What effect do screens have on our sleep?

We all notice it—our eyes have nev­er spent so much time star­ing at screens. Wheth­er at work, at home, in pub­lic trans­port, or in wait­ing rooms, look­ing at them has become almost instinct­ive. Accord­ing to the 2022 Digit­al Baro­met­er¹, 89% of French people over the age of 12 own at least one com­puter, wheth­er desktop or laptop, per­son­al or pro­fes­sion­al. That num­ber rises to 92% when con­sid­er­ing mobile phones alone.

A grow­ing field of research has emerged to invest­ig­ate their poten­tial impact on human health, par­tic­u­larly in rela­tion to sleep qual­ity. While it is widely believed that blue light dis­rupts our bio­lo­gic­al rhythms and affects sleep, some research­ers sug­gest a more nuanced per­spect­ive. Among them is Rus­sell Foster, pro­fess­or of cir­ca­di­an neur­os­cience and Dir­ect­or of the Sleep and Cir­ca­di­an Neur­os­cience Insti­tute at the Uni­ver­sity of Oxford, who warns against draw­ing con­clu­sions solely from labor­at­ory stud­ies: “Many recom­mend­a­tions we hear today are largely, if not exclus­ively, based on labor­at­ory stud­ies. How­ever, study­ing the impact of light on human beha­viour in an arti­fi­cial envir­on­ment may lead to mis­lead­ing conclusions.”

Foster, who has writ­ten extens­ively on sleep and cir­ca­di­an rhythms ², ded­ic­ates much of his work to cla­ri­fy­ing what sci­ence actu­ally knows about sleep. How­ever, he acknow­ledges that even if we know that sleep qual­ity is strongly linked to over­all health, many aspects remain unclear: “We do not really have a good mech­an­ist­ic under­stand­ing of the con­nec­tions between poor health and sleep quality.”

“One strik­ing example is a study con­duc­ted at Har­vard a few years ago, explains the pro­fess­or. This study examined the effects of pro­longed expos­ure to an e‑reader screen (sim­il­ar to a Kindle) set at its max­im­um bright­ness (about 30 lux) for four hours before bed­time, over five con­sec­ut­ive nights. How­ever, before their expos­ure to the e‑reader, the par­ti­cipants had already spent sev­er­al hours in a labor­at­ory envir­on­ment illu­min­ated at around 90 lux. After five days of expos­ure, par­ti­cipants’ sleep onset was delayed by nine minutes. And that res­ult was just stat­ist­ic­ally significant.”

Foster draws an addi­tion­al con­clu­sion from this study: “Melaton­in levels rise in anti­cip­a­tion of night­fall, peak­ing around 4 AM, which has led to the belief that melaton­in is a sleep hor­mone. How­ever, the e‑reader exper­i­ment clearly showed both a sup­pres­sion of melaton­in and a marked delay in the cir­ca­di­an rhythm of melaton­in. But this did not dir­ectly impact upon sleep/wake beha­viour. Even though bio­lo­gic­ally sig­ni­fic­ant changes were observed in melaton­in, the beha­vi­our­al impact was far less pro­nounced.” Anoth­er key point was that the sub­jects were exposed to dim light, around 90 lux, in the labor­at­ory pri­or to the e‑reader. A few years later anoth­er group repeated the exper­i­ments but exposed the par­ti­cipants to around 550 lux for 6.5hr dur­ing the day. The effect of this was to com­pletely abol­ish the effects of e‑reader use both on sleep and melaton­in. It seems that “light his­tory” can have a big impact upon how sens­it­ive we are to light in the evening.

This study focused on a spe­cif­ic type of screen. E‑readers are designed to min­im­ize screen bright­ness and facil­it­ate read­ing, and most mod­els (except the one used in this exper­i­ment) only dis­play gray­scale text. This raises broad­er ques­tions about the spec­trum of emit­ted light and its intens­ity across dif­fer­ent wavelengths.

“What sci­ence has determ­ined is the sig­ni­fic­ant link between our expos­ure to nat­ur­al light and the reg­u­la­tion of our cir­ca­di­an rhythm,” the pro­fess­or asserts. “And if spe­cify­ing that the light source is nat­ur­al mat­ters, it is primar­ily a ques­tion of intens­ity. In the even­ing at home, ambi­ent light is estim­ated to be around 100 – 300 lux. The bright­est arti­fi­cial light, typ­ic­ally found in offices, reaches approx­im­ately 400 lux. By com­par­is­on, nat­ur­al light is far more intense—a cloudy day out­doors provides at least 10,000 lux, while a sunny day can exceed 100,000 lux – even in Eng­land! It is estim­ated that 30 minutes of expos­ure to 10,000 lux is enough to reg­u­late our intern­al clock.”

