Written evidence submitted by Professor Andrew K Przybylski,
Netta Weinstein, and Amy Orben (SMH0140)

Submitted by:

Andrew K. Przybylski, Associate Professor, Oxford Internet Institute, Dept. of Experimental Psychology, University of Oxford

Netta Weinstein, Senior Lecturer, School of Psychology, Cardiff University

Amy Orben, Lecturer & D.Phil. Candidate, The Queen’s College, Dept. of Experimental Psychology, University of Oxford

As experts who have been studying the influence of social media and digital screen-use on young people's mental health for many years, we would like to thank you for providing the opportunity to weigh in on this important topic of intense and widespread public debate. With not a week going by without widespread coverage about technology effects on young people, we recognise the increasing difficulty to differentiate between high quality evidence and anecdotal or biased coverage. In this written evidence we build on our extensive research experience to outline the (lack of) evidence for technology effects on young people and the policy changes we would recommend to parliament.

Term of Reference 1: What evidence there is on the effects of social media and screen-use on young people’s physical and mental well-being — for better and for worse — and any gaps in the evidence

The proliferation of digital screens has fundamentally changed how humans work, play, and socialize. In the span of a decade, the amount of time adolescents spend online has more than doubled: from an average of 8 hours per week in 2005 to 18.9 hours [1]. While the amount of time young people spend with technology is on the rise, it is not clear that this has had a measurable impact on their well-being or health.

It is popularly assumed that the time spent using technology is intrinsically worse for young people than the time spent on ‘analogue pursuits’, which are traditionally portrayed as more worthwhile. Yet the evidence base for the effects of social media and digital screen-use is mixed and generally low in empirical quality. Some demographic factors have been successfully correlated with digital screen time, such as age [2], body mass [3][4], and minority ethnicity [4][5]. The evidence suggesting that children’s well-being [5][6], caregiver education [6,7][7,8], and household income [8,9][9,10] may be negatively related to screen time [10][11], however, is not clear-cut.

Research about how technology affects well-being and functioning is even less clear. Many studies show only small or null effects once proper control variables are used [11,12]. Moreover, more recent research, based on data provided by more than 120,000 British adolescents [13], indicates that the link between well-being and digital screen time is most probably non-linear (see Figure 1). With major components of their social life taking place online, time with technology can foster adolescent well-being: allowing them to develop their identity and build life and social skills [14,15]. If there indeed are benefits for moderate social media and smartphone use, it may be because technology provides opportunities to pursue these developmental challenges in a satisfying way.

Research results investigating the effects of screen time on sleep outcomes are similarly complex. One recent review of 87 studies  found that screen use is unrelated to sleep duration in 15 cases, linked by way of individual and contextual factors in 13 cases, and negatively related in 59 cases. [16]Results from longitudinal studies hint that this relationship is complex and may [17] or may not be bi-directional over time [18][14]. In other words, the results suggest that individuals who are unable to sleep are more motivated to use digital screens to manage their sleep problems, instead of a simple displacement effect where technology use directly decreases sleep time.

Term of Reference 2: The areas that should be the focus of any further research needed, and why

Across the board, a lot more research is needed to obtain even just an initial understanding of how technology use affects children. As researchers in the area, we find it concerning that, although there are exceptions (e.g. [13]), nearly all of the studies published to date investigate technology effects through an exploratory empirical lens, that is, the sampling and analysis plans are finalized after the data are collected.

It is important for policy-makers to understand that all studies on digital well-being are not created equal. Most findings are tentative because researcher biases can, and do, influence the analysis and interpretation of results [19][20][17][15]. In an area so heavily implicated in public debate, researchers will need to collect and analyse data in more rigorous ways to provide the concrete evidence necessary to underlie policy change. Given the policy importance and mixed findings, large-scale and pre-registered studies are needed.

To encourage this pre-registered research, U.K. research councils would need to place special emphasis on project proposals that integrate large-scale social data (e.g. Understanding Society, Millennium Cohort Study) with pre-specified analytic plans. These plans need to specify, in a detailed way, exactly how the data will be analysed before the data are collected. Only this level of methodological rigour will let researchers, and by extension policymakers, draw robust inferences about technology effects.

