Much of the revised literature on technology use and sleep used binary significance testing (which reflects the nature of the science during this time). In other words, researchers were primarily testing whether there was a significant relationship between technology use and sleep, and significance was usually defined as obtaining a significance level of p < 0.05. More recently, binary significance testing has been criticized [30]. For example, a relationship may occur between two variables, but it may be so small that it does not mean much in the real world.

The field of technology use and sleep seemingly neglected the size of the real-world effect between these two variables. However, a meta-analysis by Bartel et al. [31] was able to estimate the magnitude of the effects between various technological devices and teenagers’ sleep, along with a multitude of other risk (and protective) factors. Surprisingly, the correlations between technological devices and sleep were negligible. The use of technology was not related to sleep latency, and only computer use was associated with a decrease in total sleep time. Technology did appear to correlate, to a small extent, with bedtime. Namely, as video gaming, phone use, computer use, and Internet use increased, bedtime became later [31]. Figure 4.3 provides an illustrative look of the relative protective and risk factors for adolescents’ sleep, including “technology use.” The segments of the pie chart demonstrate the percentage of variance from each factor. We have used the mean-weighted correlations between “technology use” and “sleep” (i.e., between Internet use and bedtime, which showed an r = 0.212) from the meta-analytical findings from Bartel and colleagues – thus, at best, technology use represents just a sliver of a contribution toward adolescents’ sleep. At a glance, there appear other more important contributions, including family influences (i.e., parent-set bedtimes, family environment). The most obvious piece is represented by the question mark. Normally, this would mean we do not know what this extra contribution is and may elect to claim it is a measurement error, other methodological anomalies, or simply things we do not know. However, Bartel and colleagues were unable to meta-analyze biological contributions to teenagers’ sleep (i.e., circadian rhythm timing, sleep homeostasis, genetics), which are known to be major influences on teenagers’ sleep.

Our insight into whether the technology use of adolescents occurs after sleep becomes problematic comes from a cross-sectional study of 2,546 Belgian adolescents. The 2006 study, titled “Nodding off or switching off? The use of popular media as a sleep aid in secondary-school children,” was likely ahead of its time. Eggermont and van den Bulck [32] asked seventh- (13.2 years) and tenth-grade (16.4 years) adolescents how often they used either computer games, television, books,² or music to help them fall asleep. Respondents answered either never, rarely, sometimes, or often. One in five adolescents used television at least occasionally as a sleep aid, one in ten used computer games, and one in three used books, but it was music that was used the most, with almost one in every two adolescents using music to help them fall asleep. Despite that only books led to an earlier bedtime, more sleep, and less next-day tiredness (compared to other sleeping aids), this was the first study of teenagers exploring technology use as an aftereffect of trouble sleeping. We are likely overstating this claim, as technically adolescents were not asked to report if they had a sleep problem. We can only infer that by asking teenagers if sleep onset was assisted with an associated technological activity, that at least for some adolescents, difficulty initiating sleep may have occurred before the use of such technology.

Perhaps the best evidence to date attempting to answer the question about whether sleep can affect technology use comes from a prospective study of adults. Tavernier and Willoughby [33] followed 942 emerging adults (age at Time 1 = 19.0 years) for 3 years and measured their average weekly hours of TV watching and online social networking (e.g., Facebook, Myspace, Twitter, e-mail, Messenger), as well as their typical sleep duration, as well as an adapted version of the Insomnia Severity Index (items included difficulty initiating sleep, staying asleep, early morning awakening, and sleep satisfaction [34]), which provided a continuous variable known as “sleep problems.” Cross-lagged analyses showed that sleep problems at Time 1 predicted Time 2 television watching and online social networking, but the reverse relationship (i.e., technology use predicting later sleep problems) was not supported. Interestingly, no prospective relationships existed between technology use and sleep duration, which supports the meta-analytical findings in adolescents [31]. Nevertheless, these prospective findings suggest young adults’ perceptions of their sleep, including perceived sleep difficulties, are an important perspective for researchers to consider, as like bedtime data, difficulties in sleeping appear to show significant relationships with technology use (e.g., [35]).

When we began our first discussions on planning this chapter about technology and sleep, we knew we wanted to “spin readers’ thinking” about this relationship. We do not mean to claim that using a technological device does not lead to sleep problems in young people. We have observed this, whether in our own children or teenagers presenting with sleep problems to our Child and Adolescent Sleep Clinic at Flinders University (Fig. 4.4), albeit temporarily. Instead, we believe we need to work harder than usual to convince readers that, overall, the effect between technology use and sleep is small to negligible. If anything, we lack data to analyze whether technology use fills the void while young people wait for sleep onset to arrive [32]. Sleep problems may already exist in young people [36], and it is possibly better to avoid lying in the quiet darkness ruminating about past events and catastrophizing about future events [37, 38] by distracting oneself with a screen or the sound of music. The displacement hypothesis proposes that technology use may replace other activities, including sleeping [13]. So far, the data provide support for this hypothesis, as if anything, bedtime is related to technology use (more than sleep latency or sleep itself). However, there have been extremely few studies that have attempted to experimentally manipulate technology use and observe the effect on bedtimes [28]. If using technological devices accounts for at best 4 % of the variation in teenagers’ sleep (or more accurately, bedtimes; Fig. 4.3), then doesn’t this suggest we should turn our attention toward other culprits for why young people may sleep too little and too late [39]? We conclude this chapter by directing readers to Box 4.1, which lists areas for future research – including areas that do not involve technology use.