A hallmark of human cognition is the capacity to think about observable experience in ways that are nonobvious—from scientific concepts (genes, molecules) to everyday understandings (germs, soul). Where does this capacity come from, and how does it develop?
Contrary to what is classically assumed, young children often extend beyond the tangible “here-and-now” to think about hidden, invisible, abstract, or nonpresent entities. . . . The standard developmental story may be backward: for young humans, going beyond the obvious can be easy, and sticking with the here-and-now can be a challenge. [This has] implications for how children learn, what is basic in human thought, and how tendencies that make us so smart and sophisticated can also be sources of distortion and bias.
The paper concludes:
Nonobvious concepts are foundational to human thought—not just for scientists but also for young children. By preschool age, young children go beyond the obvious in quite basic yet important ways. They display robust essentialist beliefs, they refer to abstract kinds in their generic language, and they incorporate object history in their reasoning and inferences. . . .
This work has three key implications for how children learn.
First and most evidently, this suggests that knowledge is not just built up piecemeal or unidirectionally from sensory or perceptual observations. Rather, abstract and placeholder representations can structure how knowledge is represented and built upon.
Second, it points to the importance of learning from others . . . Nonobvious concepts require social input, as children cannot just rely on information from their own senses. Children hold expectations about social cues (both linguistic and nonlinguistic) that provide rich infor- mation about the world.
Third, the ability to reason about nonobvious aspects of the world can be a driver of intellectual development, not merely its outcome. A contrast between appearance and reality generates exploration and discovery, placeholder structures invite innovation and allow for conceptual change, and imagined possibilities permit vicarious learning.
All this resonates with me. It connects fundamental ideas of cognition and behavior with what I have personally observed in statistical and scientific workflow. There’s so much here to chew on.
Also this, which relates to our work on essentialism in public opinion:
The cognitive tendencies that make us so smart and sophisticated can also be sources of distortion. and bias. Each of the phenomena reviewed in this article has benefits for navigating the complexity of the world: Essentialism motivates a curiosity about, and search for, new and nonobvious commonalities within a category. At the same time, it leads to exaggerating group differences, prejudice and intergroup hostilities, and misconstruals of scientific constructs such as genetics and evolution. Generic language permits efficiently transmitting hard-won discoveries to others, yet also glosses over within-category variability and contributes to stereotyping and essentialism. And attention to object history permits tracking contamination, contagion, and discovering causal patterns more generally, but also may contribute to superstitious behaviors and super- natural beliefs. An important question for the future is whether we can harness the potential of our remarkable ability to consider realities beyond the obvious and discover ways to counter its negative consequences.
Good observations, important questions.
I am glad that your paper took a balanced approach. When I read “young children often extend beyond the tangible “here-and-now” to think about hidden, invisible, abstract, or nonpresent entities” I immediately thought this might parallel the “ability” of LLMs to invent things that aren’t there. The ability to abstract is powerful but not necessarily in revealing truth. The more we seem to learn about the way the brain works, the less convinced I am that it is fundamentally different than the way AI works. I know that is controversial, and I also acknowledge that the actual mechanisms of biological brains and computers must be different – but the common statements that only human brains are capable of thinking strikes me as less and less obvious.
“whether we can harness the potential of our remarkable ability to consider realities beyond the obvious…”
We don’t seem to have any problem doing this and turning it into important progress. Our problem is that we also:
– frequently create and become driven by *non* realities beyond the obvious
– frequently (especially in the last few decades) outsmart ourselves and reject obvious realities for the not-obvious creations of our imagination
All of which makes this bit the most important:
“discover ways to counter its negative consequences.”
“frequently (especially in the last few decades) outsmart ourselves and reject obvious realities for the not-obvious creations of our imagination”
So when Copernicus discovered that the earth actually rotates around the sun, he…oh, never mind.
I know Chipmunk is just owning the libz again, but I would say the opposite about the last few decades. The ancient Greeks were deep thinkers but hindered by a lack of basic knowledge about the world. Then came the Dark Ages, when abstract thinking was actively discouraged in favor of explanations involving God’s will or witches. Now us modern humans – even the libz – live in a world where gods and witches don’t go on the cause tree and we are encouraged to “think out of the box.” This milieu allows our ability to reason in the abstract to outpace our ability to evolve bigger and better brains.
“Now us modern humans….live in a world where gods and witches don’t go on the cause tree”
From the population bomb to peak oil to the power pose to critical race theory, modern witches are finding Jesus in their toast everywhere in the modern world. I love the idea of replacing human perception with “mathematical perception” that fails to measure up to reality. Of course, the usual applies: “mathematical perception” itself isn’t the problem. When it’s tested against reality and proven to reproduce it, it is an excellent tool.
OTOH, when it’s not tested – and especially when the error bars are wider than reality! – it’s just a tool to for a bunch of geeks to express their imagination.