Are Human Reasoning Abilities Declining?
In a piece at the Financial Times (and in a Bluesky thread about it), John Burn-Murdoch, the chief data reporter for the newspaper, goes over some of the worrying findings that might support a conclusion that human capacities to reason and understand are “deteriorating,” especially since the early-to-mid 2010s.
He clarifies that what he is talking about are not changes to “the fundamental biology of the human brain” over the past 10 to 15 years, but rather the effects of social, technological, educational, and situational changes on our ability to think well.
While I’m not in a position to assess whether Burn-Murdoch is giving us a complete picture of the relevant data, what he does share seems worrisome.
The data includes:
- declining average scores for teens on science, reading, and math assessments
- declining average scores for adults on numeracy and literacy
- increased reports by 18-year-olds in “difficulty thinking or concentrating” and “trouble learning new things”
- decline of leisure-time reading
a compilation of graphs by John Burn-Murdoch (click to enlarge)
Burn-Murdoch notes that the early-to-mid 2010s drop-offs we see across much of this data corresponds with “our changing relationship with information, available constantly online.” He continues:
Most discussion about the societal impacts of digital media focuses on the rise of smartphones and social media. But the change in human capacity for focused thought coincides with something more fundamental: a shift in our relationship with information. We have moved from finite web pages to infinite, constantly refreshed feeds and a constant barrage of notifications. We no longer spend as much time actively browsing the web and interacting with people we know but instead are presented with a torrent of content. This represents a move from self-directed behaviour to passive consumption and constant context-switching.
The data seem relevant to the discussion here and elsewhere earlier this month about “the average college student at the average college” these days.
It may also point to an opportunity for colleges and universities to respond to what could be considered a “critical reasoning crisis” by supporting and promoting philosophy departments and other departments whose work and teaching are primarily oriented around reasoning well. Admittedly, there is only so much that additional critical reasoning courses at the college level could likely do to move the needle on this data. Broader cultural changes are needed, including, perhaps, a reconsideration of the educational environment of children and young adults.
(via Victor Ronsin)
*Irrelevant to the topic*
“We have moved from finite web pages to infinite, constantly refreshed feeds …”
Another instance of a suspicion of mine: the word “infinite” in natural language in fact means large finite numbers…
I think this usage here is closer to the etymology – “infinite”, meaning “without end”. Modern logicians tend to distinguish the Aristotelian “potential infinity” of a thing that has no bound from an “actual infinity” of a thing that already has more than any finite number of elements, but there is an important connection in these meanings! The issue being addressed is not one where web pages have gotten really big – it’s rather one where web pages (in particular, social media feeds) are designed in a way that they always grow past whatever point one has scrolled to, which really does make them “infinite” in a sense that some gigantic seven million word screed on a 1990’s website was not. (You might object – but what if someone scrolls past literally every Facebook or Twitter post that has ever been made? Well, scrolling takes time, and both Facebook and Twitter are growing faster than people can scroll. And in any case, I believe they are both designed to keep re-displaying highly-algorithmically-recommended posts if you keep scrolling past the lower-rated ones for you, so the feed keeps going, even if it’s only finitely many posts, constantly re-mixed.)
Cultural explanations no doubt have their place, but I wonder how much of the picture is chemical, given that our brains are riddled with microplastics and we are breathing ever higher concentrations of carbon dioxide. Both seem to be intractable problems that are only getting worse.
Carbon dioxide seems unlikely to be relevant here – even well-ventilated indoor spaces have always had CO2 concentrations higher than the ones forecast for outdoors a century from now, and often two or three times as high. Since reading has almost always taken place indoors, I would expect that changes in CO2 concentration would be relatively negligible.
Microplastics can’t be ruled out on these sorts of grounds, because they are genuinely new in the past few decades, but I don’t think there’s any particular reason to suspect that they would have effects like these. And the changes being mentioned seem to have picked up speed around 2012, rather than in 1980 or so, when plastics were already becoming widespread in the environment.
Technological changes in how people interact with information, by contrast, seem like precisely the kinds of changes that could have effects like these, and they are also clearly very widespread.
Also, I think it’s important to distinguish these technological changes from cultural ones, since technological changes are immediate, while there are additional cultural changes in response to these technological changes that take time to develop and diffuse. (Compare the spread of the printing press, books, newspapers, and pamphlets to the slower spread of the cultural changes like freedom of religion and freedom of press and “don’t trust everything you read” that developed in response to these technological changes.)
