The clearest evidence that attention is under pressure today comes from studies of multitasking, switching, and the technologies that surround studying. The so-called multitasking myth claims we can perform two attention-demanding tasks at once, but in reality the brain rapidly switches, paying a reconfiguration cost each time and making more errors; genuine parallel processing only happens for automatic behaviors, so reading while messaging is not two tasks in parallel but one task, repeatedly interrupted. The widely quoted 40% efficiency loss comes from lab switching trials and is routinely over-generalised. Ophir, Nass, and Wagner (2009) found that self-reported heavy media multitaskers performed worse on tasks requiring filtering of irrelevant information and controlled attention, though the design is correlational and cannot show that multitasking caused the deficit, and a 2017 meta-analysis found the effect small and inconsistently replicated.
Phones and laptops have come under particular scrutiny. Ward et al. (2017) reported a brain drain effect in which participants with a phone face-down on the desk did worse on working-memory and fluid-intelligence tasks than those who left it in another room, though later replication attempts have largely failed to reproduce this specific effect. The better-supported cost is behavioural: a reachable phone invites checking, and each check fragments attention and interrupts encoding. Glass and Kang (2019) found that students in classroom sections where devices were permitted scored about 5 percentage points lower, roughly half a letter grade, on end-of-term exams, and strikingly within-class comprehension was unaffected; the damage showed up only in long-term retention, which is why the cost is invisible while it is happening. Mueller and Oppenheimer (2014) reported that longhand note-takers outperformed laptop users on conceptual questions, attributing it to verbatim transcription, though a large 2021 multi-site replication failed to reproduce the effect; the better-supported problem with laptops is not handwriting but internet access, since off-task browsing during lectures is reliably associated with lower grades.
The mechanism by which phones and notifications damage focus is more subtle than the time spent on them. Studies of notification effects find that merely receiving a notification degrades performance on a concurrent task even when the phone is never picked up, because the cue sets off task-irrelevant thoughts about what it might be, and silent-but-visible still beats buzzing, while out of the room beats either. The famous 23 minutes to refocus figure from Gloria Mark's observational studies of office workers is the average time to return to the original task after detouring through other work, not a measured recovery-of-concentration time and not measured on students, so it should be treated as an illustration of switching cost rather than a constant. Each check also adds attention residue, the cognitive leftover from a previous task that lingers after switching and impairs performance on the new task until it dissipates, plus the time to rebuild place in the material, so a ten-second glance can carry a multi-minute tail that is not felt subjectively. Studies comparing self-reported phone use against automatically logged use find people consistently underestimate how often and how long they use their phones, often by a wide margin, which means the intuition that one's phone is not really affecting study is produced by the same attentional system that is failing to register the interruptions.
A range of design choices and environments can protect attention in the face of these pressures. A phone parking lot, a designated physical location where phones are placed during deep work, creates a friction barrier that prevents habitual checking. A shutdown ritual, popularized by Cal Newport, reviews what got done, notes the next step, and closes loops to reduce rumination about unfinished work that can leak into the evening and degrade sleep. Digital minimalism, the philosophy of keeping only digital tools that strongly support values and eliminating the rest, directly addresses the attention costs of optional connectivity, while a second brain, an external digital knowledge management system, offloads information from biological memory to free working memory for creative and analytical thinking. Cognitive offloading more broadly, using lists, calendars, and external tools, reduces mental workload, though over-reliance can weaken memory and attentional capacity over time, and an information diet, the deliberate curation of media inputs, treats the rate of information consumption as a variable that affects mental energy and concentration capacity. The attention economy, a framework that treats human attention as a scarce commodity, makes these design choices easier to see: tech platforms profit by capturing and selling attention, so awareness of these incentives is essential for protecting study focus. Working memory capacity itself is severe: Miller's classic 1956 paper proposed about \(7 \pm 2\) chunks, while more recent estimates (Cowan, 2001) suggest roughly 4 chunks for adults, highlighting the bottleneck that limits what we can hold in mind during complex study.