtimes but in turn also to later rise times (Custers & van den Bulck, 2012).

Particularly for the use of smartphones, sleep displacement might also occur after sleep onset during the night, when incoming messages interrupt sleep. Several studies reported that smartphones lead to nighttime awakenings (Fobian et al., 2016; van den Bulck, 2003), and these nighttime awakenings might negatively influence sleeping patterns in the long run (Foerster et al., 2019). Therefore, adolescents who take their devices to bed might not only fall asleep later but might also be awakened by these devices during the night. Based on the existing literature, it is very likely that sleep displacement is a contributing factor for the detrimental impact of digital media on sleep. However, it is likely not the only factor because sleep displacement can only account for effects on sleep quantity but to a lesser account for the effects on sleep quality.

The bright screen light emitted by electronic devices has also been considered one of the main culprits for the effects of digital media on sleep. It has been argued that the artificial light emitted by electronic devices may lead to a disruption of the circadian rhythm, leading to increased alertness, and deteriorating sleep quality (Cho et al., 2015). When considering the effects of artificial light on sleep at least three factors need to be considered: the intensity of the emitted light, the duration of light exposure, and the type of light (Cho et al., 2015). Bright light is more disruptive for sleep, as well as short-wave and blue light. Electronic devices, such as smartphones, emit short-wave blue light that is said to suppress the production of the hormone melatonin, which plays an important role in making people sleepy and supporting healthy sleep.

Several studies found negative effects of screen light on subsequent sleepiness and sleep quality (Cajochen et al., 2011; Chang et al., 2015; A. Green et al., 2017). For example, exposure to a very bright LED-backlit computer screen affected melatonin levels and sleepiness of male adults (Cajochen et al., 2011). Similarly, negative effects of reading an e-reader before going to sleep were found (Chang et al., 2015). Importantly, the effects of screen light might be stronger for adolescents than for adults, as adolescents seem to be more affected by short-wave light than adults (Nagare et al., 2019).

Despite several studies finding effects of screen light on sleep quality, it is still highly debated in the field whether the light emitted from tablets, e-readers, TVs, and smartphones is bright enough to interfere with melatonin secretion and sleep. In a recent study, no or only very small and clinically insignificant effects of a bright tablet screen were found (Heath et al., 2014). Moreover, in those studies that found effects on melatonin secretion and/or sleep, sample sizes were rather small, and participants were exposed to rather extreme artificial light conditions, such as five hours of an extremely bright screen (Cajochen et al., 2011), or four hours of a bright e-reader screen (Chang et al., 2015). The clinical relevance of these findings is therefore still debatable. Overall, it is rather unlikely that the light emitted from digital devices is the only or even the most influential mechanism in explaining the effects of digital media on sleep.

The final mechanism that has been put forward is arousal. It is assumed that specific media content might lead to increased physiological arousal, which in turn makes it difficult for people to fall asleep after media use. This mechanism has received the least research attention, and a comprehensive theoretical conceptualization is missing. More specifically, we lack a clear conceptualization of which content characteristics lead to which effects on which mediator (e.g., physiological arousal, cognitive alertness). Bedtime media use might differ widely among adolescents, and from a media psychological perspective it is likely to assume that not all content is equally detrimental to all adolescents’ sleep. Although adolescents might use media for the same amount of time before going to bed, their usage patterns might differ tremendously, and their sleep might be differentially affected by their use. For example, one teenager might be listening to relaxing music on their smartphone when in bed, while another teen is actively posting and reacting on their social media accounts. It is likely that these different types of media use lead to very different effects on arousal and sleep.

Moreover, not only the type of content that adolescents consume might have an effect on sleep but also how these media are used. For example, interactive media (i.e., video games) seem to have a stronger negative impact on sleep than the passive use of media (i.e., watching a DVD; McManus et al., 2020; Weaver et al., 2010). Similarly, engaging in media multitasking is also related to sleep problems among adolescents (van der Schuur et al., 2018). These studies stress the importance of investigating not only screen time but examining more specifically the types of digital media use and the ways digital media are used.

There is limited understanding about the mechanisms that link varying content types and usage behaviors to sleep quantity and quality. So far, it has been frequently suggested that digital media use leads to heightened physiological arousal (Exelmans & van den Bulck, 2017b). However, specific types of media content may not necessarily increase physiological arousal but might lead to increased cognitive alertness that prohibits sleep (Weaver et al., 2010; Wuyts et al., 2012). Empirical investigations of these mechanisms for digital media are largely missing. One study showed small effects of video game play on alertness but not on arousal, stressing the importance of differentiating between these two processes (Weaver et al., 2010).

In sum, our current understanding of which digital media content factors are related to sleep, and through which mechanisms, is very limited. We know very little about whether specific content and usage patterns affect the varying sleep indicators differently and through which underlying mechanisms content affects sleep (see also Hale & Guan, 2015).

Future Research Directions for Digital Media and Sleep

Although concerns that media negatively affect the sleep of adolescents have a long tradition, these worries are exacerbated with the rise of smartphones and social media as these media types are used more than any other type of media by youth, and are often carried with them to bed. To avoid negative effects of digital media on sleep, the standard advice to adolescents is not to use any types of digital media in the two hours before going to bed (LeBourgeois et al., 2017). This is also reflected in current intervention studies that solely focus on removing digital media from the bedroom altogether (Martin et al., 2020). Although this advice is common and accepted by many, there are at least two problems related to this advice.

First, this strategy is in stark contrast to adolescents’ lived experience and developmental needs, and consequently it is unlikely that adolescents will agree to completely ban these devices from their bedrooms. Second, this advice is based on a rather simplistic view on the effects of digital media on sleep that considers the use of the device as universally detrimental. However, how exactly adolescents use digital media before bedtime can vary tremendously, plausibly resulting in differential effects on their sleep quantity and quality. Despite years of research into the effects of digital media on sleep, there are still important shortcomings in the literature that make it difficult to draw final conclusions about the effects of digital media on sleep. Solving these issues in future research is critical to being able to provide adolescents with effective advice on how to use digital media in healthy ways.