One particular worry I have is that previous technological changes in media production and consumption often had decades or centuries to develop cultural responses before new ones arrived, but the internet arose just a few decades after television, and then smartphones and social media hit just a decade after the internet, and AI is hitting just a decade after that.
I would include the adoption of new technology under “culture”, material culture specifically. As for the CO2, I think it’s important to recognise that the exposure isn’t just occuring during reading and task performance, but from gestation onwards. You are probably right however that the lion’s share of causality presently goes to techonolgical/cultural changes.
Returning to the plastics for a moment, a few interesting factoids for the interested.
Evidently, a large proportion of microplastics are from rubber tire dust; research on ocean microplastics put the proportion at 78%, although I don’t know how that carries over to non-ocean measures, how representative it is, etc.
https://www.systemiq.earth/wp-content/uploads/2020/07/BreakingThePlasticWave_MainReport.pdf
Nature made a splash a few months ago by citing a researcher who necropsies brains measuring microplastics to the effect that every adult has *7-10 grams of microplastics just in their brain*; helpful analogies were made with an unused crayon or a disposable plastic fork.
https://www.nature.com/articles/d41586-025-00405-8 (Nature magazine article; now paywalled, but very widely glossed in other media)
https://www.nature.com/articles/s41591-024-03453-1 (Nature Medicine article from February; given the timing I assume that may have prompted the magazine article)
And – from this December NPR piece, some “reassuring” news:
“[Microplastics researcher Matthew] Campen says you don’t see a correlation between age and the concentration of microplastics in human tissue. In other words, it doesn’t appear to endlessly accumulate inside us. It’s possible that our bodies may reach some sort of “equilibrium” based on how much is around us.
“We don’t have enough data,” he says, “but [our work] suggests there’s a very rapid time to saturation –– you do hit a limit and eventually you’re clearing it.”
A study of Zebrafish (sometimes used in biomedical research) found the uptake of microplastics did plateau at a certain point and levels decreased when the animals weren’t being exposed. The problem is the saturation point went up proportionally to how much the animals were exposed to, says Campen.
“That’s basically where we are right now,” says Campen. “Our environmental exposure keeps going up because we’re doing nothing to stop it.”
https://www.opb.org/article/2024/12/18/are-microplastics-harmful-to-our-health-scientists-are-trying-to-find-out/
I feel like there are two ways this discussion often turns into people speaking past one another, or mistrusting one another’s motivations and rigor.
A. People who agree there seems to be worsening systematic problems with reading comprehension, attention, and critical ability often posit one of the following causes, or a combination:
1. smartphones
2. social media
3. primarily inhabiting a mental and informational world that exists on the post-2010 internet
4. the (presumably transient, 1990s-00s) misstep of whole language literacy education potentially having permanently stunted a generation of readers – the “Selling a Story” reading thesis.
Because each of these proposed causes are fairly plausible, quite often people proceed as though the causal relevance of their proposed explanation(s) is too obvious to merit argument. Most of all, particularly outside of academic discussion, the form I see is “well it’s obviously the smartphones, and anyone pretending it’s something other than the phones is simply in willful denial because they’re also on their phones.” I think this is a mistake. I’m not sure how easy it is to come to better conclusions (the “whole language vs. phonics” stuff could be looked at quantitatively based on where it was adopted or not adopted, but the rest seems pretty difficult to disentangle) but assuming we understand the causes when we don’t isn’t rigorous.
B. And then, if people are given to splitting up into different rhetorical camps based on strong assumptions about causes, there’s *also* a tendency to split into different camps about whether this or that argument is borne out of elitism or pedagogical curmudgeonliness; a lot of the discussion of the Stephen Hales post, here an on bsky, ended up turning into positioning of this kind. (As I saw he comments here: I take Dr. Hales’ sincerity and urgency at face value and the things that scare him scare me; but that post would have been a slow pitch straight right over the plate for ‘anti-elitists’ even without the “amazed they ever leave their goon caves at all” line.)
There is also potentially a political valence to questions along the general lines of “what’s up with kids these days,” “are lenient marking and grade inflation causal factors,” “are students being spoonfed ever more pre-digested learning,” etc.; people with identical politics may differ on the importance of pedagogical severity or lenience, including in ways contrary to the stereotypical views associated with their politics, but one can easily see where cultural or political disagreements could also probably strongly color peoples’ opinions and discussion.
I haven’t even touched on LLMs, as I think most people’s timeline for “what’s going on” predate their mass-adoption, but they couldn’t be a worse technology, coming at a worse time, human thought-wise.
This probably correlates pretty well with the decline in philosophy majors starting in the early to mid 2010s.