Readers Respond to the April 2025 Issue
NEWS | 26 August 2025
NEURODIVERSITIES In “A Complex Diagnosis,” journalist Paul Marino writes about his experience with a type of complex motor stereotypy, a neurological condition that involves recurring involuntary movements. As I read the article, my hands began to move in excited recognition, each tapping an independent part of a polyrhythmic pattern, my equivalent of Marino’s motor stereotypy. I am autistic. Many in our community engage in what are called stims, which can sometimes be more physically dramatic than either Marino’s or my movements and are often consciously suppressed. On supporting science journalism If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today. Some recent evidence appears to link these kinds of neurodiversities with a lack of synaptic pruning. The majority of people grow synaptic connections at a fast rate when very young, yet as they mature, their brain eliminates “unneeded” connections. That may not happen as much for the neurodiverse (ND). It’s possible this is why ND folks often have more neurological “noise” but may also make connections and detect patterns when no one else does. If that’s the case, it might not be a coincidence that creative fields have a disproportionate share of ND people. The fact that up to 20 percent of the global population is ND indicates that nature considers us valuable to the human species. Many of the “disabilities” we ND people experience are really an inability to conform to arbitrary societal norms. Science needs to stop treating us as defective. These traits are indeed phenomena, not disorders. MICHAEL A. LEVINE VIA E-MAIL FIT TO BE TIED? “Are You a Good Judge of Knot Strength?” by Clara Moskowitz and Jen Christiansen [Graphic Science; March], concerns different types of knots and humans’ inadequate ability to assess their strength. Perhaps humans haven’t learned to judge the strength of a knot because they intuitively know it’s much more dependent on what the rope is made of. Having tied hundreds of sutures in my career as a dentist, I can tell you that gut sutures will practically untie themselves (unless properly tied), whereas silk sutures will hold fast with practically any knot. Similarly, a nylon rope will slip easily, but a jute or hemp rope will not. TOM N. TODD OKLAHOMA CITY, OKLA. THOUGHT AND LANGUAGE In “Thinking without Words” [Q&A March], by Gary Stix, neuroscientist Evelina Fedorenko comments on the ability of large language models (or LLMs) to produce language without thought. But I disagree with the notion that what LLMs do is parallel to the way humans learn language. Such artificial-intelligence models ingest enormous amounts of material and produce their “language” based on statistics: which word is the “most likely” one to use following a given word, phrase or sentence. There is no intelligence behind that language. Children do not learn that way. There is much more selection in what they hear. And their acquisition of language, while subject to trial and error, is strongly influenced by the desire for specific results. This desire is entirely absent from artificial intelligence—at least so far. In other words, the processes by which LLMs and children learn language, and the feedback systems by which each gets better and better at language, are different at their cores. I don’t think, as Fedorenko suggests, that AI tools can help us examine the question of how a system of thought can interact with a separate system that stores and uses linguistic expression, because there is no reasoning behind the language abilities of LLMs. TONY STEIN VIA E-MAIL EXPRESSIVE FACES In “Why People Like Expressive Faces” [Mind Matters], Eithne Kavanagh, Jamie Whitehouse and Bridget Waller wonder why some people remain “comparatively inexpressive across situations.” They speculate that this could be some kind of cost-benefit or alternative strategy. For some people, facial expressions just don’t come naturally. As an autistic person, I’ve practiced them in front of a mirror and tried to learn when they apply. It’s exhausting, feels fake and is often perceived as fake. GEORGE WIMAN VIA E-MAIL TOWERING TASK In the Tower of Hanoi puzzle described by Heinrich Hemme in “Move the Tower” [Advances; January], it is easy to see that for low values of n, moving a tower with n disks will take 2n – 1 moves. A nice challenge for the reader is to prove this for all values of n. A fairly simple recursive proof will do it. Given the result of 264 – 1, which is greater than 1019, moving a 64-level tower with one move per second would take more than 580 billion years. The monks were quite right that the end of the world would come before they finished the task! JAMES R. PAULSON PROFESSOR OF CHEMISTRY EMERITUS, UNIVERSITY OF WISCONSIN–OSHKOSH CLARIFICATION “Refreezing the Arctic,” by Alec Luhn [June], said that the Bright Ice Initiative had scattered tiny glass beads on glaciers in Iceland and India to reflect sunlight. The organization initially used that approach in a field test in Iceland. It subsequently switched to a clay-based material, which it used in a field test in the Himalayas. ERRATA “Language on the Mind,” by Gayoung Lee [Advances; May], should have said that Cas W. Coopmans explained that verbs sometimes come near the end of a sentence in Dutch language structure. And it should have included the word “because” at the beginning of the example “I ate a cookie with chocolate.” The question for 41-Down in the Science Crossword also should have included “because” in that example. “Cosmic Dawn,” by Rebecca Boyle [June], should have said that sometime between 50 million and 100 million years after the big bang, gravity drew hydrogen atoms together and ignited the first stars. In “Pay Dirt,” by Elizabeth Anne Brown [July/August], the first image caption misspelled the name of Marie Aagaard Larsen.
Author: Aaron Shattuck.
Source