Causality

Although there is consistent evidence in the literature for a negative relationship between digital media use and sleep, the direction of this relationship is less than clear. The vast majority of the existing studies are based on cross-sectional designs, making it impossible to draw conclusions about the direction of the relationship (Exelmans & van den Bulck, 2019). Although it is generally assumed that the use of digital media deteriorates sleep, it could also be that the relationship is reversed in that adolescents who sleep less tend to use more digital media. For example, adolescents who do not sleep well might use digital media as a means to cope with stress and insomnia, or because they are depleted and do not have the capacities for regulating their media use efficiently. For example, university students used more social media on days they had slept less during the previous night (Mark et al., 2016). Similarly, sleep-deprived children watched more TV during the day in an experimental study (Hart et al., 2017).

Findings like this cast doubt on the idea that there is a simple cause-and-effect relationship between digital media and sleep. Recent advancement in media effects theories conceptualize media use and effects as reciprocal, evolving dynamically over time (Slater, 2007). In the case of media use and sleep this could mean that adolescents suffering from sleep problems are more likely to use more media that in turn may further deteriorate their sleep. This dynamic and reciprocal nature for smartphone use and sleep is understudied as it demands assessing use and effects over longer time periods in the natural environment of adolescents. One two-wave study found some evidence by showing that media use and sleep times were reciprocally related in adolescents over a one-year period (Poulain et al., 2019). Understanding the nature of the relationship between digital media use and sleep is of key importance for our understanding of the effects of digital media and for intervention and prevention programs.

Individual Responses and Potential Facilitating Effects

Recent theoretical advances in media effects research stress the importance of individual susceptibilities to media effects (Beyens et al., 2020; Valkenburg & Peter, 2013). Also, for the relationship between sleep and digital media, individual differences are likely to be of importance. First, individuals differ in how they use digital media before sleep. For example, Scott and Woods (2018) showed that adolescents with higher levels of fear of missing out tended to use social media longer before sleep time and were more cognitively aroused before falling asleep. Thus individualized usage patterns might lead to varying effects. This is also important because not all evening media diets might be problematic. Some adolescents might use their smartphones in a way that benefits their sleep by actually decreasing their aroused state. This assumption builds on established media effects paradigms that argue that media are used to regulate arousal levels and to establish physiological homeostasis (Zillmann, 1988). For example, people can use apps to seek out social support, relax, and regulate sensory stimulation (Harrison et al., 2019). Research has shown that some people report that they use media in bed to wind down from the day (Eggermont & van den Bulck, 2006). However, little research has investigated whether digital media can be used in ways that benefit adolescents’ sleep. Understanding such effects could help to educate adolescents to use their smartphones in more beneficial ways.

A second reason why it is important to study individual differences is that, while uniform effects of some content are possible, adolescents likely differ in their individual responses to digital media content. For example, one study found that adolescents who used social media more frequently slept less well than those who used social media less frequently. However, this effect disappeared when social media stress was taken into account, showing that only those respondents who experienced high levels of stress from their social media use suffered from sleep problems (van der Schuur et al., 2019). Moreover, this study showed that social media use was more problematic for the sleep of girls and early adolescents. Similarly, others found that only those who were more emotionally invested in their social media use slept less well (Woods & Scott, 2016), and that physiological reactions to violent game play differed depending on previous game experience (Ivarsson et al., 2013). Investigating these individual responses to smartphone use is crucial to understand why specific content is problematic for some adolescents but not for others.

Improved Measurement

The vast majority of existing studies relied on self-reports of media use and/or sleep. Self-reports for media use and sleep have been shown to be unreliable and it is thus likely that existing studies suffer from substantial measurement errors. Luckily recent developments in digital media and sleep tracking make it easier to assess digital media use as well as sleep unobtrusively and objectively. For example, there is a multitude of commercially available sleep trackers available with some studies showing promising results using them. We therefore hope that future research will try to combine self-reports with more objective measures for both digital media use and sleep. Assessing the complexity of digital media use objectively will be a crucial step to move beyond investigating screen time toward understanding differential effects of specific content (Carter et al., 2016; Hale et al., 2019; Scott & Woods, 2019).

Overall Conclusion

Parents, educators, and researchers alike are interested in the effects that our digitalized society has on adolescents. Whether digital media impairs attention and sleep has been investigated in a large amount of studies. Yet, the conclusions that we can draw are still limited. Overall, there is compelling evidence that adolescents who use social media more frequently and who are engaging in media multitasking more frequently are more likely to show attention problems in their everyday lives. Moreover, using digital media before bedtime is related to less sleep and more sleep problems. However, the key question of whether digital media causally impairs attention and sleep cannot yet be conclusively answered. To answer this question, it is crucial for the field to advance the theoretical as well as methodological approaches that we currently employ.

Concerning theory development, it is of key importance to identify content characteristics and affordances of digital media that drive such effects. Extracting these factors is crucial to understand not only the effects of today’s digital media landscape but also the effects of future media technologies that will emerge (see also Orben, 2020). Moreover, identifying content characteristics will allow us to differentiate potential detrimental from facilitating digital media use. For some adolescents, specific types of media use might have beneficial effects, for example, when they use relaxing smartphone content before they go to bed. Such beneficial effects are oftentimes neglected in current research.

Once we have a clearer theoretical understanding of the content characteristics that drive effects, we need to employ methodological techniques that are able to empirically test those effects in more precise ways. For this, it is important to move beyond cross-sectional studies relying on self-reports of general “screen time” toward assessing digital media in its complexity. Current technological developments facilitate the tracking of digital media use and sleep unobtrusively, objectively, and continuously. Moreover, current advancements in computational methods allow us to integrate, extract and analyze these types of complex data in more efficient ways. This will pave the way toward more advanced studies that examine the dynamic nature of digital media use, sleep and attention in unprecedented ways, and that will accelerate our knowledge of the effects of digital media on youth.