“In con­trast to our vis­ion, the cir­ca­di­an sys­tem is incred­ibly insens­it­ive to light, and we still don’t fully under­stand how light intens­ity, length of expos­ure, light his­tory, how old we are and the col­our (wavelength) of the light all inter­act to reg­u­late our cir­ca­di­an rhythms. What we do know is that bright white light or around 10,000 lux for 30minutes seems to be effect for most people.”

On the basis of the data we have so far, it is not so much the light from elec­tron­ic devices such as e‑readers, smart phones or com­puter screens that is the prob­lem but rather the stim­u­lat­ing effects these devices induce. Social media, gam­ing, watch­ing a film and emails will act to make us more alert and this will delay sleep. The key point is that these men­tally enga­ging activ­it­ies delay sleep, with little impact from the emit­ted light.

“The wavelength of light has also been widely debated. In one study³, we demon­strated that nov­el photore­cept­ors in the eye, dif­fer­ent from the visu­al photore­cept­ors, the rods and cones, and called pho­to­sensit­ive ret­in­al gan­gli­on cells (pRGCs) are most sens­it­ive to 480-nano­met­er wavelengths, in the blue por­tion of the spec­trum,” explains Rus­sel Foster. “But this find­ing applies only to an isol­ated response from these cells—observed in labor­at­ory mice that lacked rods and cones. If those were present, the spec­tral responses were dif­fer­ent.” At the time of their dis­cov­ery, research­ers ten­ded to dis­tin­guish between visu­al and non-visu­al responses to light. Cones and rods were believed to be respons­ible for visu­al responses, while pRGCs were thought to reg­u­late non-visu­al pro­cesses. “The truth is that they com­mu­nic­ate with each oth­er,” cla­ri­fies the pro­fess­or. “What we con­cluded is that rods likely con­trib­ute to the dim light sens­it­iv­ity of the intern­al clock, cones prob­ably integ­rate flick­er­ing stim­uli, and pho­to­sensit­ive ret­in­al gan­gli­on cells essen­tially func­tion as bright­ness detect­ors. How­ever, how exactly they inter­act remains unclear, and this is an act­ive area of study.”

Although the rods, cones, and pRGCs inter­act, study­ing these inter­ac­tions is com­plic­ated and some of the pub­lished stud­ies have got it wrong. If you want to com­pare the effect of dif­fer­ent wavelengths of light you need to deliv­er the same num­ber of photons at dif­fer­ent wavelengths. Blue light has more energy than red light, and in many cases, research­ers have com­pared the same energy levels and not the same num­ber of photons. As a res­ult, there would be few­er photons of high energy blue light com­pared to low energy red light, alter­ing the appar­ent sens­it­iv­ity of the response. These stud­ies have fur­ther con­fused the pic­ture of what is going on.

Inter­est­ingly, there are pro­grammes that shift the col­our of screens from “blue-enriched” dur­ing the day to “red enriched” dur­ing the even­ing. These were developed to pre­vent screens delay­ing the sleep/wake cycle in the even­ing. How­ever, there are no data that show this actu­ally works. Of course, screen-induced eye fatigue is a real issue, and while blue light is often blamed, its impact is primar­ily due to its high­er per­ceived intens­ity. Screen bright­ness ranges from 30 to 300 lux, which is very little com­pared to sun­light. So why does­n’t the sun cause the same eye strain? The answer seems to be that screen usage requires con­stant visu­al accommodation—our eyes must adjust to a bright focal point against a sig­ni­fic­antly dark­er sur­round­ing envir­on­ment. This con­trast is what leads to eye strain.

And that brings us to a more con­cern­ing issue asso­ci­ated with lack of nat­ur­al light: Myopia. “Foster men­tioned a 2019 study⁴ which showed  an alarm­ing trend: between 70 to 90% of young urb­an people under 18 in South­east Asia are affected by myopia. It seems that these indi­vidu­als are spend­ing very little time out­side in bright nat­ur­al sun­light, and too much time inside look­ing at their com­puters. Bright sun­light seems to stop the eye elong­at­ing dur­ing devel­op­ment. A elong­ated eye causes an image to be formed in front of the ret­ina which then has to be cor­rec­ted by glasses. The light from screens is simply not bright enough to pre­vent eye elong­a­tion. So young people need to spend time out­side for healthy eye development.”

Pablo Andres