Term of Reference 3: The well-being benefits from social media usage, including for example any apps that provide mental-health benefits to users

Though this work is similarly tentative, as there has not been much high quality work done in this research area, some research indicates there might be benefits of social media usage. For example, many may think of gaming as a socially isolating activity, yet research indicates that online gaming handles are one of the first pieces of information shared by 38% of adolescent boys when they meet someone they want to become friends with [20]. Similarly, 83% of adolescents believe social media makes them more connected to their friends, and 68% say they have received social support by using these technologies in tough or challenging times [20]. Taking this together, there is good reason to think digital engagement, in moderation, may not be disruptive, and that it may even be a supportive part of development. The most promising area for research here might be in smartphone-based mental health apps [21]. These initiatives, either delivering short courses of Cognitive Behavioural Therapy or connecting citizens with mental health services, have been associated with lower levels of depression in a growing number of clinical trials. 

Term of Reference 4: The physical/mental harms from social media use and screen-use, including: safety online risks, the extent of any addictive behaviour, and aspects of social media/apps which magnify such addictive behaviour

Like the research literature concerned with technology effects on well-being, the evidence base concerning technology addiction is highly problematic and far removed from the public debate and media coverage [22,23]. The issues with addictive technology behaviours have been detailed in two recent debate papers considering steps taken by the American Psychiatric Association and WHO [24,25]. Unfortunately technology addiction is a highly-hyped area, and the term ‘addiction’ has become commonplace in public conversation. The unfortunate result of this is that the connotation of technology addiction has lost most of its technical meaning for both researchers and the general public.

With that understood, research based on American Psychiatric Association guidelines for technology addiction using open and preregistered methodological can tell us about the clinical reality behind the potential disorder. These studies indicate that participant response biases, if uncorrected, can inflate addiction prevalence estimates considerably: survey respondents are prone to exaggerating their technological addiction [26]. Further, these studies show gaming ‘addiction’ is significantly less prevalent than gambling disorder [27] in U.K. samples. Finally, and most importantly, these studies demonstrate that measures of gaming addiction are not linked to psychological health, physical health, or well-being over a six-month time period [28].

That said, the level of transparency among game developers with respect to ‘addictive’ mechanics is very low and would benefit from transparent collaborations with academics. There is a fundamental asymmetry between the quality and amount of data industry stakeholders have access to compared with academic researchers.

Term of Reference 5: Any measures being used, or needed, to mitigate any potential harmful effects of excessive screen-use — what solutions are being used?

Do date, the primary methods employed to mitigate the potential harmful effects of excessive screen use have been professional advice and statutory steps.

Based on work concerning television use, the American Academy of Pediatrics has issued a number of policy statements regarding digital screen time. They first advised children under two against any screen use and called for restricting older children to fewer than two hours of screen use each day [29]. More recently, advice suggests caregivers limit young children to an hour of “high-quality” screen time each day, while paediatricians should discuss children’s screen-based media use directly with caregivers, taking family media use histories, and counselling families to minimize exposure at each visit.  Though this advice is itself not evidence based, research indicates that parents find these proposed limits impossible to keep [36].

Based on fears about the deleterious effects of technology, mainly on school marks, the governments of China and South Korea recently pursued aggressive steps to limit screen engagement. Both governments require localised apps that track screen time in some fashion. In the Chinese case [37,38], game developers implement systems which diminish in-game rewards after three hours of daily play and a series of game addiction clinics have been founded. There are no English-language outcome studies we are aware of that examine these interventions. It has, however, been reported that there are deaths at aforementioned technology ‘addiction’ clinics [39]. In the South Korean case [40], a 2011 law blocks under 16 year olds from accessing games, apps, and some social media platforms from midnight to 6am. This bold initiative was widely hailed as a promising technology-based intervention, preserving sleep and limiting addiction. Yet two independent analyses of this nation-wide natural experiment indicated that the intervention saved children two minutes of sleep each night [41], and had no practically significant effect on limiting internet use [42].

These political and technological approaches to limiting screen time present formidable up front and continued maintenance costs, and there is no reason to believe that similar age-verification based technological solutions would be any more effective if deployed in the U.K.

Term of Reference 6: The extent of awareness of any risks, and how awareness could be increased for particular groups — children, schools, social media companies, Government, etc.