14 Digital Media, Suicide, and Self-Injury

Kaylee Payne Kruzan and Janis Whitlock

While interest in the relationship between media use and young people’s mental health is not new, the complexity of newer media technologies present novel research challenges – largely due to the interactive, multidimensional nature of contemporary communication technologies, such as those typified by social media environments. While early media studies focused primarily on effects of “screen time,” studies of modern-day social media must grapple with a number of overlapping and influential factors since effects are no longer related to mere exposure to potentially harmful content, but to the interactions that take place as individuals use and shape these platforms, as well.

The relationship between social media and self-injurious behaviors – specifically suicidal thoughts and behaviors and nonsuicidal self-injury (NSSI) – emerged as a primary research focus soon after social media came into widespread use, perhaps due to the well-established links between both media exposure and well-being (Wartella & Reeves, 1985) and to media effects and suicidal thoughts and behaviors (Phillips, 1974). This focus was reinforced by studies linking widely covered suicides (Niederkrotenthaler et al., 2012) and popular shows depicting suicide (Swedo et al., 2020) to upticks in self-injury and suicide-related activity.

This chapter is devoted to examining the relationship between social media and self-injurious thoughts and behaviors. Self-injurious thoughts and behaviors (SITB) describe thoughts and behaviors with (e.g., suicidal ideation, suicide plans, gestures, and behaviors) and without (e.g., NSSI) suicidal intent (Miller & Prinstein, 2019). While the developmental trajectories of NSSI and suicidal thoughts and behaviors differ from one another (Fox et al., 2015), SITB are not always clearly delineated from one another in the literature, in part because they commonly co-occur and in part because they each contribute to an increased risk for future suicide attempts (Kiekens et al., 2018). Such conflation applies to the literature on which this chapter draws. For simplicity, we will use the term SITB to refer to self-injury with, and without, intent in this chapter and we will refer to more specific constructs within this broader term when studies focus on a narrower sample.

Chapter Aims

This chapter includes two overarching aims: (1) to summarize research on the risks and benefits of social media use for SITB-related outcomes, including what is and is not known about primary mechanisms at play in these relationships and (2) to identify high-level implications, including opportunities and challenges for future research, intervention, and prevention efforts. The first section provides an overview on the prevalence and presentation of SITB in adolescence and the role of social media in SITB, while the second section summarizes findings related to the risks and benefits of social media use for SITB, and key mechanisms involved in these relationships. The final section covers implications for research, practice, and policy, through high-level opportunities and challenges.

Background

Adolescence and SITB

Understanding and addressing SITB is of major public health importance. Suicide is the second leading cause of death among young people between the ages of 10 and 24 globally (Curtin et al., 2016). Among US-based adolescent populations, lifetime prevalence of suicidal thoughts and behaviors is between 3.1% and 8.8% for suicide attempts and between 19.8% and 24.0% for suicidal ideation, with a marked increase in both suicidal ideation and behavior between the ages of 12 and 17 (Nock et al., 2008). Rates of NSSI – “the deliberate, self-inflicted damage of body tissue without suicidal intent and for purposes not socially or culturally sanctioned” (International Society for the Study of Self-Injury, 2018) – range from 17% to 37% among adolescents and young adults (Jacobson & Gould, 2007; Swannell et al., 2014).

Self-injurious thoughts and behaviors typically emerge in early- to mid-adolescence, with average age of onset for NSSI between 13 and 15 (Gillies et al., 2018), and mid- to late-adolescence for suicidal thoughts and behaviors (Nock et al., 2013). Older adolescents and young adults are more likely to die by suicide (Cha et al., 2018), when compared to younger adolescents – a pattern consistent with the idea that risk of engagement in serious suicide-related behaviors increases over time as experience of trauma and/or distress accumulates and interacts with bio-psycho-social developmental changes in ways that enhance vulnerability to cognitive and emotional challenges (Steinberg, 2010). Adolescence is also characterized by a highly social orientation, increased propensity for risk taking, and individuation/identity formation – each of which may interact with social media use in ways that amplify, or increase susceptibility to, potential media effects.

The Role of Social Media and SITB

Three decades of experience with, and empirical study of, unidirectional media affirms the potency of media influence on behavior, particularly for adolescents and children (Brown et al., 2006). The empirical link between exposure to violent media content and child and adolescent aggression was central to early media concerns and resulted in coordinated policy responses (US Senate, 2000). More recent efforts to understand the effects of social media on youth mental health retain a heightened focus on the potential adverse effects, such as: cybervictimization (John et al., 2018; Massing-Schaffer & Nesi, 2020), internet addiction (Jasso-Medrano & López-Rosales, 2018), and exposure to graphic self-injury and suicidal content (Arendt et al., 2019). It is thus not surprising that there are serious concerns about the impact that social media may have on individuals who bring preexisting vulnerabilities to online exchanges, such as SITB-vulnerable young people.

While attention to each of these domains has translated into research on social media effects of value to professionals, researchers, and platform designers, it has not yet led to robust understanding of the precise risks that social media pose to youth mental health – largely due to the number of contingencies that require disentangling and a need for methodological innovation (Whitlock & Masur, 2019). Moreover, while concern about the impact of social media on youth continues to be a regular feature of public worry and headlines, it is also recognized that social media offers important support to users, including SITB-vulnerable individuals, by (1) facilitating social connection (Duggan et al., 2012), (2) extending the reach of prevention/intervention efforts (Thorn et al., 2020), (3) linking young people who are already engaging in SITB with much needed information and support (Lavis & Winter, 2020; Lewis & Michal, 2016), and (4) increasing public awareness of SITB and reducing stigma (Li et al., 2018; Nathan & Nathan, 2020). A balanced and nuanced approach that takes into account both the risk and benefits of social media for SITB outcomes is needed to effectively consider the many factors that likely mediate and moderate social media effects.

