What the data actually shows
Studies of online reading behaviour, including work by Ziming Liu, find that people read digital text differently: more skimming and scanning, more jumping around, spotting keywords, and reading non-linearly, with less sustained, in-depth reading than on paper. This is a reasonably consistent description of how screen reading happens, even if the long-term consequences are less clear.
Maryanne Wolf's work on the reading brain emphasises that deep reading — the slow, immersive, reflective engagement that supports inference, empathy, and critical analysis — is not innate but built through practice, and therefore could in principle be weakened if we practise it less. Nicholas Carr's 'The Shallows' (2010) popularised the argument that the web's design trains distraction and shallow processing. These are influential, carefully argued cases, but they are partly interpretive rather than the verdict of settled experiments.
Some experimental evidence complicates the simple 'screens are worse' story. Several studies and meta-analyses find that comprehension can be somewhat better on paper than on screen, especially under time pressure and for demanding texts — a difference sometimes called the screen inferiority effect — but the gaps are not always large or consistent, and depend heavily on the task, the reader, and how reading is measured. The fair summary is that medium and habit matter, but the research is mixed, evolving, and resistant to sweeping conclusions.
Why this feels different from how it actually is
It feels like the internet has shortened our attention because the difficulty is so noticeable: sitting with a long, dense book can feel harder than it used to, and the pull toward checking, clicking, and switching is real. That felt experience is genuine, even where the underlying brain-change claim is uncertain.
The environment also constantly rewards the shallow mode. Online text is engineered to be scannable and to offer a more interesting next link, so the fast, fragmented style is reinforced thousands of times a day, while deep reading is rarely prompted by anything. We naturally drift toward the behaviour the environment keeps rewarding, and then experience that drift as a change in ourselves.
And we tend to attribute a diffuse sense of distraction to a single cause. 'The internet ruined my attention span' is a tidy story for a messy mix of habits, fragmented schedules, stress, and constant interruption. The web is a real part of the picture, but the feeling of changed reading is broader than any one clean mechanism the research has nailed down.
Behaviour has shifted, deep reading is worth protecting, and the strong brain-change claims remain unproven.
What the research says to do about it
Because deep reading appears to be a practised skill, the most consistent implication is simply to practise it: setting aside uninterrupted time for sustained, linear reading of demanding material. Wolf's framing suggests the deep-reading capacity is maintained by use, so deliberately exercising it is the most direct response to worries about losing it.
Reducing interruptions during reading has clearer support than the broad rewiring claim. The well-established attention research shows that interruptions and task-switching degrade comprehension and focus, so reading in a setting without constant notifications and tabs is a reasonable, evidence-aligned step regardless of how the bigger debate resolves.
Matching the medium to the goal also fits the evidence: skimming and scanning are genuinely useful for locating information, while deep reading suits understanding and reflection. The point is not that screens are bad and paper is good, but that the fast mode and the slow mode are different tools, and it helps to choose deliberately rather than always defaulting to the scannable one.
What the research says does not help
Treating the rewiring claim as proven fact does not help, because it overstates evidence that is genuinely mixed and can tip into fatalism — the belief that your attention is permanently broken, which the research does not support. Deep reading appears trainable, not destroyed.
Assuming screens are uniformly worse than paper is also not well supported. Comprehension differences exist but are inconsistent and task-dependent, and plenty of demanding reading happens perfectly well on screens. Swearing off digital reading entirely treats a nuanced, contested finding as a blanket rule it is not.
Conversely, dismissing the concern entirely — insisting nothing has changed — ignores the reasonably consistent finding that online reading behaviour really does skew toward skimming and switching. The honest position sits between the panic and the dismissal: behaviour has shifted, deep reading is worth protecting, and the strong brain-change claims remain unproven.
The fast mode and the slow mode are different tools — it helps to choose deliberately rather than always defaulting to the scannable one.
What this looks like in real life
Built for movement
Online text comes wrapped in hyperlinks, headings, snippets, and an endless next thing, so the fast, fragmented style gets reinforced thousands of times a day while deep reading is rarely prompted by anything. You drift toward the behaviour the environment keeps rewarding — and then experience that drift as a change in yourself.
The book that feels harder than it used to
Sitting with a long, dense book can feel more effortful than it once did, and the pull toward checking, clicking, and switching is real. That felt experience is genuine — but it doesn't prove your attention is permanently broken. Deep reading appears trainable: setting aside uninterrupted time for sustained, linear reading is the most direct way to exercise and maintain it.
Real numbers in context
This is a topic where precise numbers should be treated with real caution, because the strongest claims are interpretive rather than the product of clean, settled experiments. What is reasonably well supported is qualitative: online reading tends toward skimming, scanning, keyword-spotting, and non-linear movement, with less sustained deep reading than on paper, as documented in Ziming Liu's research.
On comprehension, meta-analytic work suggests a modest tendency for paper to outperform screens on demanding or time-pressured reading — the screen inferiority effect — but the gaps vary widely and are not always found. The fair takeaway is that the medium and your reading habits matter somewhat, that deep reading is a practised and protectable skill, and that anyone claiming the internet has definitively rewired the human brain is reaching well beyond what the evidence can currently bear.