Unfortunately accurate awareness of technological risks is quite low. This is problematic, as the public then pays too little attention to actually harmful aspects of technological innovations. Taking the example of online bullying, public polling indicates that British parents believe online bullying is more prevalent than face-to-face bullying [43]. In our view, this dynamic is fuelled by sensationalised media coverage of reports from academics, pressure groups, and charities. The reality of the situation is that face-to-face bullying is eight times more prevalent than online bullying, more impactful on child well-being. More than nine out of ten of those who are bullied online are also bullied in face-to-face [44]. The only way to address this is by providing resources and support to good media actors such as the charity The Science Media Centre (http://www.sciencemediacentre.org/about-us/).

Term of Reference 7: What monitoring is needed, and by whom

The potentially addictive mechanics employed by some game and platform developers should be indexed and regulated by a governmental body. Unfortunately, because the evidence base is so spare, a more detailed reply to this term of reference is not possible.

Term of Reference 8: What measures, controls or regulation are needed

As a matter of statute, the government should compel industry actors to participate in a monetary- and data-based levy with independent academic collaborators. This step is required to collect the data necessary to draw empirical inferences about the regulatory frameworks that are needed.

Term of Reference 9: Where responsibility and accountability should lie for such measures

Social media and digital entertainment companies collect, store, and profit from extremely rich and sensitive data on our daily lives. They are indispensable partners for the large-scale transparent scientific investigations that will lead to actionable evidence-based policy insights.

Closing Remarks

We have hereby stressed the absence of good, concrete and actionable evidence regarding the effects of technology on young people. While there are notable exceptions, most of the research used to back up claims about the health effects of social media is not of the high quality necessary to instigate policy change. As scientists, we wholeheartedly believe that strong claims need strong evidence, and that the evidence for the strength of claims we routinely face about technology use is not present in the slightest.

Yet we understand that the absence of good evidence is not evidence of absence for technology effects on young people. We reflect regularly on this, and what this means for policy. Hence, we believe that policy needs to instigate statutory data and financial levies to facilitate the construction of a solid and high quality evidence base. Only with better quality data and better quality science will we be able to provide policy with the understanding to instigate evidence-based and successful policy interventions.

We are deeply troubled by how our scientific field cannot yet provide the policy and the public with the necessary quality of research in a time of such heavy debate. We see this occurrence as strong evidence that the ‘lone genius’ model of social science research is not producing the solid science necessary for policymaking in this domain. If researchers need headline grabbing publications, and many of them each year, to survive in the current funding landscape, the field will not produce the results the public needs. We therefore urge you most strongly to consider the need to change the scientific landscape: embracing movements like open data, open science and computational reproducibility. Only once the reward system of academic research improves, will our endeavour to use public funds, in concert with industry data from a statutory levy, to provide much needed unbiased, openly accessible and robust evidence be fruitful.

April 2018

References

1               Ofcom. Children and parents: Media use and attitudes report 2015. 2015. http://stakeholders.ofcom.org.uk/market-data-research/other/research-publications/childrens/children-parents-nov-15/ (accessed 25 May 2016).

2               Certain LK, Kahn RS. Prevalence, Correlates, and Trajectory of Television Viewing Among Infants and Toddlers. Pediatrics 2002;109:634–42. doi:10.1542/peds.109.4.634

3               Dennison BA, Erb TA, Jenkins PL. Television viewing and television in bedroom associated with overweight risk among low-income preschool children. Pediatrics 2002;109:1028–35.

4               Thompson DA, Sibinga EMS, Jennings JM, et al. Television viewing by young Hispanic children: evidence of heterogeneity. Arch Pediatr Adolesc Med 2010;164:174–9. doi:10.1001/archpediatrics.2009.257

5               Hinkley T, Verbestel V, Ahrens W, et al. Early Childhood Electronic Media Use as a Predictor of Poorer Well-being: A Prospective Cohort Study. JAMA Pediatr 2014;168:485. doi:10.1001/jamapediatrics.2014.94

6               Barr R, Danziger C, Hilliard M, et al. Amount, content and context of infant media exposure: A parental questionnaire and diary analysis. Int J Early Years Educ 2010;18:107–22. doi:10.1080/09669760.2010.494431

7               Horodynski MA, Stommel M, Brophy-Herb HE, et al. Mealtime television viewing and dietary quality in low-income African American and Caucasian mother-toddler dyads. Matern Child Health J 2010;14:548–56. doi:10.1007/s10995-009-0501-2