Brief Overview of Methods Used to Study the Relationship Between SITB and Social Media

A brief historical overview on the methodological approaches most commonly used in social media and SITB research is both helpful in contextualizing the risks and benefits and in surfacing methodological frontiers in this domain. In general, SITB-focused research aims have (1) described online content and activity related to SITB, (2) explored the relationship between online activity and SITB, and (3) identified risks germane to intervention efforts. While these efforts have laid the theoretical and empirical foundations necessary for inferring and anticipating risks and benefits and for understanding key mechanisms, they have been less effective in surfacing and disentangling clear causal relationships between social media use and SITB behaviors or in describing the moderating role preexisting SITB vulnerability plays in these relationships.

In general, research documenting potential effects of SITB-related content and exchange has been more straightforward to generate than research aimed at understanding causal relationships between online activity and SITB; in part because the latter requires innovative methods that balance privacy and ethical concerns with the need for cross-ecological and granular approaches capable of disentangling effects. Moreover, because the nature of communication technologies is so dynamic, the research methods required to understand effects must also be dynamic. Most early work focused on content and thematic analyses to investigate common themes in online discussions about self-injury and suicide (Rodham et al., 2007; Whitlock et al., 2006). Surveys were (and still are) used to assess motives for social media use and to understand the perceived effects of use (Lewis & Michal, 2016).

Recent advances in the application of computational methods to social media research have paved the way for investigation of links between online activities and SITB risk, largely through tracking patterns in linguistic and behavioral markers (De Choudhury et al., 2016; Du et al., 2018). Ecological momentary assessments (EMA), or diary methodologies, have been used to understand the relationship between social media use and outcomes related to mental health. For example, EMA methods were used to understand what behaviors young people engage in instead of self-injury (Fitzpatrick et al., 2020). Longitudinal studies have begun yielding results, but even these are limited by challenges in disentangling between- from within-effects of media use, understanding risks and benefits accrued to vulnerable subgroups, and the way that both developmental stage and specific social media affordances interact with social media use (Schemer et al., 2020). In sum, research focused on the intersection of SITB and social media use has evolved from a focus on more static content in online communities (precursors to social media) to more dynamic interactions between user behaviors, content, and offline markers over time. While important methodological challenges remain, much has been learned; this is the focus of the following sections.

Risks of Social Media for Self-Injury and Suicide

Study of the ways in which use of social media increases SITB risk reveals a complex portrait of effects, some of which clearly enhance risk of SITB behavior and others that may protect against such risk. This section details the dominant categories of risk identified thus far including: (1) exposure to SITB content, (2) normalization and narrative reinforcement, (3) contagion, (4) cyberbullying, and (5) heavy social media use.

Exposure to Suicide and Self-Injury Content

As with traditional media, at least some research documents a link between exposure to suicidal and self-injury social media content and increased risk for SITB experiences. Exposure to digital SITB-related content is not infrequent – in one study, 25% of young people were exposed to suicide stories through social media (Dunlop et al., 2011). This is concerning because increased exposure to self-injury-related content has been associated with decreased aversion to self-injury and to future suicidal ideation in past work (Franklin et al., 2016) and because habituation to SITB content may reduce barriers to, and increase the acquired capability for, suicide (Massing-Schaffer & Nesi, 2020). Moreover, such risks may not diminish over time. For example, in a study of effects of exposure to self-harm content on Instagram, researchers found that lifetime exposure to self-harm content was associated with increased SITB risk. Furthermore, exposure was related to an increase in self-harm behaviors, suicidal ideation, and hopelessness one month later, even when controlling for preexisting SITB vulnerability (Arendt et al., 2019).

While it is possible that well-moderated sites could minimize harm resulting from unregulated exposure to triggering content, empirical evidence suggests that even with site moderation individuals can be exposed to triggering graphic or emotional images or text (Baker & Lewis, 2013; Lewis & Michal, 2016), including tips on concealment, suicidal ideation, or plans (Dyson et al., 2016). Indeed, in the aforementioned Instagram study, only 20% of those who reported seeing self-harm content intentionally searched for it (Arendt et al., 2019). Further, some studies indicate that a subgroup of individuals access online communities in order to sustain or trigger self-injury and share maladaptive techniques (Lewis & Seko, 2016; Whitlock et al., 2006).

Awareness of the potential for social media content to have harmful effects has led to an increase in moderation efforts, often by platform developers themselves. Popular social media platforms like Instagram, for example, have built in “sensitivity screens” (i.e., trigger warnings) that are meant to shield content related to self-injury and other harmful behaviors enabling users to view content if they clear the shield (Carman, 2019). However, even these efforts require empirical study since, in this case, evidence suggests that use of trigger warnings to decrease risk of SITB-related harm has relatively limited effects on distress (Sanson et al., 2019) and may increase anticipatory anxiety in some cases (Gainsburg & Earl, 2018). Effects of what a user does in response to a trigger warning is also less intuitive than it might seem. For example, a study focused on self-injury related activity on TalkLife, a mobile peer-support app, showed that choosing to dismiss a trigger warning and view self-injury content was related both to greater intentions to injure and greater ability to resist injuring within a week’s time (Kruzan et al., 2021). Notably, posting triggering content was related to increased odds of both self-injury thoughts and behaviors. In sum, more work is needed to explicate both the factors that contribute to effects related to exposure to SITB content and the potential protective value of moderation efforts, like trigger warnings.

The Downside of Social Connection on Social Media: Normalization and Narrative Reinforcement

The fact that self-injury and suicide-related posts so frequently co-occur with themes of loneliness underscores the important role that social connection plays in mental health and well-being (Cavazos-Rehg et al., 2017). Indeed, the promise of rich social connection is one of the factors that makes participation in social media so appealing. However, empirical evidence suggests that the “social” part of social media is simultaneously a risk and a protective factor for SITB. While the perceived and actual social support that comes from social media’s ability to connect young people struggling with self-injury and suicide can be beneficial and SITB-protective, regular exposure to SITB content and association with other individuals struggling with SITB may expose vulnerable adolescents to communities where self-injury is normalized or encouraged, even if not overtly or consciously (Rodham et al., 2007; Whitlock et al., 2006). This “normalization effect” is commonly seen in studies of online communication about self-injury where young people discuss self-injury thoughts and behaviors in detail and often minimize the severity of self-injury and its consequences (Dyson et al., 2016). Moreover, the tendency for individuals to co-construct and then reinforce foundational narratives, sometimes termed “narrative reinforcement,” that essentially justifies the need for and use of SITB-linked activities, can lead to desensitization and normalization of behavior, especially when self-injury is depicted as painless and effective (Whitlock et al., 2007).

