It is a question that sounds performative and is not. Something has been happening to how people pay attention, hold complex material, and work with ideas over time. Not to everyone, not uniformly, and not necessarily irreversibly — but the pattern is too large to be explained as individual failure, and the question of whether we are losing some collective capacity for deep thought is one worth taking seriously even if the answer turns out to be partially no.
The empirical evidence is mixed in ways that matter. Gloria Mark’s attention research is solid and consistent. Johann Hari’s synthesis is popular and uneven. Jonathan Haidt’s youth-mental-health argument has been seriously contested by researchers who think the causal claims outrun the data. The academic literature on cognitive offloading finds real effects without clear agreement on how much they matter outside laboratory conditions. Someone who wants a confident answer in either direction — we are losing our minds to our phones or we are fine, these moral panics always blow over — will find evidence for that answer and have to ignore the countervailing evidence to maintain confidence.
The second difficulty is that “thinking” is multiple things. Sustained focus on complex material is one. Holding contradictions in the mind long enough to work through them is another. Forming original judgments rather than accepting pre-formed ones is a third. Remembering things that a device could retrieve is a fourth, and a fifth is navigating the world without a map application. These different kinds of thinking are affected differently by the technologies at hand, and the question can I still think? lumps them together in a way that makes empirical resolution difficult. Some kinds of thinking may be undergoing serious erosion; others may be augmented.
The third difficulty is the collective action problem. A person can, in principle, opt out — fewer notifications, more paper books, extended deep-work sessions, distance from social media. Doing this alone buys back some individual capacity. It does not address the environment everyone else is living in, the schools children are growing up in, or the information infrastructure that civic life now runs on. The individual solution is not the same as the civic solution, and confusing the two makes the civic question harder.
The fourth difficulty is that the framing itself can be used as avoidance. Worrying about can I still think can be a substitute for thinking, in the way that reading about how to be productive can be a substitute for being productive. The question has to be answered, if it is answered at all, by practice rather than by further metaphysics.
Films from the Future treats Limitless as a thought experiment about cognitive enhancement — what would it mean, to whom would it be available, what would it cost in terms of selfhood. The sibling question the book did not raise in 2018, but that the current moment forces: what if the drug running in the background of most people’s daily lives is an attention-shredding one rather than an attention-enhancing one? What if the answer to what would cognitive enhancement do to us? is more immediately visible in what cognitive degradation is doing to us, at scale, by architectural design?
The book’s Deception, Manipulation, and Convenient Lies framework applies particularly to the self-deception dimension. The comforting story — I am just distracted today, I will catch up tomorrow, my capacity for sustained attention is fine — becomes harder to maintain in the face of the longitudinal data. The harder story is that the architecture of most people’s information environment has been optimised for engagement, not for thought, and that the costs of this are borne in a currency conventional risk assessment does not count.
The book’s Risk Innovation framework is directly relevant. The risks at stake are not physical harm. They are threats to dignity, autonomy, and the capacity for self-direction. Those are categories the book’s expanded risk thinking takes seriously. They are categories the regulatory infrastructure around digital technologies mostly does not.
The Science, Belief, and Ways of Knowing framework asks a civic question. Democratic deliberation and serious inquiry both require sustained attention to complex material. If the capacity for that attention is substantially eroded across a population, the downstream consequences for governance and for collective problem-solving are large. This is not an argument for nostalgia about an imagined pre-digital past that was not, in fact, attention-rich for most people. It is an argument that whatever conditions make sustained thought possible are conditions worth defending.
The productive reframing — borrowing the Never Let Me Go move again — is that can I still think may be the wrong question, or at least the wrong entry point. The more tractable question is: what infrastructure supports the kinds of thinking a person and a society need, and what infrastructure undermines it? That question is actionable. Parents, educators, employers, platform regulators, policy-makers — each has a version of the question they can address. The metaphysical framing resolves to a policy question, which is where the book’s frameworks would send it.
One honest concession: this page is being written by an AI, with all that implies about the epistemic position from which the argument is being made. That is worth naming rather than hiding.