8               Schmidt ME, Rich M, Rifas-Shiman SL, et al. Television Viewing in Infancy and Child Cognition at 3 Years of Age in a US Cohort. Pediatrics 2009;123:e370–5. doi:10.1542/peds.2008-3221

9               Vandewater EA, Rideout VJ, Wartella EA, et al. Digital Childhood: Electronic Media and Technology Use Among Infants, Toddlers, and Preschoolers. Pediatrics 2007;119:e1006–15. doi:10.1542/peds.2006-1804

10               Duch H, Fisher EM, Ensari I, et al. Screen time use in children under 3 years old: a systematic review of correlates. Int J Behav Nutr Phys Act 2013;10:102. doi:10.1186/1479-5868-10-102

11               Przybylski AK, Weinstein N. Digital Screen Time Limits and Young Children’s Psychological Well-Being: Evidence From a Population-Based Study. Child Dev Published Online First: 13 December 2017. doi:10.1111/cdev.13007

12               Parkes A, Sweeting H, Wight D, et al. Do television and electronic games predict children’s psychosocial adjustment? Longitudinal research using the UK Millennium Cohort Study. Arch Dis Child 2013;98:341–8. doi:10.1136/archdischild-2011-301508

13               Przybylski AK, Weinstein N. A Large-Scale Test of the Goldilocks Hypothesis: Quantifying the Relations Between Digital-Screen Use and the Mental Well-Being of Adolescents. Psychol Sci 2017;28:204–15. doi:10.1177/0956797616678438

14               Luyckx K, Soenens B, Goossens L, et al. Identity Exploration and Commitment in Late Adolescence: Correlates of Perfectionism and Mediating Mechanisms on the Pathway to Well–Being. J Soc Clin Psychol 2008;27:336–61. doi:10.1521/jscp.2008.27.4.336

15               Yarcheski A, Mahon NE, Yarcheski TJ. Social support and well-being in early adolescents: the role of mediating variables. Clin Nurs Res 2001;10:163–81.

16               Hale L, Guan S. Screen time and sleep among school-aged children and adolescents: a systematic literature review. Sleep Med Rev 2015;21:50–8. doi:10.1016/j.smrv.2014.07.007

17               Magee CA, Lee JK, Vella SA. Bidirectional relationships between sleep duration and screen time in early childhood. JAMA Pediatr 2014;168:465–70. doi:10.1001/jamapediatrics.2013.4183

18               Tavernier R, Willoughby T. Sleep problems: predictor or outcome of media use among emerging adults at university? J Sleep Res 2014;23:389–96. doi:10.1111/jsr.12132

19               Wagenmakers E-J, Wetzels R, Borsboom D, et al. An Agenda for Purely Confirmatory Research. Perspect Psychol Sci 2012;7:632–8. doi:10.1177/1745691612463078

20               Lenhart A, Smith A, Anderson M, et al. Teens, Technology and Friendships. 2015. http://www.pewinternet.org/2015/08/06/teens-technology-and-friendships/

21               Firth J, Torous J, Nicholas J, et al. The efficacy of smartphone-based mental health interventions for depressive symptoms: a meta-analysis of randomized controlled trials. World Psychiatry 2017;16:287–98. doi:10.1002/wps.20472

22               Orben AC, Przybylski A. Social media is nothing like drugs, despite all the horror stories. The Conversation. 2017.http://theconversation.com/social-media-is-nothing-like-drugs-despite-all-the-horror-stories-79382 (accessed 2 Oct 2017).

23               Przybylski  y, Orben A. Why it’s too soon to classify gaming addiction as a mental disorder. the Guardian. 2018.http://www.theguardian.com/science/head-quarters/2018/feb/14/gaming-addiction-as-a-mental-disorder-its-premature-to-pathologise-players (accessed 27 Mar 2018).