Even when a user is trying to minimize exposure to triggering content, most studies show that it is common for pro-recovery messages and encouragement to occur alongside pro-self-injury posts and comments, such as advice on how to injure safely and how to conceal wounds (Lavis & Winter, 2020; Whitlock et al., 2006). This may not only normalize self-injury, but may also trigger SITB-impulses or discourage use of alternative coping strategies or professional help seeking (Dyson et al., 2016; Smithson et al., 2011). In sum, while the emotional support received through social media sites can positively influence the recovery process, this support may detract from the severity of the behavior, potentially slowing the change process (Dyson et al., 2016).

Contagion: Spread and Scale of Social Media Messages

The idea that exposure to a behavior through media may be “contagious” is a subject of long-standing research interest. Research shows both an increase in the number of SITB themes in on- and offline media, and concomitant concern that such content may contribute to onset or maintenance of SITB among vulnerable individuals, mostly likely through social modelling (Jarvi et al., 2013). While the adverse impact of SITB social media content on individuals with existing vulnerabilities is intuitive, recent work suggests that even individuals without existing vulnerabilities may be at risk of adverse outcomes from SITB-related themes in social media. For example, there is evidence that viewing suicide-cluster-related posts (e.g., vigils, memorials), online news articles related to suicide, and watching the Netflix series 13 Reasons Why (which features suicidal content) is associated with increased odds of suicidal ideation and attempts, among students both with and without prior self-injury history (Swedo et al., 2020). This study did not control for other known risk factors, like depression or anxiety, and it cannot rule out the possibility that other important preexisting vulnerabilities exist, but it does suggest that even individuals without prior self-injury history are adversely affected by some media content. This possibility is also implicit in research that finds an over 14% increase in population-based suicide trends for young people between 10 and 19 (Niederkrotenthaler et al., 2019) and “excess” hospitalizations for suicide attempts among young people (Cooper et al., 2018) following the release of 13 Reasons Why.

In a similar vein, research reveals that individuals who post suicidal content are more tightly clustered in friend, or reposting groups, than users who do not post suicide-related content. This supports the idea that individuals tend to gravitate to like-minded others online in ways that may heighten likelihood of narrative reinforcement, and concomitantly, risk of spread among those most vulnerable (Colombo et al., 2016). However, the authors also note that re-tweeting behavior connects users whose posts contain suicidal ideation with users whose posts do not, providing evidence for the potential of contagion across diverse networks.

Contagion and Social Media “Challenges”

Social media challenges allow users to pose a behavioral challenge to followers who then receive online community recognition for meeting the challenge – most often over a series of days or weeks. While potentially harmless, or even beneficial, challenges can also heighten individual SITB risk. The Blue Whale Challenge, which occurred through social media from 2013 to 2017, is purported to encourage youth to participate in a series of tasks over 50 days that involve self-harm and culminate in a suicide challenge (Sumner et al., 2019). Not only is the challenge itself associated with heightened SITB risk, but YouTube media covering this challenge often violated Suicide Prevention Resource Center guidelines (Khasawneh et al., 2020). Such challenges also underscore the ways in which the very features that make social media so attractive also present novel risks.

Cyberbullying

Bullying is a long-standing source of stress for young people and this holds as true in online social settings as it does in offline social settings (John et al., 2018). Cyberbullying, a term used to describe bullying that occurs online, is also associated with heightened risk for SITB. Notably, it is not just the victims of cyberbullying who are at elevated SITB risk. A recent meta-analysis shows that youth victims of cyberbullying are over twice likely to engage in self-harm, to report a suicide attempt, and to report suicidal thoughts, when compared to nonvictims (John et al., 2018). Even one episode of cybervictimization increases risk of suicidal ideation (Hirschtritt et al., 2015). Moreover, the risk of SITB after a cyberbullying incident increases significantly among individuals with existing vulnerabilities. Indeed, in a study of adolescents presenting to Canadian emergency departments for mental health complaints, those reporting histories of cybervictimization were over 11 times more likely to report suicidal ideation (Alavi et al., 2017). Also, being both a victim and perpetrator of cyberbullying doubles the risk of reporting suicidal thoughts when compared to those who have one of these experiences (Bonanno & Hymel, 2013; John et al., 2018).

Heavy Social Media Use

Research has also shown that risk of NSSI and SITB increases with heavy social media use (Lee et al., 2016; Twenge & Campbell, 2019). Indeed, in a study of Canadian high school students, those who spent more than two hours a day on social media had were five times more likely to experience suicidal ideation when compared to peers reporting fewer than two hours of social media use a day (Sampasa-Kanyinga & Lewis, 2015). Similarly, adolescents who report heavy digital media use are twice as likely to report suicidal thoughts, suicide plans, and suicide attempts when compared to light users, according to a large survey study (Twenge & Campbell, 2019). And, in a recent review of seven studies researchers documented a direct association between heavy social media/internet use and suicide attempts (Sedgwick et al., 2019).

Interestingly, some studies show that some social media use is better than no use (Kim, 2012; Lee et al., 2016). These findings are consistent with broader literature on social media use and well-being that suggests curvilinear relationships between social media use and well-being with benefits derived from some use, versus no use, and risks increasing most significantly from low or moderate to heavy use (Kim, 2012; Przybylski & Weinstein, 2017; Twenge & Campbell, 2019). Specifically, risks increase most significantly from low (<1 hour a day) or moderate to heavy use (>5 hours a day) (Twenge & Campbell, 2019). One explanatory theory is that time spent on social media displaces other activities that could be beneficial for mental health, such as physical activity, in-person social interaction, and sleep – all risk factors for suicide (Porras-Segovia et al., 2019; Sedgwick et al., 2019; Verkooijen et al., 2018).

Benefits of Social Media for Reducing Self-Injury and Suicide

While risks associated with social media use are a focus of continued empirical investigation, salutary effects have also been documented. Reviews focused on social media and SITB (deliberate self-harm: Biernesser et al., 2020; Dyson et al., 2016 and self-harm and suicide: Daine et al., 2013; Marchant et al., 2017; Memon et al., 2018) converge in their identification of tangible benefits, including enhanced: (1) social support and connectedness, (2) self-knowledge/expression, and (3) access/exchange of resources/information. Key empirical findings for each area are described below.

Social Support and Connectedness

One of the primary perceived benefits of social media use is the exchange of social support not bounded by time or geography. This is important because social support is known to buffer effects of negative life events, enhance mental health and well-being (Cutrona & Suhr, 1992), decrease feelings of isolation, lead to sense of purpose, and to promote feelings of acceptance or being understood (Daine et al., 2013). Opportunities for social support through social media can be powerful for young people with SITB, since stigma is often an impediment to offline help and support seeking. Online environments allow for anonymity and carry few clear social penalties for candid sharing, which makes such environments particularly attractive to individuals concerned about disclosing SITB-related behaviors or impulses to people in their offline lives (Duggan et al., 2012). And, since social support is a critical protective factor for SITB (Joiner et al., 2012), social exchange in social media forums offers a promising alternative to offline sharing.

It is thus unsurprising that empirical evidence suggests that young people with SITB histories use the Internet more often than their peers (De Riggi et al., 2018; Memon et al., 2018) and that it is a preferred means for seeking and receiving help (Frost & Casey, 2016). For example, youth with suicidal ideation are more likely to report online-only friendships, relative to those without suicidal ideation, and these friendships appear to buffer the harmful effects of relational victimization and stress (Massing-Schaffer et al., 2020). Nearly one-third of young people with a history of self-injury had reported online help seeking in one study – and those who sought help online were more distressed and suicidal than those who had not (Frost & Casey, 2016). Additionally, adolescents with more recent NSSI have higher levels of online support seeking, compared to those with past or no NSSI history (De Riggi et al., 2018). Even when individuals have a strong support system offline, they may have trouble accessing support in times when they need it (Kruzan et al., 2021; Lavis & Winter, 2020). The immediate nature of social support exchange on social media may be important for individuals who struggle with SITB given that intense urges are commonly cited as a key barrier to behavior change (Kruzan & Whitlock, 2019) and findings showing that young people frequently look for, and receive, emotional support online when they are experiencing an urge (Lewis & Michal, 2016; Rodham et al., 2007).

Not all social support is equal, however. While some work suggests that young people perceive benefits from participation (Brown et al., 2020; Lewis & Michal, 2016), others note the “mundane” or safe nature of the advice, which leads to questions of actual utility (Smithson et al., 2011). The availability and immediate accessibility of such support is nonetheless quite appealing – as is the fact that support is exchanged among peers with shared experience and experiential knowledge (Marchant et al., 2017; Thoits, 2011). Research consistently documents a preference for peer versus professional support for NSSI and the tendency for young people to confide SITB in peers versus others in their social network (De Riggi et al., 2018), something social media facilitates organically.

The question of whether such peer support is helpful for SITB outcomes remains nascent. Early work showed positive associations between social support received and decreased self-injury behaviors (Murray & Fox, 2006), but research directly connecting social support through social media use to its effects on SITB outcomes is limited. One experimental study varying exposure to hopeful or hopeless YouTube videos, found that hopeful messages were associated with increased positive attitudes toward recovery, suggesting shifts in recovery-oriented subjective norms (Lewis et al., 2018). Interestingly, there were no attitudinal changes in those viewing hopeless messages.

Self-Knowledge and Expression

Beyond the use of social media as a source of social support is its role in facilitating self-expression and exploration. Being able to connect and provide mutual support, narrate experiences, and self-reflect, while also maintaining autonomy and anonymity, are all identified as clear benefits to social media use among individuals with SITB history (Coulson et al., 2017; Rodham et al., 2013). Indeed, self-oriented motivations such as understanding NSSI experience or expressing oneself through narrative description or other forms of creative expression are potent motives of online activity (Seko et al., 2015). Insight gleaned through sharing one’s story and encountering resonance in others’ stories is important in recovery and is associated with active information seeking, increased self-efficacy, and enhanced self-awareness (Kruzan & Whitlock, 2019). Since young people frequently provide advice to others online (Seko et al., 2015; Whitlock et al., 2006), it is also possible that seeing oneself as a valued mentor to others with shared struggles may increase commitment to recovery processes. Online self-presentation and expression can assist in developing self-understanding, and be associated with beneficial shifts in self-perceptions (Kruzan & Won, 2019; Valkenburg, 2017).

Exchange of Resources and Information

Use of social media to both identify and exchange coping techniques is also common and potentially beneficial (Duggan et al., 2012) for individuals navigating self-injury or suicidal thoughts and urges (Lavis & Winter, 2020; Lewis & Michal, 2016). Tips on how to reduce the urge or replace self-injury behaviors are also highly salient. For example, in a study of three different social media sites (Reddit, Instagram, Twitter) researchers found a rich exchange of coping advice related to visual, distraction, and sensory techniques effective in reducing urges (Lavis & Winter, 2020). There is also evidence that topics related to professional help seeking for SITB are a feature of some online exchange (Lavis & Winter, 2020), but whether this is common remains unclear since there is work suggesting that online exchange does not lead to increased professional help seeking (R. C. Brown et al., 2020) and because this line of inquiry remains underexplored.

The power of social media exchange to alter offline behavior does open opportunity for development of more formal intervention. Online peers may be uniquely positioned to provide advice on treatment and coping strategies, and this advice may be easier to digest, and apply, when coming from someone who has “been there” (Naslund et al., 2016). Such exchange can be considered a unique and potent form of expertise (Marchant et al., 2017) that can be leveraged to deliver coping- and recovery-supportive messages and resources. Since not all resources exchanged through social media are evidence-based, and some can be harmful or depict self-injury as an effective coping strategy (Lewis & Baker, 2011; Seko & Lewis, 2018), it is crucial that the nature of naturally occurring exchange is understood and mitigated when potentially harmful.