24               Aarseth E, Bean AM, Boonen H, et al. Scholars’ open debate paper on the World Health Organization ICD-11 Gaming Disorder proposal. J Behav Addict 2016;:1–4. doi:10.1556/2006.5.2016.088

25               van Rooij AJ, Ferguson CJ, Colder Carras M, et al. A weak scientific basis for gaming disorder: Let us err on the side of caution. J Behav Addict 2018;:1–9. doi:10.1556/2006.7.2018.19

26               Przybylski AK. Mischievous responding in Internet Gaming Disorder research. PeerJ 2016;4:e2401. doi:10.7717/peerj.2401

27               Przybylski AK, Weinstein N, Murayama K. Internet Gaming Disorder: Investigating the Clinical Relevance of a New Phenomenon. Am J Psychiatry Published Online First: 2016. doi:10.1176/appi.ajp.2016.16020224

28               Weinstein N, Przybylski AK, Murayama K. A prospective study of the motivational and health dynamics of Internet Gaming Disorder. PeerJ 2017;5:e3838. doi:10.7717/peerj.3838

29               Council on Communications And Media. Children, Adolescents, and the Media. Pediatrics 2013;132:958–61. doi:10.1542/peds.2013-2656

30               Anderson SE, Economos CD, Must A. Active play and screen time in US children aged 4 to 11 years in relation to sociodemographic and weight status characteristics: a nationally representative cross-sectional analysis. BMC Public Health 2008;8:366. doi:10.1186/1471-2458-8-366

31               Boone JE, Gordon-Larsen P, Adair LS, et al. Screen time and physical activity during adolescence: longitudinal effects on obesity in young adulthood. Int J Behav Nutr Phys Act 2007;4:26. doi:10.1186/1479-5868-4-26

32               Iannotti RJ, Kogan MD, Janssen I, et al. Patterns of Adolescent Physical Activity, Screen-Based Media Use, and Positive and Negative Health Indicators in the U.S. and Canada. J Adolesc Health 2009;44:493–9. doi:10.1016/j.jadohealth.2008.10.142

33               Foster EM, Watkins S. The value of reanalysis: TV viewing and attention problems. Child Dev 2010;81:368–75. doi:10.1111/j.1467-8624.2009.01400.x

34               Griffiths LJ, Dowda M, Dezateux C, et al. Associations between sport and screen-entertainment with mental health problems in 5-year-old children. Int J Behav Nutr Phys Act 2010;7:30. doi:10.1186/1479-5868-7-30

35               Mistry KB, Minkovitz CS, Strobino DM, et al. Children’s television exposure and behavioral and social outcomes at 5.5 years: does timing of exposure matter? Pediatrics 2007;120:762–9. doi:10.1542/peds.2006-3573

36               Houghton S, Hunter SC, Rosenberg M, et al. Virtually impossible: limiting Australian children and adolescents daily screen based media use. BMC Public Health 2015;15:5. doi:10.1186/1471-2458-15-5

37               China imposes online gaming curbs. 2005.http://news.bbc.co.uk/1/hi/technology/4183340.stm (accessed 27 Mar 2018).

38               France-Presse A. Chinese internet giant limits online game play for children over health concerns. the Guardian. 2017.http://www.theguardian.com/world/2017/jul/04/chinese-internet-giant-limits-online-game-play-for-children-over-health-concerns (accessed 27 Mar 2018).

39               China ‘internet addict’ death sparks fury. BBC News. 2017.http://www.bbc.co.uk/news/world-asia-china-40920488 (accessed 2 Oct 2017).

40               CNN BJL. South Korea pulls plug on late-night adolescent online gamers. CNN. https://www.cnn.com/2011/11/22/world/asia/south-korea-gaming/index.html (accessed 27 Mar 2018).

41               Lee C, Kim H, Hong A. Ex-post evaluation of illegalizing juvenile online game after midnight: A case of shutdown policy in South Korea. Telemat Inform Published Online First: July 2017. doi:10.1016/j.tele.2017.07.006

42               Choi J, Cho H, Lee S, et al. Effect of the Online Game Shutdown Policy on Internet Use, Internet Addiction, and Sleeping Hours in Korean Adolescents. J Adolesc Health 2018;0. doi:10.1016/j.jadohealth.2017.11.291

43               Mascellino A. Children are ‘more likely to be bullied online than in the playground’. mirror. 2017.http://www.mirror.co.uk/tech/your-children-safe-online-parents-9720544 (accessed 19 May 2017).

44               Przybylski AK, Bowes L. Cyberbullying and adolescent well-being in England: a population-based cross-sectional study. Lancet Child Adolesc Health 2017;1:19–26. doi:10.1016/S2352-4642(17)30011-1