Key Mechanisms: Moderators and Mediators of Effects on SITB

Individual, developmental, and social-contextual factors are all empirically and theoretically relevant when considering susceptibility to SITB and media effects, especially since young people with preexisting vulnerabilities, such as other mental health conditions, are more likely to be exposed to harmful content (Dyson et al., 2016). SITB-specific individual-level factors such as prior SITB history may moderate social media effects (Dyson et al., 2016). Cyberbullying may also moderate or mediate social media effects (John et al., 2018), and while underexplored, factors such as offline support and prior SITB help seeking are likely to moderate the effect of social media on SITB. For example, social media effects, particularly negative effects, might be less damaging to individuals who have rich social supports outside of social media. A review of the most acknowledged likely mediators follows.

Mental Health History

Just as prior mental health history has the potential to moderate the effects of social media use on SITB outcomes, it can also mediate this relationship. In some work, the relationship between heavy social media use and NSSI was mediated by factors such as suicidality, anxiety, and affective and psychotic disorders (Mészáros et al., 2020).

Affect and Intentions

Emotional affect and motives for use are also likely mediators of the relationship between social media and SITB. The connection between NSSI and affect is well established, and may be particularly important in understanding interactions that lead to risks or benefits of social media use, since both NSSI (Klonsky, 2007) and social media use can be ways to modulate emotion (Rideout & Fox, 2018). Indeed, young people can deliberately seek out uplifting, distressing, or neutral messages that reflect, and may impact, their own affective state. While few studies have examined the role of mood in the relationship between SITB and social media use, young people with lived NSSI experience often discuss mood as part of their use of social media and related technologies (Seko et al., 2015).

Interactional Factors

In addition to the amount of use, the way someone uses social media is consistently connected to mental health outcomes (Verduyn et al., 2017). This trend holds for SITB-related studies, as well, but the patterns of effects are not entirely intuitive. In a cross-sectional study of the association between SITB (both NSSI and suicidal thoughts and behavior) and social media use type among Norwegian university students, researchers found that active public social media use (e.g., posting, commenting) was associated with increased odds of NSSI ideation and behaviors and suicide attempts, whereas social private use (e.g., messaging friends) was associated with reduced odds of all NSSI and suicide outcomes (Kingsbury et al., 2021). Passive nonsocial use (e.g., reading news) was associated with decreased odds of NSSI ideation, NSSI, and suicidal ideation, and active nonsocial use (e.g., for studies) was associated with decreased odds of suicide attempt. In parallel with the broader literature on social media effects on well-being, these findings suggest a nuanced relationship that differs by types of engagement.

Social Comparison Processes

Social comparison is a primary mechanism through which social media use impacts mental health and well-being (Appel et al., 2016; Kruzan & Won, 2019; Wang et al., 2017). Upward social comparison – wherein individuals compare themselves to those who are perceived as better off – has been associated with reductions in self-esteem, increased negative affect, and envy (Appel et al., 2016; Wang et al., 2017). Consonant with this general trend, Kingsbury et al., (2021) found that the presence of social comparison is associated with increased odds for all NSSI and suicidal outcomes. However, social comparison processes may look slightly different on social media sites or forums that are structured almost entirely around conversations about SITB (e.g., TalkLife) where the general positivity bias documented in mainstream social media does not exist. In light of its influence, the role of social comparison for SITB risk in social media should be explored further.

Opportunities and Challenges

Despite limitations, social media and related platforms, like mobile apps, offer excellent opportunities to leverage modern communication technologies in ways that provide timely and scalable intervention and, ideally, prevention. Such opportunities, however, present unique challenges related to methodological innovation and strategies for effectively addressing privacy and ethical considerations.

Opportunities: Amplifying the Beneficial Potential of Social Media

In addition to the opportunities inherent in the nature of the technology’s design, such as the possibility for enhanced social connection and belonging, there are unique opportunities for: (1) identification/detection, (2) intervention, (3) prevention, and (4) awareness/stigma reduction.

Identification/Detection

Automated methods for predicting SITB risk and social media effects are promising as they are capable of considering complex combinations not likely to arise from more traditional assessments (Walsh et al., 2017). Creative use of machine learning has been successful in early efforts to detect and address suicidal content, particularly when used to detect and intervene with novel online risks, such as pro-suicide games (Sumner et al., 2019). This same method can also be used to identify at-risk users. Natural language processing and topic modeling have been leveraged to understand changes in suicide-related content following national reports of celebrity suicides (Kumar et al., 2015) and changes in emotional expression and self-attentional focus are consistently identified as indicators of higher suicide risk, for example (Coppersmith et al., 2018; De Choudhury et al., 2016). However, most work has focused on high-level trends, rather than individual risk patterns, which would be useful for tailoring interventions. An exception to this is a study that was able to differentiate between users who are at risk of transitioning to suicidal ideation (De Choudhury et al., 2016). While discerning posts related to self-injury with, and without, suicidal intent is more difficult, it is a promising area for further investigation.

Intervention

As the ability to detect at-risk users who could benefit from additional resources improves, scalable interventions delivered through social media will be possible. Preliminary evidence suggests that young people would be receptive to digital interventions, such as those through social media (Naslund et al., 2016) and that digital interventions focused on acquisition and implementation of evidence-based SITB coping skills are likely to be efficacious in reducing self-injury (Rizvi et al., 2016; Schroeder et al., 2018). Such interventions could also serve as a decisional tool for future help-seeking behaviors, for both those at risk of SITB and concerned friends and family (Rowe et al., 2018).

Two frameworks particularly promising for early intervention in the social media environment are: (1) single session interventions (Schleider & Weisz, 2017) and (2) digital micro interventions (Baumel et al., 2020). Single session interventions (SSIs) – brief, but potent, treatments designed to last one session – have shown promise in reducing many mental health outcomes in adolescent populations (Schleider & Weisz, 2017). These interventions are scalable, potentially capable of reaching young people who are unlikely to come into contact with more formal/traditional services, and are flexible enough to be disseminated in multiple contexts, including social media. Additionally, the potential value of SSIs in reducing SITB has already been noted (Dobias et al., 2020).

Digital micro interventions (DMIs) are small “bite-sized” interventions designed to fit seamlessly into an individual’s natural use of media (Baumel et al., 2020). In contrast to the linear and/or single-platform approach DMIs work across a number of platforms (e.g., social media apps, text messaging) and involve a series of smaller, dynamic touch points that are responsive to young people’s media habits. Since at least one suicide prevention study suggests that young users want preventive interventions embedded in the platforms they already frequent (Thorn et al., 2020), DMIs may be particularly well suited for delivering SITB early intervention and prevention.

Prevention

Social media can be leveraged to increase awareness, reduce stigma, and provide psychoeducation at scale (Robinson et al., 2016). Simulation studies in this area demonstrate that suicide prevention efforts on social media have the potential to reach at-risk populations at a much larger scale than traditional methods (Silenzio et al., 2009). Despite this potential, few prevention efforts for SITB to date have been disseminated on social media. Some of the more innovative work in this area engages young social media users in codesigning workshops aimed at developing a social media campaign (the #chatsafe project) focused on safe communication about suicide online (Thorn et al., 2020). The project demonstrated that it is feasible to safely engage young people in codesigning a suicide prevention intervention (Robinson et al., 2018; Thorn et al., 2020). A number of auxiliary but useful key takeaways surfaced through this process, including finding that young people wanted to see guidelines through sharable content – including videos, animations, photographs – and that they want to feel visible in the media campaign (Thorn et al., 2020).

Awareness and Stigma Reduction

Destigmatizing mental health struggles and increasing positive discourse and disclosure is another opportunity for social media to address SITB. Social media can be used to gauge public perceptions of suicide, determine needs for literacy, and deliver psychoeducation when needed (Nathan & Nathan, 2020). Social media mining can also be leveraged to improve the performance of stigma reduction programs (Li et al., 2018). However, more research is needed to better understand how social media can be used to reduce stigma and promote open and nuanced discussions.

Challenges: Minimizing the Negative Potential of Social Media

Some of the challenges of studying and understanding the relationship between social media use and SITB outcomes are broadly related to (1) creating and maintaining a safe environment, (2) methodological innovation, and (3) privacy and ethical considerations.

Creating and Maintaining a Safe Environment

The need to attenuate negative effects of social media use and prevent further “digital harm” – or “online communication and activity that leads to, supports, or exacerbates, non-suicidal yet intentional harm or impairment of an individual’s physical well-being” (Pater & Mynatt, 2017) (p. 1501) is critical to creating and maintaining safe online environments. While much of the work focused on social media and SITB risks focuses on moderation, it is also useful to think about how spaces can be designed to facilitate connection and supportive exchanges and to make negative interactions less likely. To accomplish this, however, understanding of how platforms can be designed to protect users against negative experiences (e.g., cyberbullying) without sacrificing opportunities for user agency (including peer-to-peer intervention) at interaction and platform level must be enhanced and leveraged. Researchers in fields like human–computer interaction are particularly well suited to address these concerns due to their person-centered approaches, especially when working in collaboration with experts in both SITB and adolescent development and well-being.

Methodological Innovation

The dynamic nature of social media environments coupled with the broad-reaching messaging power presents new and important methodological challenges for research – all of which merit careful attention from scholars in various technical and clinical disciplines. Social media data has improved our understanding of the needs and struggles of young people with SITB histories and has been linked to markers of SITB risk. However, both automation and platform use preferences evolve rapidly – necessitating a flexible approach. While use of automated methods has powerful potential, algorithms are “black boxes,” and utility is not likely to be of universal ease or impact across platforms. Therefore, understanding variations in speed, efficiency, and utility of methods across platforms will be a key component of augmenting utility. It will be similarly important for researchers to consider how to best translate findings from sophisticated detection algorithms into practice and to have a set of guidelines for developing, and validating, social media interventions.

Two of the greatest needs for future research are to examine the temporal relationship between online activities and behavior change, and to discern which mechanisms contribute to desirable outcomes. To do this, it will be important to triangulate different types of data and methods (Lavis & Winter, 2020) and to consider new methodological approaches capable of tracking what participants actually see and do online. Combining EMA with tracking (logging media use), for example, may assess states rather than traits, reduce recall bias, and link fluctuations to the manifold situational factors and circumstances outlined in this chapter (Whitlock & Masur, 2019). Future research should also consider the bi-directional relationships between SITB and social media engagement (Lavis & Winter, 2020). To date, most work has focused on the impact of social media use on SITB risk; however, it is equally important to understand how individual histories of SITB and risk influence social media use.

Privacy and Ethics

Such methodological approaches pose significant ethical challenges and will require care in balancing potential ethical challenges inherent in such methods with the benefits they provide. One of the biggest challenges for platform designers, researchers, and policy-makers is navigating user privacy and ethics while also safeguarding against potential harms of free expression – both in terms of platform affordances and the research needed to better understand the complex interactions between social media use, SITB risk, and individual-level factors such as developmental stage and other risk and protective factors (Whitlock & Masur, 2019). There is also a need to establish universal protocols for how risk detection and accuracy is measured and applied across platforms (Westers et al., 2020). This will likely require continuous monitoring and updating of algorithms as the data available expands and brings with it questions about privacy and opting-in to such monitoring.

Conclusion

Evidence that young people go online, exchange support, and share relatively openly about their experiences is promising in that it presents grounds to understand young people’s experiences, detect needs, and design and deliver scalable preventative interventions. However, there are also risks associated with the social media environment such as exposure to, and the quick spread of, potentially harmful content. To better understand how we can best amplify the beneficial potential of social media, while minimizing the negative consequences, further research focused on disentangling factors that contribute most to the SITB–social media relationship is needed.